Category Award Winners

2007 AAAR Annual Conference Abstracts
TUESDAY PLENARY
Single Particle Analysis All the Way up to the
Stratosphere Daniel Murphy, National Oceanic and
Atmospheric Administration (NOAA), Boulder, CO
At NOAA, we have been studying the composition of
single atmospheric particles with a laser ionization
mass spectrometer for almost 10 years. Such mass
spectrometry provides tremendous sensitivity and
statistical significance. It is also a very general
technique that can measure almost anything present in
particles from, organics to mineral dust. For this talk, I
will focus on two topics where single particle analysis
provides information not available from bulk analyses.
The first topic is the stratosphere. There, we can
distinguish particles that were formed below, within,
and above the stratosphere. That information has
implications as diverse as the rate at which organics are
oxidized in atmospheric particles to the flux of meteors
hitting the earth. The second topic is the nucleation of
ice crystals in the atmosphere. There are large
differences in composition between particles that are
good at nucleating ice and those that are not. The ice
nuclei also contain a wide variety of chemical
compounds, so the problem is a good match to the
capabilities of a mass spectrometer. We were able to
confirm some known properties of ice nucleation, such
as that mineral dust particles can be excellent ice nuclei.
We also found some new properties, such as the effect
of organics on particles that aren’t such good ice nuclei.
Besides the science, I’ll show a few pictures of what it
is like to put such an instrument on airplanes.
WEDNESDAY PLENARY
Inhaled Insulin and the Marvelous New Innovations
in Aerosol Medicines, John Patton, Nektar
Therapeutics, San Carlos, CA
In contrast to 10 years ago when the only inhaled
medicines were for asthma, inhaled products are now
being developed for virtually all types of lung disease
(lung disease is among the top 3 killers of humans
world wide) and also for the systemic delivery of small
and large molecules.
After more than 15 years of development, the first
inhaled insulin, Exubera, was approved for marketing
in January 2006 in both Europe and the US. In
addition to the first non-injectable insulin and the first
inhaled systemic drug to be approved in 40 years,
Exubera represents a number of technical achievements.
It contains the first room temperature stable insulin
formulation, the first amorphous insulin formulation,
the first spray dried aerosol product, the first unit dose
blister powder packaging system, the first fine powder
filling system and the first inhalation device as reliable
as injections. The development of the product was
prolonged by safety concerns (immunogenicity and
lung function effects) which were eventually overcome
by extensive long term data. Now numerous
companies are working on their own inhaled insulins.
The past 15-20 years have seen a large number of
innovations in the science and engineering of aerosol
medicines. In the area of dry powders, with the use of
special safe excipients, particle engineering technology
is now to the point where almost any therapeutic
molecule can be formulated into a dispersible, stable
powder which can be used in small breath-activated
devices. In addition, with the new highly dispersible
powders, the mass of drug that can be inhaled into the
lungs in one breath, has increased from 100s of
micrograms (ugs) to >30 milligrams (mgs), which now
enables the delivery of anti-infectives in a few breaths
as compared to 20-30 minute nebulizer regiments.
Spray drying has now become the method of choice for
controlling the size, shape and solid state of small
inhalable powders (1-3 microns).
Among the liquid delivery systems the vibrating mesh
is proving to be an important advance in the generation
of “soft mist” aqueous aerosols. In contrast to the older
inefficient jet nebulizers which require large amounts
of compressed air, the vibrating mesh can generate low
velocity high efficiency aerosols essentially air free.
This enables anti-infectives to be delivered to
ventilated patients who are at high risk for pneumonia
without having to interrupt the ventilator breathing
settings.
Other innovations include the development of
combination products which contain two drugs in the
same powder and solution formulations in the metered
dose inhaler systems (MDIs, the traditional small
asthma inhalers) where ethanol is used to solubilize
drugs in the hydrofluoroalkane (HFA) propellants.
Finally, the immunoglobulin receptor in the airways
has been shown to actively and efficiently shuttle
potent therapeutic proteins (i.e. interferons,
erthropoietin) into the systemic circulation from the
airways via large molecular weight, immunoglobulin
fragment – therapeutic protein conjugates.
____________
Patton J S, Byron, P.R. Nature Reviews/Drug
Discovery 6:67 (2007)
Weers J, et al. Exp. Opin. Drug Delivery (in press)
(2007)
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
THURSDAY PLENARY
The Devil is in the Details: On the Role of
Molecular Structure in Secondary Organic Aerosol
Chemistry. Paul Ziemann, University of California,
Riverside, CA
Measurements indicate that a significant fraction of the
mass of atmospheric aerosol particles is organic matter,
the majority of which consists of oxidized compounds
that are the products of gas-to-particle conversion
(secondary organic aerosol, SOA). The chemical and
physical processes involved in the formation of SOA
are complex and can include reactions of volatile
organic compounds (VOCs) with various atmospheric
oxidants (primarily O3, and OH and NO3 radicals), as
well as surface and condensed-phase reactions,
homogeneous nucleation, and gas-particle partitioning.
It should come as no surprise that understanding and
accurately modeling these processes is a major
challenge that has not yet been achieved. In this talk, I
will focus specifically on the impact of VOC molecular
structure on SOA chemistry. Using examples from
laboratory studies, I will demonstrate some of the ways
in which changes in structure can alter SOA products
and yields (which in turn can affect particle properties
such as hygroscopicity, CCN activity, light scattering
and absorption, and toxicity), and suggest explanations
for these effects based on current understanding of
chemical reaction mechanisms.
FRIDAY PLENARY
CNN, Clusters, Nanoparticles and Nucleation:
Connecting the Dots, M. Samy El-Shall, Virginia
Commonwealth University, Richmond, VA
Nucleation is one of the most ubiquitous and important
phenomena in science and technology. It plays a
central role in the formation of clusters, nanoparticles
and crystal growth. In spite of the fundamental and
applied interest in nucleation, the nucleus for
condensation remains one of the most elusive entities
known in chemical physics, and has never been
observed directly. Only the consequences of its
presence, e.g. droplet formation, precipitation, etc. are
observed. For example, in vapor phase nucleation
studies, the nucleation rate is often obtained by
measuring the rate of production of macroscopic liquid
droplets from the vapor phase.
A perspective of this important phenomenon will be
addressed with particular emphasis on ion-induced
nucleation. The application of Resonant Enhanced
Multiphoton Ionization (REMPI) in supersaturated
vapors to selectively generate specific ions of interest
and study their nucleation behaviors will be discussed.
The REMPI nucleation method has tremendous
amplification and detection capabilities that can
provide valuable and novel analytical tools for the
identification of trace components in the vapor phase.
Other examples of nucleation-based processes dealing
with the formation and properties of molecular clusters
and cluster ions, vapor phase synthesis of nanoparticles
and polymer nanocomposites, and the condensation of
supersaturated vapors on nanoparticles are discussed.
The study of gas phase clusters provides
information on how the properties of matter evolve as
the size of a material system ranges from molecular to
macroscopic dimensions. Detailed information on the
structures and conformational changes of molecular
and cluster ions can be obtained using the massselected ion mobility technique. The mobility
measurements provide structural information on the
ionized clusters and oligomers on the basis of their
collision cross-sections, which depend on the
geometric shapes of the ions. The application of ion
mobility to investigate the polymerization of ionized
acetylene clusters and the formation of Polycyclic
Aromatic Hydrocarbons (PAHs) will be discussed. The
cluster reactions of small molecules provide novel
mechanisms for the formation of large PAH and
complex molecular ions.
The application of the laser vaporization controlled
condensation (LVCC) technique coupled with a
differential mobility analyzer (DMA) to synthesize
size-selected semiconductor, metal and intermetallic
nanoparticles from the vapor phase will be addressed.
The assembly of nanoparticles in the vapor phase into
filaments and fibers in the presence of an electric field
will be discussed. Enormous electrostatic interaction
due to dipole forces is observed between nanoparticles
to form chain filaments, and between the chains to
form tree-like assemblies. The filaments display stretch
and contraction properties depending on the strength of
the applied field. These observations have significant
implications to the ductility and the plastic behavior of
the materials formed from consolidated nanoparticle
assemblies.
Finally, a new technique to study the condensation
of supersaturated vapors on nanoparticles under welldefined conditions of vapor supersaturation,
temperature and carrier gas pressure will be presented.
The nanoparticles can be activated to act as
condensation nuclei at supersaturations significantly
lower than those required for homogeneous nucleation.
The question of the sign effect in the condensation of
supersaturated vapors on charged nanoparticles will be
addressed.
This presentation seeks to connect the fields of gas
phase clusters, nucleation and nanoparticles and place
the integrated fields in context with the aerosol
formation mechanisms, properties and applications.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.1
Simultaneous On-line Size and Chemical Analysis of Gas
Phase and Particulate Phase of Mainstream Tobacco Smoke.
JOHN McAUGHEY, Conor McGrath, British American
Tobacco; Thomas Adam, Christoph Mocker, Ralf Zimmermann,
University of Augsburg.
Tobacco smoke is a complex and dynamic matrix
consisting of gaseous and particulate material, in which
about 4800 constituents have been identified. The
chemical composition and partition between phases of the
smoke can change continuously and is strongly influenced
by time, temperature, chemistry and dilution of smoke.
We present an experimental set-up consisting of gas phase
and particulate phase on-line instrumentation for
comprehensive analysis of mainstream tobacco smoke,
that is, the smoke that emerges from the mouth end of the
cigarette during a puff.
Cigarettes comprising Burley, Virginia or Oriental
tobacco at 3 filter ventilation levels were smoked at two
flow regimes, with particle diameter and concentration
measured by electrical mobility. Chemical composition
was characterised on-line by resonance-enhanced multiphoton ionisation (REMPI) and single photon ionisation
(SPI) techniques with time-of-flight mass spectrometry
(TOFMS).
Count median diameter (CMD) averaged over the
cigarette varied from 182 - 260 nm and increased with
increasing filter ventilation and lower puff flow rates; a
consequence of increasing smoke residence time and
coagulation within the rod. Puff-by-puff data showed
increasing particle concentration and decreasing diameter
as the tobacco was consumed and the coagulation period
decreased.
Initial mass spectrometry data show that most smoke
constituents feature a continuous increase from the first to
the last puff. However, there are some substances, in
particular unsaturated hydrocarbons e.g. butadiene,
isoprene, and propyne, which show a completely different
behaviour by having the highest amounts in the first puff.
This is likely to be related to the different combustion and
pyrolysis conditions when the cigarette is lit.
11A.2
Highly Time-Resolved Ambient Measurements of Organic
Molecular Markers and Air Toxics in Pittsburgh Using
Thermal Desorption Aerosol GC-MS (TAG). ANDREW T.
LAMBE, Jennifer M. Logue, Allen L. Robinson, Neil M.
Donahue, Carnegie Mellon University; David R. Worton, Brent
J. Williams, Allen H. Goldstein,University of California,
Berkeley; Nathan M. Kreisberg, Armond Gauthier, Susanne V.
Hering, Aerosol Dynamics Inc.
A significant fraction of airborne PM2.5 consists of
organic carbonaceous material, with hundreds of
individual compounds identified in ambient aerosols
including source specific molecular markers and known
carcinogens such as PAHs. Speciation of organic aerosol
traditionally involves offline GC-MS analysis of solvent
extracted samples, which is labor-intensive and results in
measurement resolution of 12-24 hours. This study builds
on work introduced by Williams et al (AS&T, 2006; JGR,
2007) using Thermal Desorption Aerosol GC-MS, a novel
method for automated collection of ambient particles with
online GC-MS analysis. TAG affords hourly resolution
and has the advantage of fully automated operation,
therefore holding promise as an alternative to
traditional organic speciation techniques.
TAG has two modes of operation: ambient sampling with
concurrent GC-MS analysis of the previously collected
sample, and thermal desorption of the previously
collected sample onto the GC column. Particles are
humidified to increase adhesion and minimize bounce
before inertial impaction onto a collection substrate.
During ambient measurements, sampling is set on a 26hour cycle that includes collection of ambient samples as
well as filtered ambient samples to determine contribution
of gas-phase compounds and zero air blanks to determine
internal contamination levels. In this work we monitor
ambient aerosol loadings in the vicinity of a large
construction project on the
Carnegie Mellon campus. These highly time-resolved
measurements provide insight into the relative
contribution of local point sources to an area dominated
by regional transport. The extent of this contribution
would be more difficult to discern using traditional
analysis techniques with measurement resolution
approaching regional mixing timescales, and
demonstrates an application uniquely suited to TAG. We
also evaluate instrument performance for nonpolar
compounds (i.e. n-alkanes, hopanes and steranes, PAHs)
relative to solvent extracted samples collected
concurrently during ambient measurements and smog
chamber experiments.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.3
Analysis of Organic Aerosols Using Methods of Highresolution Mass Spectrometry. YURY DESYATERIK,
Pacific Northwest National Laboratory; Maggie L. Walser,
Sergey A. Nizkorodov, University of California, Irvine; Julia
Laskin, Alexander Laskin*, Pacific Northwest National
Laboratory.
Traditional methods for characterization of the
chemical composition of PM2.5 are limited to relatively
low molecular weight (MW) thermally stable compounds.
However, recent studies demonstrated that up to 80% of
unidentified organic matter in field-collected aerosols
corresponds to high-MW humic-like substances (HULIS).
Development of new analytical approaches is crucial for
accurate determination of chemical composition of PM2.5
and estimating the effects of prescribed burning on the air
quality and climate. This presentation reports on our
efforts aiming to develop a novel analytical platform for
unequivocal identification of high-MW molecules in
aerosol samples using high-resolution mass spectrometry
combined with electrospray ionization. The presented
work is focused on characterization of the chemical
composition of secondary organic aerosol formed during
the ozonolysis of undecylenic acid and limonene in
laboratory experiments, and then on molecular speciation
of organic compounds characteristic of field collected
biomass burning aerosols. Accurate mass measurement
using high-resolution (100,000 mass resolution) mass
spectrometry is used for determination of the elemental
composition of high-MW constituents while tandem mass
spectrometry (MS/MS) experiments are used for
structural characterization of complex molecules. High
mass resolving power of the LTQ/OrbitrapTM-MS
instrument allows us to assign the empirical elemental
composition of hundreds to thousands of peaks in each
given sample. The complexity of the spectra can be
partially reduced using the Kendrick diagram, in which
groups of homologous species that differ by the number
of repeating or additional groups (e.g, -CH2-,-CH=CH-, OH, -COOH etc.) can be identified.
11A.4
Measurements of Organic Nitrogen Budget in Atmospheric
Aerosol. ANDREY KHLYSTOV, Ming-Yeng Lin, Duke
University.
Ammonium and nitrate are the main inorganic nitrogencontaining aerosol constituents. However, measurements
of only inorganic nitrogen do not provide complete
information on the total nitrogen content of ambient
aerosol. Despite growing evidence that organo-nitrogen
compounds may constitute a significant fraction of the
aerosol nitrogen, very little is known about this fraction.
We report high time resolution measurements of chemical
composition of PM2.5 aerosol at the Duke Forest
Research Facility with the emphasis on quantification of
the contribution of organo-nitrogen compounds (ONC) to
the total nitrogen budget in the aerosol as well as its
dynamics as a function of ambient temperature and
relative humidity. The main water-soluble inorganic
components (SO42-, NO3-, Cl-, NH4+, Na+, K+), total
and water soluble carbon, as well as organic nitrogen are
measured using a modified Steam-Jet Aerosol Collector
(SJAC). The total concentration of aerosol carbon and
nitrogen are measured using a high sensitivity TOC/TN
unit. The concentration of organo-nitrogen in the aerosol
is determined as the difference between the total nitrogen
concentration and the concentration of inorganic nitrogen
species. The time resolution of the measurements is 30
min.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.5
Characterization of Nitrogen Containing Organic Species in
Atmospheric Aqueous Samples and Aerosol Particles Using
a High Resolution Time-of-Flight Aerosol Mass
Spectrometer. Yele Sun, QI ZHANG, University at Albany,
SUNY.
Despite the fact that nitrogen-containing organic species
are ubiquitous in atmospheric particles and water droplets
and likely play important roles in atmospheric chemistry
and ecosystem health, very little is known about the
concentration or composition of this class of compounds.
Here we present the development of a new method that
allows the quantification and bulk-characterization of
water-soluble organic nitrogen (WSON) species in
atmospheric fog and cloud waters using an Aerodyne
High-Resolution Time-of-Flight Aerosol Mass
Spectrometer (HR-ToF-AMS). The success of this
method relies on the facts that 1) the AMS mass spectra
can be deconvolved to quantify and chemically
characterize organic material in complex mixtures such as
atmospheric aerosol particles and 2) the high resolution
capability of the HR-ToF-AMS can distinguish nitrogen
containing organic fragments (e.g., CxHyNp+ and
CxHyOzNp+) from ion fragments lack of N atom (e.g.,
CxHy+ and CxHyOz+). We evaluate this method through
HR-ToF-AMS analyses of a suite of WSON standard
compounds, including 19 amino acids, urea, peptides and
proteins. The elemental ratios of C:H:O:N are determined
in the spectrum of each compound and compared to the
theoretical values. In addition, we will present the
application of this technique in analyses of atmospheric
fog and cloud water samples and aerosol extracts.
Attempts will also be made to characterize different
classes of WSON species (e.g., amino compounds vs.
amides) in atmospheric samples based on mass spectral
signature identified in standard compounds.
11A.6
A Community Software for Quality Control and Analysis of
Data from the Aerodyne Time-of-Flight Aerosol Mass
Spectrometers (ToF-AMS). DONNA SUEPER, Aerodyne and
University of Colorado, Boulder; James Allan, University of
Manchester; Edward Dunlea, University of Colorado, Boulder;
Jonny Crosier, University of Manchester; Joel Kimmel, Peter
DeCarlo, Allison Aiken, Jose-Luis Jimenez, University of
Colorado, Boulder; Doug Worsnop, Aerodyne.
The ToF-AMS (Drewnick et al., 2005) and highresolution ToF-AMS (HR-ToF-AMS, DeCarlo et al.,
2006) are the new versions of the Aerodyne AMS, and are
rapidly superseding the quadrupole AMS (Q-AMS). The
data quality control and analysis tasks are enormous for
ToF-AMS datasets, due to the size (a typical dataset size
is 25 GB) and dimensionality (3-4 mathematical
dimensions) of the data produced. A software tool for
ToF-AMS data management and analysis called
SQUIRREL will be presented (Sueper et al., 2007). As
was the case with the Q-AMS analysis software,
SQUIRREL is also shared with the entire AMS Users
Community and improved with feedback from all users.
This leads to faster development and ensures consistent
processing of ToF-AMS data. The architecture of
SQUIRREL is centered on the premise that most data are
kept in the computer hard drive, as ToF-AMS datasets are
too large to completely reside in computer memory. The
ToF-AMS data acquisition software saves the data in
Hierarchical Data Format (HDF5), a highly structured
binary format that was developed by NCSA with NASA
funding for satellite applications. HDF5 files can be
accessed randomly, which greatly reduces seek times over
the text file system used in the Q-AMS acquisition and
analysis software. The name "SQUIRREL" reflects the
ability of the software to quickly move pieces of the data
("acorns") between the memory and the hard drive. The
basic principles of data analysis are adapted from those
from the Q-AMS (e.g. Allan et al., 2003, 2004). Along
with flexible and detailed mass calibration and baseline
fitting routines, SQUIRREL incorporates techniques to
handle high-speed recording and multiple ionization
schemes. Additional modules can be built upon
SQUIRREL for specialized analyses. An example of
such a module is the code to characterize the chemical
information contained in the high-resolution mass spectra
from the HR-ToF-AMS. Examples of application of
SQUIRREL and HR-SQUIRREL to real datasets, and
directions for future development will be given.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.7
Application of Positive Matrix Factorization (PMF) to
Aerosol Mass Spectrometer (AMS) Data: Pittfalls and
Results. Ingrid Ulbrich, JOSE L. JIMENEZ, Katja Dzepina,
Kenneth Docherty, University of Colorado-Boulder; Qi Zhang,
SUNY-Albany; Manjula Canagaratna, Douglas Worsnop,
Aerodyne Research; Dara Salcedo, Univ. Estado Morelos.
11A.8
Investigation of biomass combustion aerosol by H-NMR
spectroscopy. James Hutchings, Pierre Herckes, Arizona State
University; GAVIN MACMEEKING, Sonia Kreidenweis,
Jeffrey L. Collett, Jr., Colorado State University; Wei Min Hao,
Cyle Wold, US Forest Service; W.C. Malm, National Park
Service.
Our understanding of organic aerosols (OA) is rapidly
evolving, partially due to the influx of data from new realtime techniques. Zhang et al. (ES&T, 2005; ACP, 2005)
first applied component analysis techniques to AMS field
data, leading to the recognition of the dominance of
oxygenated organic aerosols (OOA) even in urban areas.
Several new component analysis techniques have been
recently applied to AMS datasets (Zhang et al., GRL,
2007; Lanz et al., ACP 2007; Marcolli et al., ACP 2006).
PMF is promising due to the non-negativity constraint
that reduces the rotational freedom of the solutions, the
availability of reliable software, and the accumulated
experience of community in its use. However PMF does
not produce a unique solution, but rather a matrix of
solutions (with increasing number of factors and values of
the rotational parameter FPEAK) from which the user has
to choose the "most realistic one." Although many OA
sources are known to make a contribution to ambient OA
levels, it is not clear how finely these can be retrieved
using unit-resolution EI AMS data. Using synthetic data,
we show that PMF solutions with too many factors still
produce realistic-looking time series and mass spectra,
due to "mixing" of the real components and to "splitting"
of the real components into subcomponents. Splitting
tends to yield similar mass concentrations on both new
components. The evaluation of PMF solutions through
synthetic analysis is critical to the believability of the
results. We apply PMF to ambient datasets from
Pittsburgh, Riverside, and Mexico City. PMF reveals a
small semivolatile OOA II component (8% of the OA
mass) in Pittsburgh, which doesn't change the
interpretation of this dataset by Zhang et al, (2005ab).
PMF solutions for Pittsburgh with more than 3
components appear to split the main components, rather
than to find real new components.
Biomass combustion is an important emission source of
particulate matter into the atmosphere with local (haze)
and global (climate) impacts. In recent years many efforts
have been made to characterize the particulate emissions
in terms of size, hygroscopicity, optical and chemical
properties. Chemical characterization has mainly focused
on speciation studies, quantifying individual species while
the bulk of the organic matter remained poorly
characterized beyond water solubility. Consequently little
is known on how the chemical properties including
functionality impact physical characteristics like
hygroscopicity and optical properties.
We investigated particulate matter emissions from
biomass burning through a series of large scale laboratory
experiments aimed at understanding the impact of fuel
and combustion regime on chemistry and optical
properties. For the present work, biomass samples from a
variety of fuels and burning regimes were extracted with
deuterium oxide. These aqueous extracts were then
characterized by proton nuclear magnetic resonance
spectroscopy (H-NMR). The results give insights on the
variability of the chemical structure of the water soluble
organic carbon (WSOC) fraction by quantifying the
importance of different types of protons (e.g., aromatic vs
aliphatic). Beyond functional group information, major
organic species including levoglucosan can be identified
and quantified at a molecular level using the discrete
features in the H-NMR spectra. Finally, we will discuss
how these new insights in the structure and composition
relate to the variability in optical and hygroscopic
properties observed in biomass burning aerosol.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.9
Cross flow ion mobility spectrometry. MANG ZHANG,
Anthony S Wexler, University of California, Davis.
A new instrument, the cross flow ion mobility
spectrometer (CF-IMS) was constructed and tested. CFIMS is a light-weight, low cost instrument that can
analyze the composition of the gas phase continuously
and with high mobility resolution. Unlike the traditional
ion mobilities spectrometers (IMS) and differential
mobility analyzers(DMA), CF-IMS uses a higher flow
velocity and parallel plated configuration that decreases
the characteristic dimension in the Reynolds number to
achieve higher resolution. Its electrometer sensor array
eliminates voltage scanning in the DMA and the shutter
gate in IMS, both of which decrease duty cycle in these
instruments.
Three different CF-IMS prototypes which include two
channel, 20 channel and 256 channel electrometers is
tested with different chemicals. Result will be compared
and presented.
11A.10
A New Automated Monitor for the Measurement of
Particulate Reactive Oxidant Concentrations in the
Atmosphere. PRASANNA VENKATACHARI, Philip K.
Hopke, Clarkson University.
The global burden of disease as a consequence of ambient
particulate matter (PM) remains a growing threat, and
efforts to identify and link specific components of the PM
mix with various PM-associated health effects are
intensifying. A causal hypothesis for particle toxicity is
through oxidative challenges to the lung, resulting in the
generation of reactive oxidative species (ROS) at the
target sites. However, research conducted over the past
few years has shown that ROS is present on ambient
particles to which we are exposed. It needs to be
recognized that ROS present on particles can cause the
same kind of systemic dysfunctions as endogenously
produced ROS. Few measurements have been made of
the particulate oxidant concentrations due to the
impracticability of the manual measurement methods, and
the difficulty of obtaining timely and accurate
measurements is an obstacle for the research and
regulatory communities. As a result, an automated
instrument is needed.
In atmospheric applications, the requisite sensitivity
involved in the trace determinations of atmospheric
oxidants limits the choice to luminescence methods. The
photoluminescence procedure involving the oxidation of
the non-fluorescent form of dichlorofluorescein to its
fluorescent form by ROS in the presence of a peroxidase
enzyme was found to be a sensitive, non-specific,
attractive method. A practical sample collection-flow
injection system for the semi-continuous measurement of
the total particulate oxidant concentrations in the ambient
atmosphere based on this dichlorofluorescein-oxidantperoxidase fluorescence reaction has been developed.
This monitor allows the determination of the distribution
of ambient particulate ROS as a function of location, time
of day, and day of year, and will ultimately aid in the
statistical evaluation of the role of particle-bound oxidants
in the overall toxicological impact of PM. The system
configuration, validation of its performance, and,
measurements of atmospheric oxidant concentrations with
this monitor will be presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
11A.11
Contribution of Carboxylic Acids in Ambient Aerosol to the
m/z 44 Signal of an Aerodyne Aerosol Mass Spectrometer.
NOBUYUKI TAKEGAWA, Takuma Miyakawa, Masamichi
Watanabe, Yutaka Kondo, RCAST, University of Tokyo;
Kimitaka Kawamura, Hokkaido University.
The Aerodyne aerosol mass spectrometer (AMS) employs
flash vaporization (600C) followed by 70-eV electron
impact ionization to detect organic and inorganic aerosols.
The signal at mass-to-charge ratio (m/z) 44 (mainly CO2
+) is considered the most reliable marker of oxygenated
organic aerosol, especially for carboxylic acids. We
estimate the contribution of selected low molecular
weight dicarboxylic acids (diacids) and omegaoxocarboxylic acids (omega-oxoacids) to the particlephase m/z 44 signal of the AMS mass spectrum. Ambient
measurements were conducted at a surface site in Tokyo
(35.39N, 139.40E) during August 3-8, 2003. Diacids and
omega-oxoacids were measured using a filter sampling
followed by extraction, derivation, and gas
chromatograph - flame ionization detector (GC-FID)
analysis. The mass concentrations of diacids and omegaoxoacids show tight correlation with the m/z 44 signal (r2
= 0.85-0.94) during the measurement period. Laboratory
experiments were performed to determine the fragment
patterns of selected diacids (C2-C6 diacids and phthalic
acids) and omega-oxoacid (glyoxylic acid) in ambient
aerosols. We have found that the selected organic acids
could account for 14% of the observed m/z 44 signal on
average during the measurement period. Oxalic acid (C2)
is the largest contributor, accounting for 10% of the
observed m/z 44 signal. The mass spectra of other
carboxylic acids, which include monocarboxylic acids
(monoacids) and polycarboxylic acids (polyacids), have
been investigated in the laboratory. Although these
monoacids and polyacids were not measured during the
ambient measurement, possible contributions from these
compounds are also discussed to explore the missing
source of the m/z 44 signal in ambient aerosols.
11A.12
Low-Pressure Chemical Ionization Mass Spectrometry of
Ultrafine Aerosols. SONYA C. COLLIER, Angela I. Shibata,
Denis J. Phares, University of Southern California.
Chemical analysis of organics requires an ionization
mechanism that minimizes the energy transferred to the
molecule that may lead to fragmentation. Chemical
ionization represents one such \soft\ ionization scheme.
Maintaining the chemical reaction region at low pressure
(1 to 10 Torr) minimizes the ion clustering that occurs at
atmospheric pressure, thereby allowing for easier
identification of the protonated organic molecule. In this
study, we apply low-pressure chemical ionization time-offlight mass spectrometry to the analysis of ultrafine
aerosols. Particles are classified and collected using an
inlet that resembles a cylindrical Differential Mobility
Analyzer (DMA) in that the sample flow is introduced
around the periphery of the annulus between two
concentric cylinders, and charged particles migrate
inward, depositing on a Nichrome filament. The particles
are desorbed, and the resulting vapor is passed into the
low-pressure chemical ionization cell. The ions are then
focused into an orthogonal extraction TOF mass
spectrometer, providing a mass spectrum for the sizeresolved aerosol. Results are presented for organic aerosol
standards ranging in size from 20 nm to 200 nm.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
12A.1
Bridging the Gap Between Top-Down and Bottom-Up
Characterization of Organic Aerosols. MURRAY
JOHNSTON, Matthew Dreyfus, Katherine Heaton, Julie Lloyd,
Christopher Zordan, University of Delaware.
Ambient organic aerosol is complex and may contain
hundreds or thousands of individual compounds. There
are two main ways to handle this complexity: 1) bottomup, where the distribution of molecular components is
measured and properties of the total, \macroscopic\
aerosol are inferred, and 2) top-down, were macroscopic
properties are measured and the distribution of molecular
components is inferred. Usually, molecular level
measurements are performed off-line with low time
resolution. An advantage of these measurements is that
individual \marker\ compounds can be linked with
specific sources, facilitating source apportionment. In
contrast, macroscopic measurements are usually
performed on-line with high time resolution. Advantages
of these measurements are that short term variations can
be associated with meteorological conditions or transient
events and macroscopic physical properties can be
inferred. Bridging these two views and the information
they provide requires additional analytical capability,
specifically the ability to measure molecular distributions
with high time resolution.
Our group has developed an aerosol mass spectrometer
that allows molecular distributions to be measured with a
time resolution of a few minutes. The current version of
the instrument uses photoionization with vacuum
ultraviolet radiation to characterize relatively nonpolar,
semivolatile components. A new version of the
instrument uses chemical ionization and matrix assisted
laser desorption ionization to characterize polar and
macromolecular components. In either case, rapid
changes in the distribution of molecular components can
be directly compared with macroscopic chemical
measurements such as OC/EC and with meteorological
parameters. Our group has also developed an aerosol
mass spectrometer that allows the elemental composition
of individual particles to be measured with high time
resolution. The molecular distribution can be compared
with the macroscopic elemental composition, both
obtained with high time resolution, to determine how
representative the measured molecular distribution is of
the total aerosol. Results of laboratory and ambient
studies will be discussed.
12A.2
Tracing the Sources and Transformations of Oxidized
Organic Aerosols in the Atmosphere by Spectroscopic
methods: Results from Functional Group Analysis. Stefano
Decesari, MARIA CRISTINA FACCHINI, Sandro Fuzzi,
Emanuela Finessi, Italian National Research Council, Italy;
Fabio Moretti, Emilio Tagliavini, Centro Interdipartimentale di
Ricerca per le Scienze Ambientali, University of Bologna, Italy;
also at Department of Chemistry, University of Bologna, Italy.
The recent development of high-time resolution aerosol
MS (AMS) techniques has provided a new insight to the
sources and transformation of organic aerosols in the
atmosphere, by showing that spectral fingerprints can be
identified for specific fresh emissions and for processed,
chemically aged aerosols. These findings are in
substantial agreement with the results of functional group
analysis by nuclear magnetic resonance (NMR)
spectroscopy, showing that the \bulk\ chemical
composition - and not only molecular tracers - can retain
information about the sources of aerosol organic
compounds. Analogously to the results of AMS
measurements, distinct NMR compositions of oxidized
organic aerosols can be found for biomass burning
emissions and for SOA formation. Moreover, specific
spectral fingerprints of fresh biogenic SOA can be
extracted from field data collected in clean continental
forest sites, and linked to analogous features found for
synthetic biogenic SOA formed in smog chambers.
Conversely, both AMS and NMR spectral fingerprints of
oxidized organic aerosols in polluted environments could
not be reproduced by SOA formed in smog chambers
experiments employing high concentrations of aromatic
hydrocarbons as precursors. NMR functional group
compositions deviating from the oxidized organic aerosol
types identified by AMS were also observed, especially in
the free troposphere at the mid-latitudes in wintertime,
and in the marine boundary layer. Finally, NMR analysis
shows significant changes in the composition of oxidized
organic aerosols in periods of high photochemical
activity, that can only partly be attributed to the chemical
aging effects - such as the oxidation of carbonyls to
carboxyls - already shown in the AMS data record.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
12A.3
Secondary Organic Aerosol Formation Through Cloud
Processing: Acids and Oligomers from Aqueous
Methylglyoxal Photooxidation. Katye Altieri, Annmarie
Carlton, EPA; Yi Tan, Sybil Seitzinger, BARBARA TURPIN,
Rutgers University.
There is growing evidence that secondary organic aerosol
(SOA) forms through cloud processing. Specifically,
organic emissions are oxidized to form water-soluble
compounds that partition into cloud water and oxidize
further to form low volatility products. These products
remain, at least in part, in the particle phase after droplet
evaporation, forming SOA (Blando, Atmos. Environ.,
2000). This process could help explain the gap (Heald,
JGR, 2006) between measured and modeled organic PM
in the free troposphere. In-cloud SOA predictions have
mostly assumed aqueous oxidation pathways and products
(Ervens, JGR, 2004; Lim, EST, 2005). Since then,
laboratory experiments have verified that low volatility
products form from aqueous photooxidation of pyruvic
acid and glyoxal (Carlton, GRL, 2006; Altieri, EST, 2006;
Carlton, Atmos. Environ., 2007). Field measurements also
provide evidence for in-cloud formation (i.e., of oxalic
acid; Crahan, Atmos. Environ., 2004; Sorooshian, JGR,
2006).
In this work, aqueous-phase photooxidation experiments
were conducted with methylglyoxal and hydroxyl radical
(from the reaction of hydrogen peroxide with UV; 254
nm). Methylglyoxal is a major water-soluble product of
gas-phase isoprene oxidation. Precursors and products,
analyzed by HPLC, ESI-MS, FT-ICR-MS, and ESI-MSMS, included pyruvic, acetic, formic, glyoxylic and oxalic
acids and oligomers. A modified aqueous-phase reaction
mechanism is proposed for methylglyoxal that is
consistent with the measured concentration dynamics of
products, including pathways leading to the formation of
oligomers (i.e., through radical-initiated reactions) and
oxalic acid. Oligomeric products have higher molecular
weights (m/z 100-500) and somewhat lower organic mass
(OM) to organic carbon (OC) ratios (1.6-2.1) than the
organic acid products (OM/OC=2.0-3.8). Thus we expect
the oligomers to be somewhat less hygroscopic than the
organic acid products. The ESI-MS-MS fragmentation
patterns of the oligomers are consistent with the presence
of carboxylic acid functionalities.
12A.4
Comparison of Organic Functional Groups from FTIR and
Organic Mass Fragments from AMS at Six North American
Field Studies. LYNN M. RUSSELL, Stefania Gilardoni, Lelia
N. Hawkins, Scripps Institution of Oceanography, UCSD; Tim
S. Bates, Pacific Marine Environmental Laboratory, NOAA;
James D. Allan, University of Manchester; Darrel
Baumgardner, National Autonomous University of Mexico;
Peter F. DeCarlo, Edward Dunlea, Jose L. Jimenez, University
of Colorado at Boulder; Tim B. Onasch, Doug R. Worsnop,
Aerodyne Research Inc.
This study compares the organic composition of six
different sets of collocated Fourier Transform Infrared
(FTIR) spectrometry of bulk submicron particle filter
samples and Aerosol Mass Spectrometer (AMS) online
size-resolved submicron particle measurements during
field projects. The comparison includes about 30 days of
data from each of the following six field studies:
International Consortium for Atmospheric Research on
Transport and Transformation (ICARTT) during July and
August 2004 at Chebogue Point and aboard the R/V
Ronald Brown, Megacities Impact on Regional and
Global Climate (MIRAGE)/Megacity Initiative: Local
and Global Research Observations (MILAGRO) during
March and April 2006 aboard the NCAR C130, Paso de
Cortes during March and April 2006, Intercontinental
Chemical Transport Experiment (INTEX-B) during April
and May 2006 aboard the NCAR C130, and Texas Air
Quality Study / Gulf of Mexico Atmospheric
Composition and Climate Study (TEXAQS/GoMACCS)
during July and August 2006 aboard the R/V Ronald
Brown. The FTIR measures wavelength-dependent
carbon bond absorption by transmission to estimate
functional group concentrations of aromatic C=C-H,
unsaturated aliphatic C=C-H, saturated aliphatic C-C-H,
organic hydroxyl O-H, organosulfur C-O-S, and carbonyl
C=O using calibrated standards. The AMS measures
mass fragments of compounds, including carbon,
hydrogen, oxygen and sulfur atoms. The correlations of
functional groups with mass fragments that result from
the comparison of FTIR and AMS measurements provide
enhanced information about the structure of the organic
compounds in ambient aerosol particles.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
12A.5
Introducing the Concept of Potential Aerosol Mass. Eunha
Kang, WILLIAM H. BRUNE, Magaret Root, Pennsylvania
State University; Darin Toohey, University of Colorado.
We introduce a new concept: potential aerosol mass
(PAM). PAM can be defined as the as the maximum
aerosol mass that precursor gases can be oxidized to form
particulate matter. This concept has many uses. In the
atmosphere, PAM can be used along with chemical
measurements in the gas and particle phases to better
understand the integrated secondary organic aerosol
(SOA) formation and evolution and to test the
completeness of the measured SOA sources, aerosol mass
yields, and oxidation pathways. When placed in networks,
the sum of aerosol mass and PAM can be used to better
understand the sources and distribution of SOA. In
environmental chambers, it can be used in a variety of
ways to understand the behavior and completion of SOA
formation and subsequent oxidation.
The PAM measurement consists of passing air
containing aerosol-precursor gases through a small
chamber that is irradiated with ultraviolet lamps. Rapid
and complete oxidation ensues in the extreme oxidizing
environment, with measured values of about 10 ppmv of
O3, 100 pptv of OH, and 2 ppbv of HO2. The airflow
exiting the chamber is sufficient that a wide range of
detection has been used, including a Tapered Element
Oscillating Microbalance (TEOM), an aerosol mass
spectrometer (AMS), and a variety of particle sizing
instruments.
We present laboratory studies that demonstrate the
feasibility of the PAM concept, with emphasis on
applications to SOA. These studies include dependence of
aerosol mass yields on oxidant levels, relative humidity,
and UV light; measurements of aerosol mass yields for
several anthropogenic and biogenic hydrocarbons; and the
chemical and size evolution of SOA in the PAM chamber.
We discuss improvements in the PAM measurement
technique that will make it useful for an even wider array
of applications.
12A.6
Developement and Application of a Soot Particle Mass
Spectrometer. Achim Trimborn, DAGMAR TRIMBORN,
Timothy Onasch, Manjula Canagaratna, Jesse Kroll, John Jayne,
Douglas Worsnop, Aerodyne Research, Inc.; Gregory Kok,
Droplet Measurement Technologies.
Black carbon containing aerosols play important roles in
governing the optical properties of atmospheric aerosol.
Absorption of incident solar radiation by elemental
carbon containing aerosol can lead to a cooling effect on
the ground, potentially offsetting the warming caused by
greenhouse gases. In addition, the absorption can lead to
heating of the airmass potentially affecting clouds. An
outstanding question is the role of coatings on black
carbon cores and how such coatings may alter the optical
properties. Instruments like the Single Particle Soot
Photometer (SP2) and Photoacoustic Spectrometer
(PASS) are well suited to characterize the black carbon
content inside internally mixed ambient particles (via
incandescence and absorption measurements). However,
these instruments lack the capability to measure the
chemical composition of absorbed species in or on black
carbon cores. We report here a new method for
measuring the chemical composition of absorbed
inorganic and organic material which is specific only to
particles containing black carbon. Black carbon particles
absorb and are heated by radiation in the 1 um wavelength
range. The vaporized constituents are then ionized by
electron impact ionization and detected using a time of
flight (TOF) mass spectrometric approach. We report a
series of measurements which demonstrate the utility of
this technique for a variety of particles with and without
black carbon cores and with various coatings. These
measurements clearly show that the method is highly
selective and sensitive only for particles containing an
absorbing core. The results also show that the mass
spectrometric signals vary linearly with the amount of the
condensed species. This allows for the quantification of
the chemical composition of this class of particles.
Signals for vaporized carbon atom clusters are also
observed and current work is aimed at interpreting the
significance of these signals.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
13A.1
Semivolatile Emissions and the Organic Aerosol Budget.
ALLEN L. ROBINSON, Neil M. Donahue, Carnegie Mellon
University.
Sources of primary organic aerosols (POA) such as motor
vehicles, wood combustion, and industrial processes emit
substantial quantities of semivolatile organics. The
amount of primary organic aerosol depends on the gas
particle partitioning of this complex mixture.
Photochemical aging of these emissions typically creates
more polar and less volatile reaction products, generating
secondary organic aerosol (SOA).
This talk combines recent field, laboratory, and modeling
results to discuss the influence of semivolatile emissions
on the organic aerosol budget. Although POA emissions
are commonly measured with dilution sampler, these
measurements are typically made at low dilution ratios.
Measurements with diesel exhaust and woodsmoke
indicate that this can substantially bias measured emission
factors because aerosol concentrations inside the dilution
sampler can be orders of magnitude higher than typical
atmospheric conditions. Unfortunately, emission
inventories and models currently treat the POA emissions
as non-volatile, implicitly assuming that the dilution
sampler measurements represent the full range of
atmospheric conditions. Accounting for partitioning of
POA in chemical transport models dramatically reduces
POA concentrations. Smog chamber experiments
indicate that semivolatile emissions may be an important
source of SOA. Current SOA models do account for
some SOA production from low volatility vapors, but
these vapors contribute little SOA compared to oxidation
products of light aromatics and biogenics. A comparison
of emission inventories with partitioning and speciation
data indicate that emissions of semivolatile organics are
grossly under-represented in the inventories, which have
largely been developed for simulating tropospheric ozone.
A major challenge is that the vast majority of the low
volatility organics cannot be speciated, and instead appear
as an unresolved complex mixture. The talk will
conclude with a discussion of how the basis set
framework can be used to better account for the effects of
both gas-particle partitioning and photochemical aging of
semivolatile emissions on the organic aerosol budget.
13A.2
Chemical Characterization of Low, Medium, and High
Volatility Biogenic Secondary Organic Aerosol Compoments
Using an Aerosol Mass Spectrometer. EVANGELIA
KOSTENIDOU, Spyros N. Pandis, Institute of Chemical
Engineering and High Temperature Chemical Processes and also
University of Patras; Byong-Hyoek Lee, Gabriella J. Engelhart,
Spyros N. Pandis, Carnegie Mellon University.
o
A thermodenuder operating at temperatures of 25-100 C
(An et al., 2007) is used to separate the secondary organic
aerosol (SOA) components by volatility. The chemical
composition of these components is then measured using
an Aerodyne Aerosol Mass Spectrometer (AMS). The
SOA is produced in the Carnegie Mellon smog chamber
by ozonolysis of alpha pinene, beta pinene and limonene.
The mass spectra of the low, medium, and high volatility
components are quantified and inter-compared.
Experiments are performed at low and intermediate RH
and at low and high NOx conditions. The effect of these
conditions on the mass spectra of the corresponding
components is quantified. As volatility decreases the
normalized organic fragments at m/z = 12, 17, 18 and 44
increase, while those at m/z = 15, 27, 29, 43, 55 and 57
show a decreasing trend.
An approach to deconvolute these spectra according to the
volatility basis-set modeling framework is developed and
applied to the data set. These AMS spectra of the various
SOA components separated by volatility can become
valuable tools in linking the smog chamber SOA
measurements with field observations of organic aerosol.
Woo Jin An, Ravi K. Pathak, Byong-Hyoek Lee and
Spyros N. Pandis (2007). Aerosol volatility measurement
using an improved thermodenuder: Application to
secondary organic aerosol. Journal of Aerosol Science,
Volume 38, Issue 3, Pages 305-314.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
13A.3
Volatility of Primary and Secondary Organic Aerosols:
Source and Field Measurements. J. ALEX HUFFMAN,
Allison C. Aiken, Ken Docherty, Ingrid Ulbrich, Jose L.
Jimenez, University of Colorado at Boulder Jesse Kroll,
Timothy Onasch, John T. Jayne, Douglas R. Worsnop,
Aerodyne Research, Inc. Paul Ziemann, University of California
- Riverside.
The volatility of organic species in atmospheric particles is
important for a number of reasons. Models need to properly
account for the gas-particle partitioning of organic species in
order to predict the organic aerosol (OA) concentrations in the
atmosphere. Semi-volatile and intermediate volatility organic
compounds (SVOC and IVOC) can react in the atmosphere to
produce significant amounts of SOA (Robinson et al, 2007).
These species are difficult to measure directly, but their relative
amounts can be inferred from the slopes of TD-volatility curves
near ambient temperature. Aerosol volatility also plays a large
role in both dry deposition and reaction losses, because dry
deposition and reactions are both typically faster for species in
the gas phase. Finally, knowledge of aerosol volatility allows an
estimation of losses of particle species due to ram and cabin
heating when sampling into aircraft.
A custom-built fast temperature-stepping thermodenuder (TD)
was coupled with an Aerodyne Aerosol Mass Spectrometer
(AMS) to allow the study of chemically-resolved volatility in
the lab and the field. The TD-AMS system was deployed during
field campaigns in Riverside, CA (SOAR-1: July-August, 2005),
Mexico City (MILAGRO/MCMA-2006: February 2006).
Particles from three additional sources were also characterized:
(a) primary biomass burning aerosols at the USDA Fire Sciences
Lab in Missoula, MT (FLAME-1: June, 2006), (b) secondary
organic aerosols (SOA) from the photochemical reaction
chamber at the University of California - Riverside, and (c)
primary anthropogenic particles from combustion and meatcooking sources in Boulder, CO. Urban SOA was less volatile
than urban POA. Biomass burning OA exhibited a wide
variability in its volatility, but in most cases was of similar or
larger volatility than urban POA. In each case of mixed aerosol,
the less oxygenated species were shown to evaporate at lower
temperatures than the more oxygenated species.
13A.4
Hourly Measurements of Organic Marker Compounds using
an In-Situ Thermal desorption Aerosol Gas chromatograph
(TAG). BRENT WILLIAMS, Allen Goldstein, University of
California Berkeley; Nathan Kreisberg, Susanne Hering,
Aerosol Dynamics Inc.; Laura Shields, Kimberly Prather,
University of California San Diego.
Thermal desorption Aerosol Gas chromatograph (TAG) is
a new in-situ instrument to identify and quantify organic
aerosol chemical composition with one hour time
resolution. Atmospheric particles are collected by means
of humidification and inertial impaction. The sample is
then thermally desorbed onto a GC column, where it is
separated into individual compounds which are identified
and quantified using a quadrupole mass spectrometer
(MS) and flame ionization detector (FID). With the
exception of periodic manually applied calibration
standards, TAG is fully automated, offering around the
clock measurements to determine diurnal, weekly, and
seasonal patterns in organic aerosol composition.
We report ambient aerosol measurements made in
southern California during the 2005 Study of Organic
Aerosol at Riverside (SOAR). We use hourly
measurements of over 300 individual organic compounds
to define both primary and secondary particle sources.
The compound classes include alkanes, branched alkanes,
alkenes, PAHs, branched PAHs, acids, ketones,
aldehydes, phthalates, furanones, terpenes, nitrogen
containing organics, sulfur containing organics, chlorine
containing organics, phosphorous containing organics,
and more. The particle sources defined include primary
anthropogenic sources such as vehicle emissions, meat
cooking, biomass burning, pesticide use, herbicide use,
along with primary biogenic sources such as plant
emissions and plant waxes. We also explore secondary
particle sources (i.e. SOA) formed as a result of the
oxidation of biogenic and anthropogenic precursor gases.
These sources are then compared with similar sources
independently defined by ATOFMS single particle
measurements. Finally, we present ambient air
observations of gas-particle phase partitioning as a
function of molecular size and functional groups.
These results contradict the assumptions in almost all current
models, which treat primary organic aerosol (POA) from
biomass or anthropogenic combustion sources as completely
non-volatile, and SOA as quite volatile. The use of a chemical
detector after the TD allowed this study to characterize the
volatility of bulk OA in ambient air for the first time.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
13A.5
Biomass Burning and Pollution Aerosol over North
America: Organic Components and their influence on
Spectral Optical Properties and Humidification Response.
ANTONY CLARKE, Cameron McNaughton, Vladimir
Kapustin, Yohei Shinozuka, Steven Howell, Jingchuan Zhou,
Vera Brekhovskikh, Mitchell Pinkerton, University of Hawaii;
Jack Dibb, University of New Hampshire; Bruce Anderson
NASA-LaRC; Harold Turner; University of Alabama.
Thermal analysis of aerosol size distributions provided
size resolved volatility up to temperatures of 400C during
extensive flights over North America (NA) for the
INTEX/ICARTT experiment in summer 2004. Biomass
burning and pollution plumes identified from trace gas
measurements were evaluated for their aerosol physiochemical and optical signatures. Measurements of soluble
ionic mass and refractory black carbon (BC) mass,
inferred from light-absorption, were combined with
volatility to identify organic carbon at 400C (VolatileOC)
and the residual or refractory organic carbon,
RefractoryOC. This approach characterized distinct
constituent mass fractions present in biomass burning and
pollution plumes every 5-10 minutes. Biomass burning,
pollution and dust aerosol could be stratified by their
combined spectral scattering and absorption properties.
The \non-plume\ regional aerosol exhibited properties
dominated by pollution characteristics near the surface
and biomass burning aloft.
VolatileOC included most water-soluble organic
carbon. RefractoryOC dominated enhanced shortwave
absorption in plumes from Alaskan and Canadian forest
fires. The mass absorption efficiency of this
RefractoryOC was about 0.63 m2g-1 at 470 nm and 0.09
m2g-1 at 530nm. Concurrent measurements of the
humidity dependence of scattering, gamma, revealed the
OC component to be only weakly hygroscopic resulting in
a general decrease in gamma with increasing OC mass
fractions. Under ambient humidity conditions, the
systematic relations between physio-chemical properties
and gamma lead to a simple dependency on the
absorption per unit dry mass for these plume types that
may be used to challenge remotely sensed and modeled
optical properties.
13A.6
Investigating the Volatility of SOA in Different Urban
Environments. CHRISTOPHER J. HENNIGAN, Amy P.
Sullivan, Richard E. Peltier, Rodney J. Weber, Christos
Fountoukis, Athanasios Nenes, Georgia Institute of Technology;
Delphine Farmer, Paul J. Wooldridge, Ronald C. Cohen,
University of California, Berkeley.
The formation of secondary organic aerosol (SOA)
remains a poorly understood area of aerosol science.
Current theory on SOA formation suggests that volatile
organic carbon compounds (VOC's) in the gas phase
undergo oxidation in the atmosphere to form gaseous
products that are less volatile than the parent compounds
(i.e., the vapor pressures of the product compounds are
lower than those of the parent compounds). The
oxidation products can, if equilibrium conditions exist,
partition to the particle phase through a sorption process.
SOA volatility is important because it may provide insight
into the chemical nature of SOA, as well as information
about the formation mechanisms and precursors. In this
study, using water-soluble organic compounds (WSOC)
in fine particles as a measure of SOA, the volatile nature
of SOA has been investigated in two different urban
environments. VOC emissions in the Mexico City
Metropolitan Area (MCMA) are dominated by local
anthropogenic sources (predominantly mobile sources)
while VOC emissions in Atlanta, GA are a mix of
anthropogenic emissions and a regional biogenic source.
Using a combination of aerosol composition
measurements (both organic and inorganic), the
ISORROPIA aerosol thermodynamic equilibrium model,
and the application of a thermal denuder, we investigate
the volatility of SOA in two urban environments with
predominantly different sources.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
14A.1
Measurements and Interpretation of the Effect of Soluble
Organic Surfactants on the Density, Shape and Water
Uptake of Hygroscopic Particles. ALLA ZELENYUK,
Pacific Northwest National Laboratory; Dan Imre, Imre
Consulting; Luis A. Cuadra-Rodriguez, Barney Ellison,
University of Colorado at Boulder.
A large fraction of atmospheric particles are composed of
hygroscopic salts that are mixed with variety of organic
molecules, of which surfactants represent an important
class. Because of the tendency of surfactant molecules to
coat the particles' surfaces, a monolayer might be
sufficient to drastically alter particle hygroscopic
properties, their CCN activity, and reactivity. Moreover,
because the aliphatic chains are exposed to the oxidizing
atmosphere they are expected to be transformed through
heterogeneous chemistry, yielding complex products with
mixed properties. Given the important role that is played
by the interaction of particles with the ambient relative
humidity it is critical to develop an understanding of the
impact surfactants may exert on particle hygroscopic
properties. We will report the results from a series of
observations on ammonium sulfate, sodium nitrate,
sodium chloride and sea salt particles coated with two
types of soluble surfactant molecules: sodium dodecyl
sulfate and sodium oleate. We have been able to measure
the effective densities and hygroscopic growth factors of
internally mixed particles with a range of surfactant
concentrations that start below a monolayer and extend all
the way to particles composed of pure surfactant. For
many of the measurements the data reveal a rather
complex picture that cannot be simply interpreted in terms
of the known pure-compound densities and growth
factors. We show that the observed particle density
provides evidence that the density of the surfactant
fraction changes with concentration and that once this is
properly taken into account the water uptake data can be
quantitatively understood. For unsaturated hydrocarbons
we observed and quantified the effect of oxidation by
ozone on particle size, effective density, hygroscopic
growth factor and individual particle mass spectral
signatures.
14A.2
Evolution of SOA Mass Spectra from Photo-oxidation of
Diesel Exhaust. AMY M. SAGE, Emily A Weitkamp, Allen L.
Robinson, Neil M. Donahue, Carnegie Mellon University.
Regional chemistry models predicated on laboratory yield
curves significantly underpredict the particle-forming
capacity of aging urban air masses. The high-flux,
volatile organic compounds included in these models
cannot account for the large quantities of organic material
that condense downwind of anthropogenic sources.
Furthermore, the mass spectra of laboratory-generated
SOA from traditional precursors do not agree with those
observed in aged air masses. Atmospheric abundance is
not the sole criterion for identifying SOA precursors. We
have suggested that precursor vapor pressure also plays an
important role, and we have used the large suite of semivolatile compounds emitted by combustion sources to test
this hypothesis.
UV-initiated oxidation chemistry of diesel exhaust carried
out in our environmental chamber results in significant
particle growth after illumination. We calculate that the
mass formed exceeds that expected from known
precursors by a factor of ten. Here we wish to explicitly
consider the chemical nature of the SOA that is formed.
To that end, the composition of suspended particulate
matter was monitored throughout several experiments
using a Q-AMS
Using the known mass spectrum of the primary emissions
recorded prior to illumination, we can confidently subtract
the primary contribution from the total spectrum. The
resulting residual spectrum reveals the chemical
transformations occurring in the condensed phase as SOA
is formed. Our analyses show that the chemical
composition of the SOA that is formed is not constant
over time. As SOA continues to form, signal in the
residual mass spectrum shifts from larger masses (m/z <
70) to smaller, oxygen-containing fragments, suggesting
that later-forming SOA must be more functionalized
before it can condense. After four to five hours of
oxidation, we produce aged organic aerosol whose
spectrum matches well with ambient observations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
14A.3
HR-ToF-AMS Study of the Yield and Chemical Composition
of alpha-Pinene SOA as a Function of Organic Particulate
Loading. JOHN SHILLING, Qi Chen, Stephanie King, Thomas
Rosenoern, Scot Martin, Harvard University; Jesse Kroll,
Douglas Worsnop; Aerodyne Research Inc.; Peter DeCarlo,
Allison C. Aiken, Donna Sueper, Jose L. Jimenez, University of
Colorado and CIRES.
Recent reports in the literature indicate that models
invoking equilibrium partitioning of semivolatile VOC
oxidation products into the organic phase are unable to
reproduce measured SOA concentrations. This
discrepancy between the modeled and measured data
demonstrates that the chemical pathways responsible for
SOA formation are poorly understood. In an effort to
elucidate these pathways, the Harvard Environmental
Chamber has been used to generate secondary organic
aerosol (SOA) from the ozonolysis of alpha-pinene. The
total organic loading was varied by changing the alphapinene concentration from 1 - 100 ppbv while holding all
other reaction conditions constant. An Aerodyne HRToF-AMS was employed to determine the yield and
chemical composition of the alpha-pinene SOA as a
function of particulate loading.
The yield of SOA from this reaction agrees with
previous measurements at the highest particulate loadings
studied. However, at low particulate loading, yields were
higher than previously reported. Revised yield
measurements can help interpret the discrepancy between
measured and modeled atmospheric SOA levels.
HR-ToF-AMS results indicate that the chemical
composition of the aerosol is a strong function of the
organic particulate loading. The spectra show that SOA
produced from alpha-pinene ozonolysis is significantly
more oxygenated under atmospherically relevant organic
loadings than previously reported. Analysis of the highresolution data shows that the fraction of oxygenated
organic compounds (CxHyOz) in the SOA increased at
the expense of hydrocarbon-like compounds (CxHy) as
the total loading decreased. The carbon-to-oxygen ratio of
the organic material was also determined and decreased
with loading. Oligomeric material was observed in the
SOA, even at the lowest organic loadings. Furthermore,
oligomers composed an increasing fraction of the total
organic material as loading decreased. A mechanism for
the polymerization of organic material was developed
based on the mass spectra. These results significantly
advance the understanding of SOA formation and
partitioning.
14A.4
Incorporating GCxGC-TOFMS Information on
Compositional Complexity of Chamber-Derived Aerosol in
Models of Secondary Organic Aerosol (SOA) Formation and
Aging. KELLEY BARSANTI, James Smith, National Center
for Atmospheric Research; James Pankow, Oregon Health &
Science University.
Achieving a quantitative understanding of the formation
and aging of secondary organic aerosol (SOA) remains a
considerable challenge for the accurate prediction of
organic particulate matter (OPM) levels in the
atmosphere. Currently, most large scale SOA models
assume two products (2p) per parent hydrocarbon (HC);
when N parent HCs are present, this is denoted here as the
N-2p approach. Recent studies have shown that the N-2p
approach leads to significant underprediction of
atmospheric OPM levels. At least some portion of this
problem is due to failure of the N-2p approach to
adequately represent the complex mixture of condensable
products that can form from a given parent HC, and the
aging processes affecting those products. SOA models
are needed that consider: a) the time dependence in the
number of products that form from a particular parent
HC; b) the time-dependent properties of the products (e.
g., increasing polarity due to continued oxidation and/or
fragmentation); and c) the formation of essentially nonvolatile polymeric material by accretion reactions.
Chamber experiments have been conducted using
atmospherically relevant levels of parent HCs from live
trees under a range of conditions. OPM was analyzed
using a Pegasus 4D two-dimensional gas chromatograph/
time-of-flight mass spectrometer (GCxGC-TOFMS).
GCxGC-TOFMS is well-suited for determining polarity
and composition (on a functional-group level) of complex
mixtures. Approaches for and implications from porting
composition data from chamber-based GCxGC-TOFMS
analyses to higher-order, time-dependent SOA models
will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
14A.5
Oxygenated Organic Aerosols: Bridging Field and Smog
Chamber Observations Using an Aerodyne Aerosol Mass
Spectrometer. M.RAMI ALFARRA, Andre S.H. Prevot,
Jonathan Duplissy, Axel Metzger, Josef Dommen, Ernest
Weingartner, Urs Baltensperger, Laboratory of Atmospheric
Chemistry, Paul Scherrer Institut; Valentin A. Lanz, Christoph
Hueglin, Empa, Swiss Federal Laboratories for Materials
Testing and Research.
14A.6
TBA
Two types of oxygenated organic aerosols (OOA I and
OOA II) have recently been identified at urban locations
in Europe and North America using the Aerodyne Mass
Spectrometer (Q-AMS) based on multivariate statistical
analysis methods. During a summer study in Zurich,
Switzerland, OOA I was characterised by a relatively high
m/z 44 to total organic ratio and it had a mass spectral
signature similar to that of fulvic acid. It was also found
to have a similar temporal behaviour to the sulphate
component of the aerosols. On the other hand, OOA II
was found to be less aged than OOA I and it was
characterised by a relatively high m/z 43 to total organic
ratio. It also had a similar temporal behaviour to the
nitrate component of the aerosol and it was sensitive to
ambient temperature (i.e. more volatile than OOA I). In
this paper, we investigate the chemical composition of
secondary organic aerosol (SOA) generated in a smog
chamber from the photooxidation of the biogenic
precursor (alpha-pinene) and compare it to the ambient
spectra of OOA I and OOA II. In particular, we present
results showing the effect of the initial precursor
concentration on the mass spectral signature of the SOA
produced and on its chemical and physical properties.
This work represents a direct application attempt of the
AMS to bridge the gap between field measurements and
smog chamber experiments with the aim of achieving an
improved understanding of SOA formation and
transformation in the atmosphere.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
16A.1
Emissions and Secondary Formation of Organic Aerosols in
the Polluted Atmosphere: New Results from the
Northeastern U.S. in 2004 and Texas in 2006. JOOST DE
GOUW, Charles Brock, Ann Middlebrook, NOAA Earth
System Research Laboratory and CIRES, University of
Colorado; Rodney Weber, Georgia Institute of Technology; Tim
Bates, NOAA Pacific Marine Environmental Laboratory.
We analyzed airborne measurements of water-soluble
organic carbon (WSOC) and ship-based measurements of
organic aerosols (OA) in urban plumes in the northeastern
U.S. in 2004. A strong growth in secondary organic
aerosols (SOA) was observed in urban plumes in the first
24 hour after emission that cannot be explained from the
measured volatile organic compound (VOC) precursors,
in agreement with the findings from multiple recent
studies. The reasons for this discrepancy are discussed
and include (i) formation of SOA from lesser-volatile
precursors that are not captured by current VOC
measurements, (ii) higher SOA yields than smog-chamber
studies indicate, and (iii) enhanced formation of biogenic
VOCs in urban plumes. A parameterization is derived that
explains a large part of the observed variability in WSOC
and OA based on measured mixing ratios of carbon
monoxide (CO) and the transport or photochemical age of
the sampled air masses. Results from the Texas Air
Quality Study (TexAQS) in the summer of 2006 provide
an interesting test case for these concepts, because the
VOC-NOx-CO composition of industrial plumes in Texas
is markedly different from that of an urban plume. Some
initial results from organic aerosol measurements during
TexAQS will be presented and discussed in the
framework of the findings from the northeastern U.S.
16A.2
Assessing Secondary Organic Aerosol Using Online Aerosol
Mass Spectrometry. James Allan, Keith Bower, Gerard Capes,
HUGH COE, Jonathan Crosier, Paul Williams, University of
Manchester, UK.
The talk will utilise recent measurements made using the
Aerodyne Aerosol Mass Spectrometer to provide
evidence for the behaviour of secondary aerosol in a
range of environments. Several major field campaigns
have taken place in the last few years around Europe: in
the UK, in the Adriatic region and the Po Valley; and also
in West Africa during AMMA. Oxygenated organic
aerosol is a major component of the submicron mass of
aerosols in all these regions and in the polluted northern
hemisphere its mass has been shown to be greatly
underpredicted by models compared to recent
measurements.
During the African Monsoon Multidimensional Analysis
(AMMA) project, the UK BAe 146 aircraft flew a series
of missions across the West African region in both the dry
season and the summertime monsoon. In the dry season
biomass burning aerosol were ubiquitous throughout the
region, at times mixing with the dust layers observed.
Evidence is presented for the spatial extent of the biomass
burning aerosol and it is shown that there does not appear
to be significant formation of secondary organic mass
with age, unlike the situation in continental polluted
environments. However, there is evidence for a change in
the chemical functionality of the organic biomass burning
aerosols with marker compounds such as levoglucosan
being preferentially removed and more oxygenated multifunctional moieties increasing with time. In the
summertime, the measured organic aerosol loading is
close to the detection limit of the instrument. However,
the concentrations observed appear to be consistent with
model estimates of secondary organic aerosol mass in
these regions. This contrasts markedly with polluted
northern hemispheres where similar models, based on
adsorptive partitioning schemes, greatly underpredict the
SOA. The contrast between northern continental midlatitudes and tropical forests will be highlighted and
discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
16A.3
Measurements of the Composition of 6 - 30 nm Diameter
Biogenic Secondary Organic Aerosols using Thermal
Desorption Chemical Ionization Mass Spectrometry.
JAMES SMITH, Jeff Rathbone, National Center for
Atmospheric Research; Markku Kulmala, University of
Helsinki; Peter McMurry, University of Minnesota.
We report measurements of the molecular composition of
6 - 30 nm diameter atmospheric aerosols performed using
the Thermal Desorption Chemical Ionization Mass
Spectrometer (TDCIMS) at two sites that are dominated
by biogenic emissions. The first of these measurements
was performed in July 2006 during the CELTIC-Niwot
Ridge study, in a subalpine forest located 35 km west of
Boulder, Colorado. The second set of measurements was
performed during the EUCAARI07 campaign in the
Boreal forest in Hyytiala, Finland from 15 April - 30 June
2007. Both sets of measurements indicate a dominant
role played by organic species in the formation of
atmospheric nanoparticles. Positive ion TDCIMS
measurements at both sites show the presence of methyl
and dimethyl amines in particles as small as 8 nm. Other
oxidized organics detected in positive ion TDCIMS
measurements are presumed to be alcohols, aldehydes, or
ketones with molecular weights as high as 400 amu.
Negative ion TDCIMS measurements show the presence
of multifunctional organics with carboxylic acid moieties,
with molecular weights as high as 400 amu. The
evolution of chemical composition of 6 - 30 nm diameter
particles will also be presented in the transition from
Spring to Summer in Hyytiala. Changes in composition
during early particle growth will also be explored.
16A.4
The search for marine organic aerosols. JAMES ALLAN,
Jonathan Crosier, Paul Williams, Keith Bower, Nick Good,
Martin Irwin, Gordon McFiggans, Michael Flynn, David
Topping, Hugh Coe, University of Manchester, UK.
Understanding the composition of marine aerosols is
vitally important, as oceans cover the majority of the
earth's surface and are known to be a large source of
particles, representing a large but relatively poorly
understood factor in radiation budgets through the direct
and indirect radiative effects. There is currently mounting
evidence for the presence of an organic component of the
submicron aerosol, linked with phytoplankton activity, in
addition to the known sea salt and sulphate components.
The effect of this fraction is potentially hugely significant,
as organic matter is known to drastically perturb the cloud
activation behaviour of particles, through the reduction of
solubility and surface tension, although this is highly
dependent on the precise concentrations and mixing states
of the components and the size of the particles.
Developments in the field of in situ aerosol
instrumentation are currently helping to address these
issues through the online measurement of composition,
size and hygroscopic behaviour in both sub- and supersaturated conditions. Aerosol measurements using
instruments such as Aerosol Mass Spectrometers, CCN
counters and Hygroscopicity Tandem DMAs from four
recent field campaigns in various remote Atlantic
locations (on both ground- and ship-based platforms) will
be presented, and ongoing work to model the potential
implications this fraction will have on cloud formation
will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Advances in Instrumentation for
Organic Aerosols
2007 AAAR Annual Conference Abstracts
16A.5
Exploring the Magnitude and Formation Mechanism of
Above-Cloud Organic Layers. SHANE MURPHY, Armin
Sorooshian, Harmony Gates, Richard C. Flagan, John H.
Seinfeld, California Institute of Technology; Graham Feingold,
National Oceanic and Atmospheric Administration; Haflidi
Jonsson, Naval Postgraduate School.
Recent field measurements indicate that there is more
organic aerosol in the free troposphere than previously
thought. Because of the potential impacts on global
climate, both through direct and indirect radiative effects,
it is important to determine the source and prevalence of
this organic aerosol. Cloud processing is one potential
pathway that could lead to the formation of organic
aerosol in the troposphere above the boundary layer.
However, few in situ, chemically-resolved measurements
of cloud-processed aerosol are available.
A compact time of flight Aerodyne aerosol mass
spectrometer (cToF-AMS) was deployed for the first time
on an airborne platform (CIRPAS Twin Otter) off the
coast of California near Monterey in July of 2005 and
then again in Houston, TX during August, 2006. The
high sensitivity of the cToF-AMS enables us to obtain
size resolved aerosol mass spectra on a 15 second
timescale and non-size resolved spectra on a 2 second
timescale. These fast chemical data, in conjunction with
the use of a couterflow virtual impactor (CVI), allow us to
track the chemical transformation of aerosol particles
within a cloud.
We have observed organic aerosol layers above marine
stratocumulus clouds off the coast of California and above
continental cumulus near Houston. Measurements of the
size distribution and chemical composition of below
cloud aerosol together with in-cloud measurements of
droplet residual chemistry indicate that most of the abovecloud organic aerosol is distinctly different from aerosol
that has been transported through the clouds. By
monitoring the chemical evolution of aerosol traveling
through the clouds, we conclude that there must be
another mechanism other than cloud processing acting to
form the observed organic layers.
16A.6
A Study on the Sources and Chemical Processes of Organic
Aerosol at the Whistler Summit with a High-Resolution
Time-of-Flight Aerosol Mass Spectrometer. QI ZHANG,
Yele Sun, State University of New York, University at Albany,
NY; Richard Leaitch, Anne Marie Macdonald, Kathy Hayden,
Shao-Meng Li, John Liggio, Peter Liu, Environment Canada;
Aaron van Donkelaar, Randall Martin, Dalhousie University;
Douglas Worsnop, Aerodyne Research, Inc.; Michael Cubison,
University of Colorado-Boulder, Colorado,
A new Time-of-Flight Aerosol Mass Spectrometer with a
high mass resolution of ~ 5000 (HR-ToF-AMS; DeCarlo
et al., Anal. Chem., 2006) was deployed at the summit of
the Whistler Mountain, British Columbia, from April 20
to May 17, 2006. With this instrument, we determined
the concentration, composition, and chemically speciated
size distributions of submicron particles (approximately
PM1) every 5 minutes. We also obtained the highly m/zresolved mass spectra, based on which the elemental
composition of most small ion fragments (m/z <100 amu)
was quantitatively determined. This improved chemical
characterization of organic aerosol mass spectra
significantly enhances our ability to address the sources
and processes of organic aerosol. Organic aerosol (OA)
prevailed at the Whistler summit (elev. ~ 2200 m),
accounting for ~ 50% of the total PM1 mass. The OA
appeared to be highly aged and was composed almost
completely of oxygenated organic aerosol (OOA).
Sulfate aerosol was usually less abundant than the OA but
episodes of strongly enhanced sulfate aerosol were
observed. A major sulfate episode occurred on May 15,
2006, during which ammonium sulfate contributed > 90%
of the total PM1 mass. This episode lasted for ~ 0.5 day
and was followed by a high organic aerosol episode. The
mass spectra reveal that the organic aerosol is more
highly oxidized during the sulfate period. Also, the high
resolution mass spectra indicate significant structure
difference in organic species between these two episodes.
Simulations from a global model of oxidant-aerosol
chemistry (GEOS-Chem), backtrajectory analysis,
evolution of the size distribution of aerosol species, and
correlation of the Whistler aerosol spectrum with previous
AMS measurements in rural British Columbia, and in
Asia provides supporting evidence of Asian influence
during the high sulfate period, in contrast with regional
influence during the high organic period.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2B.1
Cloud Droplet Activation Properties of Surface Active
Straight-Chain Fatty Acids. NOENNE PRISLE, Birgitta
Svenningsson, Merete Bilde: University of Copenhagen; Riikka
Sorjamaa, Ari Laaksonen: University of Kuopio.
Water soluble organics with surfactant properties have
been identified in atmospheric aerosol and precipitation
samples. Surface active compounds affect the equilibrium
water vapor pressure over an aqueous solution droplet and
thereby the cloud droplet activation properties of
atmospheric particles comprising these surfactants. A
characteristic feature of surfactants in aqueous solutions is
the partitioning of the surfactant molecules to the solution
surface. Partitioning becomes increasingly pronounced
with the larger surface area to bulk volume ratios for still
smaller solution droplets. Sorjamaa et al. (2004) have
shown for the surfactant model compound SDS that the
effect of surfactant properties on cloud droplet activation
is best described when this partitioning is accounted for in
both the Kelvin (curvature) and Raoult (solute) terms of
the Koehler equation.
Recent model studies by Sorjamaa et al. (2006) have
varied the different surfactant parameters over intervals
representative of several compound classes found in
atmospheric aerosols. It is shown that for the shorter
straight chain fatty acids among others, surfactant
properties in general and the effect of surfactant
partitioning in particular can be expected to significantly
affect the predicted cloud droplet activation properties of
atmospheric particles.
We have experimentally investigated the cloud droplet
activation of laboratory produced particles consisting of
the sodium salts of n-octanoic (caprylic), n-decanoic
(capric), n-dodecanoic (lauric) and n-tetradecanoic
(myristic) acid, respectively, using a Wyoming CCNC
-100B cloud condensation nucleus counter. The results
are compared to theoretical predictions derived using the
model of Sorjamaa et al. (2004) which accounts for
surfactant partitioning in the activating droplet through
both the Kelvin and the Raoult effects. Our findings will
be discussed within the context of atmospheric science.
2B.2
Effect of adipic acid (a slightly soluble organic substance)
coatings on the CCN activation of soluble and insoluble
particles. SILKE S. HINGS, Eben S. Cross, Paul Davidovits,
Boston College; Timothy B. Onasch, Douglas R. Worsnop,
Aerodyne Research, Inc.
In this study, the effect on the CCN activity of coating
soluble and insoluble particles with a slightly soluble
organic substance (adipic acid) is explored. The soluble
particles are composed of ammonium sulfate; the
insoluble particles are size selected freshly generated soot
particles. The experiments were conducted as a function
of adipic acid coating thickness (5-35 nm and 5-100 nm,
respectively) deposited via gas-to-particle condensation
onto size-selected solid ammonium sulfate and soot
particles. Pure ammonium sulfate particles were observed
to activate at the predicted water supersaturations (Kohler
equation). The uncoated soot particles - as expected were inactive over the supersaturation range investigated
(0.2-2.5 %). To evaluate the effect of adipic acid coatings
we first performed CCN activation experiments with pure
adipic acid particles. When generated under laboratory
conditions, such particles have been shown to exist as
either solid particles or supersaturated aqueous solution
droplets. Solid particles were generated via nucleation
from saturated vapor, supersaturated aqueous solution
droplets were generated via atomization. For a given
particle diameter the CCN activation exhibited by these
two adipic acid particle phases occurs at different water
supersaturations, i.e. the supersaturated aqueous solution
droplets activate at significantly lower supersaturations
than the solid particles.
In the present experiments, it is observed that with
increasing adipic acid coating thickness, the CCN activity
of the ammonium sulfate particles decreases, approaching
the Kohler curve for the supersaturated aqueous adipic
acid solution droplets. By contrast, the CCN activity of
the soot particles increases with increasing coating
thickness and approaches the activation curve for the solid
adipic acid particles. Possible ways of understanding this
behavior will be discussed.
_________________
Sorjamaa et al., Journal of Aerosol Science, 37 (2006),
1730-1736.
Sorjamaa et al., Atmospheric Chemistry and Physics, 4
(2004), 2107-2117.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2B.3
CCN Closure in the Polluted Boundary Layer over Houston,
TX During the Gulf of Mexico Atmospheric Composition
and Climate Study (GoMACCS). SARA LANCE, Athanasios
Nenes, Georgia Institute of Technology; Harmony Gates,
Varuntida Varutbangkul, Tracey Rissman, Shane Murphy,
Armin Sorooshian, Fred Brechtel, Richard Flagan, John
Seinfeld, California Institute of Technology; Graham Feingold,
National Oceanic and Atmospheric Administration; Haflid
Jonssoni, Roy Woods, Navy Postgraduate School.
2B.4
Modeling Cloud Condensation Nuclei Activation at Urban
and Background Locations: The Influence of Composition
and Mixing State. Ingrid Ulbrich, Ken Docherty, Jose Jimenez,
MIKE CUBISON, University of Colorado; Barbara Ervens,
Betsy Andrews, Graham Feingold, John Ogren, NOAA Earth
System Research Laboratory; Kerry Denkenberger, Kim Prather,
University of California- San Diego; David Snyder, James
Schauer, University of Wisconsin; Thanos Nenes, Georgia
Institute of Technology.
We performed Cloud Condensation Nuclei (CCN) closure
for boundary layer samples over Houston, Tx during the
month long GOMACCS campaign during AugustSeptember, 2006. Measurements were obtained aboard
the CIRPAS Twin Otter, and polluted air masses in and
out of cloudy regions were sampled over a total of 22
flights. Two flights are presented, the first within the
Houston regional plume, and the second over the Houston
ship channel, one of the United State's busiest sea ports
and a heavily industrialized area. During both flights we
sampled highly polluted air, with peak particle
concentrations exceeding 25,000 cm-3. To perform
closure calculations, we used Particle Into Liquid Sampler
(PILS) and Aerosol Mass Spectrometer (AMS)
measurements to constrain the particle chemical
composition and size distribution measurements obtained
with a scanning Differential Mobility Analyzer (DMA).
Data were obtained for both below cloud aerosol,
interstitial aerosol and cloud droplet residuals sampled
from a Counterflow Virtual Impactor (CVI).
The indirect influence of aerosol particles on the radiative
balance of the atmosphere through changes in droplet
number and persistence of clouds, known as the "aerosol
indirect effects", carry the largest uncertainty amongst the
presently known causes change in the radiative forcing of
climate [IPCC, 2007]. Closure studies, where it is
attempted to model the measured activation of aerosol to
cloud condensation nuclei (CCN) using measured
physical and chemical aerosol parameters (concentration,
size distribution, composition, mixing state), are key for
constraining the magnitude of the indirect effects.
In background locations, assuming the aerosol population
is internally-mixed is sufficient to reach agreement
between simple CCN models and measurements, and thus
the size distribution is the most important parameter in
determining CCN activation [e.g. Dusek, et al., 2006;
Ervens, et al., 2007]. However, in areas under the
influence of urban emissions, chemical composition
information and knowledge of the state of mixing of the
aerosol population is required to better predict CCN
activation [Broekhuizen, et al., 2006; Medina, et al., 2007;
Stroud, et al., 2007]. We present results from urban areas
and background locations highlighting where the mixing
state and composition are important factors in
determining CCN activation, develop suitable proxies for
incorporating this information in cloud models and assess
the global extent to which this applies.
Broekhuizen, K., et al. (2006). Atmospheric Chemistry
and Physics, 6, 2513-2524.
Dusek, U., et al. (2006). Environmental Science &
Technology, 40, 1223-1230.
Ervens, B., et al. (2007). Journal of Geophysical Research
(Atmospheres), 112, D10S32.
IPCC (2007). Climate change 2007: Scientific basis.
Fourth assessment of the Inter-govermental Panel on
Climate Change.
Medina, J., et al. (2007). Journal of Geophysical Research
(Atmospheres), in press,
Stroud, C. A., et al. (2007). Journal of the Atmospheric
Sciences, 64, 441-459.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2B.5
Chemical Speciation of Sulfur in Marine Cloud Droplets and
Particles: Quantitative Assessment of Methanesulfonate and
non-Sea Salt Sulfate Partitioning in Individual Sea Salt
Particles. R.J. Hopkins, Lawrence Berkeley National
Laboratory; Y. Desyaterik, R.A. Zaveri, C.M. Berkowitz,
Pacific Northwest National Laboratory; A.V. Tivanski, M.K.
Gilles, Lawrence Berkeley National Laboratory; A. Laskin,
Pacific Northwest National Laboratory.
Sea salt particles are chemically and physically modified
by condensation of various oxidation products of
dimethylsulfide (DMS), the major source of sulfur over oceans.
For many years, effective conversion of DMS to particulate
sulfate was assumed to be the dominant reaction pathway for
DMS in the marine boundary layer. However, recent modeling
studies (von Glasow and Crutzen, Atm.Chem.Phys. 2004)
indicated that under certain conditions DMS does not
2predominantly convert to sulfate (nss-SO4 ), but rather ends up
in sea salt particles in a form of methanesulfonate (CH3SO3 ),
which previously has been considered only of minor importance.
2Specifically, high CH3SO3-/nss-SO4 values have been
reported in model runs for cloudy MBL at winter conditions
(surface temperature of 3-8 C over the ocean). These findings
have been confirmed in our field study presented here. We
report that CH3SO3Na is a dominant form of the nss-sulfur
found in the sea salt particles that traveled over the areas of cold
ocean stream prior their sampling in the vicinity of Pt. Reyes,
CA, north of San Francisco. In this presentation we demonstrate
the effective and complementary coupling of a multitude of
microprobe analytical techniques (CCSEM/EDX, TOF-SIMS
and STXM/NEXAFS) to unravel the chemical composition of
individual sea salt particles with the focus on the quantitative
2assessment of the CH3SO3-/nss-SO4 partitioning of the nsssulfur in these particles. We report the particle size specific data
on the nss-S/Na and the CH3SO3-/nss-SO4^(2-) ratios measured
in dry residues of marine cloud droplets and particles collected
during the Marine Stratus Experiment (MASE) in July 2005.
Characteristic ratios of nss-S/Na < 0.10 are reported for sea salt
particles, with higher values for small particles indicating
extensive formation of sulfur containing salts in small particles.
2Characteristic ratios of CH3SO3-/nss-SO4 < 0.70 are reported
with higher values for large particles, indicating the higher
2capacity for CH3SO3- (lower conversion to SO4) for large
particles. To the best of our knowledge, this is the first time that
2CH3SO3-/nss-SO4 have been quantitatively reported based on
the individual particle measurements These observations were
rationalized with our modeling calculations showing enhanced
formation of CH3SO3- in sea salt particles that is consistently
predicted for the specifics of geophysical environment and
meteorology of the reported experiment.
2B.6
An Algorithm to Derive Size Dependent Hygroscopic
Growth Factors from Size Distribution Data. ANDREY
KHLYSTOV, Duke University.
Understanding the aerosol hygroscopic properties is of
critical importance for studies of the aerosol effect on
climate. Hygroscopic growth of aerosol particles is
usually studied using Tandem-DMA approach in which a
nearly monodisperse aerosol of a certain size is subjected
to different relative humidities (RH) and the changes in
particle size are recorded. While this approach provides
an accurate measure of aerosol growth factors, it can
study only one size at a time. An alternative approach was
recently developed for determining integrated volumebased growth factors as a function of relative humidity
which provides a way for verification of thermodynamic
models using chemical composition data as an input. In
this approach aerosol size distributions are measured at a
reference low RH and an elevated RH. The difference in
integrated volume concentrations at two RHs provides a
measure of aerosol water content.
Here an algorithm is presented that allows to derive
hygroscopic growth factors as a function of particle size
using size distribution measurements at two RHs, without
any additional information on the aerosol composition. In
the case of an internally mixed aerosol the algorithm
provides a unique solution to the size dependent growth
factors in the size range of the measured size
distributions. For externally mixed aerosol consisting of
two fractions the algorithm allows to estimate the growth
factors of each fraction and their relative abundance. For
externally mixed aerosol consisting of more than two
fractions constraints on the shape of the fractions are
required to derive their relative abundance in the aerosol.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2B.7
Optical Particle Counter Measurements of Marine Aerosol
Hygrosopic Growth. JEFFERSON R. SNIDER, University of
Wyoming; Markus Petters, Colorado State University.
2B.8
Broadening of cloud droplet size spectrum observed during
Marine Stratus/Stratocumulus Experiment (MASE). JIAN
WANG, Peter Daum, Yangang Liu, Gunnar Senum
A technique is developed for the determination of the
hygroscopic growth factor of dry particles with diameter
between 0.3 and 0.6 micro-meter and is applied to
measurements made during the second Dynamics and
Chemistry of Marine Stratocumulus (DYCOMS-II). Two
optical particle counters (OPC) are utilized; one measures
the aerosol size spectrum at ambient relative humidity and
the other dries the particles prior to light scattering
detection. Growth factors are based on measurements
made in the region of the Mie scattering curve where
scattered light intensity increases monotonically with
particle diameter, i.e. D<0.9 micro-meter. Growth factors
at approximately 90% ambient relative humidity in a
marine airmass sampled over the eastern Pacific Ocean
range between 1.4 and 1.7 and suggest that upwards of
30% of the particle mass is non-hygroscopic.
Clouds play a dominant role in determining the Earth
radiation budget. The radiative effect of clouds is
strongly influenced by cloud microphysics. One of the
unresolved problems in modeling cloud microphysics is
the discrepancy between the modeled and observed
dispersion of cloud droplet spectrum. Condensational
growth theory suggests, when exposed to the same
supersaturation, smaller droplets grow faster than the
larger droplets. As a result, droplets in clouds tend to
approach asymptotically to the same size. However, the
observed droplet size spectra are often much broader.
This problem is critical as it is related to the \growth gap\
separating the processes of droplet growth by
condensation and growth by collision-coalescence. The
mechanism responsible for the spectrum broadening can
potentially produce droplets large enough to initiate the
drizzle/rain formation.
The dispersion of the cloud droplet size spectrum is
examined for the Marine Stratus/Stratocumulus
Experiment (MASE). During MASE, aerosol
microphysics, cloud microphysics, and meteorological
parameters were measured onboard the Department of
Energy Gulfstream-1 aircraft over the Eastern Pacific
Ocean in July 2005. Flights were conducted over a
coastal site located at Pt Reyes National Seashore just
north of San Francisco, and extended west over the
Pacific Ocean to as much as 200 km offshore, and as far
south as Monterey Bay. During these flights,
measurements were conducted at multiple levels inside
the clouds, which extended from near the ocean surface to
altitudes between 300 and 450 m. In nearly all cases,
level-averaged relative dispersion decreases with
increasing altitude and liquid water content, as expected
for condensation growth. However, in each level, high
relative dispersions (spectrum broadening) were often
associated with low droplet number concentrations and
large volume-mean droplet diameters. The relationships
between CCN concentration, turbulence, and cloud
microphysics are investigated. The results suggest the
fluctuation in supersaturation induced by turbulence likely
plays an important role in the spectrum broadening
observed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2B.9
CCN, Cloud Droplet Concentrations, and Precipitation in
Clean Air. SUBHASHREE MISHRA, James G. Hudson,
Desert Research Insitute.
Concentration variations of Giant Nuclei (GN) have been
a leading hypotheses for warm rain initiation. ColonRobles et al. (2006) [C-R6] found GN well correlated
with low level horizontal wind speed (v) during the Rain
in Cumulus over the Ocean (RICO) experiment in the
eastern Caribbean in December-January, 2004-05.
However, they found the concentrations of large cloud
droplets (Nld) anti-correlated with v and GN. They also
found a high correlation between v and vertical wind (w)
and between w and total cloud droplet concentrations
(Nc), which is known to be controlled by both w and
cloud condensation nuclei (CCN) concentrations. Since
C-R6 found no correlation of either Condensation Nuclei
(CN) or accumulation mode particles with v they
concluded that Nc was predominantly controlled by w
(negatively), which also tended to negatively control Nld.
The positive correlation of CCN with Nc that we found
was not only contrary to the C-R6 aerosol measurements
but was also similar to their w-Nc correlation (correlation
coefficient. 0.70 versus 0.66). Moreover, the correlation.
coefficient of the product of w and CCN with Nc was
0.82. C-R6 found a negative relationship between w and
Nc and w and Nld for 10 of the 12 flights. We found a
similar positive relationship between CCN and Nld for
these same 10 flights. The two flights that did not fit their
pattern-6th and 12th ranked w but low Nld-had the 2nd
and 3rd highest CCN concentrations and thus the 5th and
4th highest Nc. The present study is similar to earlier
studies in more polluted clouds that also showed more
influence of CCN than GN on both Nc and Nld. These
results indicate that the variations in CCN concentrations
within these clean air masses tended to modulate
precipitation.
Colon-Robles et al. (2006), Geophys. Res. Lett. L20814,
doi:10.1029/2006GL027487
2C.1
Aerosol-Cloud Interactions: Sensitivity of Indirect Effects to
Cloud Formation Parameterization, Meteorological Fields,
and Emission Scenario SOTIROPOULOU RAFAELLAELENI, Nicholoas Meskhidze, Athanasios Nenes, Georgia
Institute of Technology.
The aerosol indirect effect (AIE) is one of the largest
sources of uncertainty in assessments of anthropogenic
climate change. The objective of this study is to assess the
uncertainties in indirect forcing and autoconversion of
cloud water to rain from differences in meteorological
fields, emission scenarios, and parameterizations of cloud
droplet formation. The uncertainty in AIE and
autoconversion is assessed with the NASA Global
Modeling Initiative (GMI), which allows easy interchange
of meteorological fields, aerosol microphysical and
aerosol-cloud interaction packages and therefore is an
ideal testbed for assessing the effects of different
parameters on AIE. “Present day” and “preindustrial”
simulations were carried out using the University of
Michigan and AEROCOM emission inventories.
Meteorological fields are provided by two global climate
models (the NASA GEOS4 finite volume and the
Goddard Institute for Space Studies version II’) and the
NASA Data Assimilation Office. Cloud droplet number
concentration (CDNC) is computed with the empirical
correlations of Boucher and Lohmann [1995] and Segal
and Khain [2006], and the mechanistic parameterizations
of Abdul-Razzak and Ghan [2000] and Fountoukis and
Nenes [2005]. Computed CDNC is used to calculate the
cloud optical depth, the autoconversion rate and the mean
top-of-the-atmosphere (TOA) short-wave radiative
forcing using a modified version of the FAST-J algorithm
[Meshkhidze et al., in preperation]. Autoconversion of
cloud water to precipitation is parameterized following
the formulations of Khairoutdinov and Kogan [2000] and
Rotstayn [1997]. Derived cloud properties, such as cloud
optical thickness and effective radius are compared with
satellite products from MODIS platform. Our results
suggest that differences in meteorological fields, cloud
droplet activation parameterizations and emission
scenarios could account for more than 30% variability in
forcing estimates for the first indirect effect and up to
50% in autoconversion rates. AIE is mostly sensitive to
CDNC parameterization; meteorological is of lesser
importance.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2C.2
Parameterization of Cloud Drop Microphysical Properties
and Evolution in Large-Scale Models. Athanasios Nenes,
WEI-CHUN HSIEH, Georgia Institute of Technology
A parameterization framework that links cloud activation
and microphysical evolution is developed for usage in
aerosol-cloud interaction studies with large scale models.
The framework computes vertical evolution of droplet
population beyond activation; this is then used to directly
compute height-dependant properties, such as
autoconversion rates. The framework also computes the
ratio between effective radius and volumetric radius,
which is used for computing cloud optical depth and
indirect forcing. In this work, we present results using an
adiabatic framework.
The framework is evaluated by comparison with a
detailed numerical parcel model and in-situ data. Good
agreement of relative dispersion between
parameterization and a detailed numerical parcel model
indicates the framework capture the physics of droplet
formation and growth. Evaluation with in-situ
measurements were done for clouds sampled aboard the
CIRPAS Twin Otter during the CRYSTAL-FACE and
CSTRIPE campaigns. On average, the predicted relative
dispersion of parcel model is a factor of 2 lower than
measurement for clouds sampled during CRYSTALFACE and CSTRIPE missions. This underestimation of
the predicted relative dispersion cause relatively larger
uncertainty in predicting autoconversion rates: -41.1 %
(CRYSTAL-FACE) and -58.4 % (CSTRIPE). The
autoconversion uncertainty associated with the predicted
cloud drop number is +3.4 % (CRYSTAL_FACE) and
+5.6% (CSTRIPE). This underestimation of the relative
dispersion is largely due to the assumption of adiabaticity
which can be relaxed using an entraining parcel
framework.
2C.3
Parameterization of Cloud Droplet Formation in Large
Scale Models: Including Effects of Entrainment. DONIFAN
BARAHONA, Athanasios Nenes, Georgia Institute of
Technology.
We present a prognostic parameterization of cloud droplet
formation and growth within the framework of an
entraining ascending parcel. Mixing of outside air is
parameterized in terms of a per-length entrainment rate.
The integration of the droplet growth is done using the
\population splitting\ concept of Nenes and Seinfeld.
Formulations for lognormal and sectional aerosol
representations are given. The concept of \critical
entrainment\, a value beyond which droplet activation is
not favored is introduced, and shown that it is important
for defining i) whether or not entrainment effects have an
impact in droplet formation, and, ii) the characteristic
temperature and pressure for cloud droplet formation. The
performance of the parameterization was evaluated
against a detailed numerical parcel model over a
comprehensive range of droplet formation conditions. The
agreement is very good, with a mean relative error 2.2%
+/- 18%; errors tend to increase (never above 40%) as
entrainment approaches the critical value. This work
offers for the first time a parameterization suitable for
large scale 3-D models which is robust, computationally
efficient, and from first principles links chemical effects
and entrainment to cloud droplet formation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2C.4
Modeling Studies of Aerosol-Cold Cloud Interactions.
TRUDE EIDHAMMER, Paul J. DeMott, Sonia M. Kreidenweis,
Colorado State University.
A Lagrangian parcel model, previously used in aerosolwarm cloud studies, is further developed to include a
simplified description of aerosol hygroscopicity and ice
formation processes in cold clouds. The parcel model is a
tool for investigating aerosol particle influences on ice
initiation, for example via homogeneous freezing
nucleation versus (or in combination with) different
heterogeneous ice nucleation mechanisms. In addition, the
competition for water vapor through the different crystal
formation and growth processes is studied. In the new
model version, the treatment of thermodynamics of
solutions is changed and is expressed in terms of a single
parameter (kappa) [Petters and Kreidenweis, 2007]. Using
several assumed size distributions combined with a range
of kappa-values allows for studying the sensitivity of
cloud properties to aerosol hygroscopicity.
Hygroscopicity is linked to influences on homogeneous
freezing nucleation rates as determined from
parameterizations using solution water activity [Koop et
al., 2000]. For equivalent air parcel initialization
conditions, the conditions for initiation and concentrations
of ice crystals formed by homogeneous freezing in the
revised model compare well with values from several
independent aerosol parcel models used in the Cirrus
Parcel Model Comparisons Project [Lin et al., 2002].
Alternate parameterizations of heterogeneous ice
nucleation have been implemented. These
parameterizations, some published and some newly
derived, originate from theory, laboratory measurements
or field study measurements. Predictions from these
heterogeneous ice nucleation routines are compared and
contrasted in simulations of cold clouds.
Koop, T., et al., Water activity as the determinant for
homogeneous ice nucleation in aqueous solutions, Nature,
406, 611-614, 2000.
2C.5
The Aerosol Modeling Testbed: A New Approach in
Evaluating Treatments of Aerosol Processes for Regional
and Global Climate Models. JEROME FAST, William
Gustafson Jr., Elaine Chapman, Douglas Baxter, Pacific
Northwest National Laboratory.
The direct (scattering and absorption of radiation) and
indirect (cloud-aerosol interaction) effects of aerosols
predicted by global climate models still contain large
uncertainties. The objective of our new project is to
develop an Aerosol Modeling Testbed that streamlines the
process of testing and evaluating refined aerosol process
modules, including those that treat the feedbacks of
aerosols and meteorology, over a wide range of spatial
and temporal scales. The Aerosol Modeling Testbed will
consist of a modular and user-friendly version of WRFchem (a fully-coupled meteorology-chemistry-aerosol
model), and a suite of tools to evaluate the performance of
aerosol process modules via comparison with a wide
range of field measurements. The primary tasks
associated with the Aerosol Modeling Testbed include: 1)
improving the modularity and \user-friendliness\ of the
aerosol process modules within WRF-chem to facilitate
evaluations; 2) developing a series of test simulations,
archived field data, and analysis tools suitable for
systematically evaluating aerosol process modules; 3)
utilizing WRF-chem and analysis tools on various
computer platforms to ensure platform portability; and 4)
implementing, testing and evaluating new aerosol
treatments. A modular model will enable various
treatments of specific aerosol processes to be
systematically compared, while all other atmospheric
processes (e.g. other aerosol processes, emissions, gas
chemistry, meteorology) remain the same. Examples of
specific aerosol processes that could be evaluated include
aerosol-cloud interactions and secondary organic aerosols.
We also discuss one of the most important components of
the Aerosol Modeling Testbed: how it could be used to
foster collaborations and coordination of effort among the
aerosol scientific community.
Lin, R.F. et al.,Cirrus Parcel Model Comparison Project.
Phase 1: The critical components to simulate cirrus
initiation explicitly, J. Atmos. Sci., 59, 2305-2329, 2002.
Petters, M.D., S.M. Kreidenweis, A single parameter
representation of hygroscopic growth and cloud
condensation nucleus activity, Atmos. Chem. and Phys.,
Accepted, 2007.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2C.6
ASDC: A Source of Remotely Sensed Data for Studying
Aerosols, Clouds, and Climate. KATHLEEN MORRIS,
Science Systems and Applications, Inc.; Michelle Ferebee,
NASA Langley Research Center.
The Atmospheric Science Data Center (ASDC) at NASA
Langley Research Center archives aerosol and cloud data
from the Cloud and the Earth's Radiant Energy System
(CERES), the Multi-angle Imaging SpectroRadiometer
(MISR), and the Cloud-Aerosol Lidar and Infrared
Pathfinder Satellite Observations (CALIPSO) projects.
These data span approximately eight years and are used to
study aerosols, clouds, and climate.
The CERES instrument measures broadband radiative
fluxes along with cloud and aerosol properties. The first
CERES instrument (PFM) was launched on November
27, 1997, as part of the Tropical Rainfall Measuring
Mission (TRMM). Two CERES instruments (FM1 and
FM2) were launched into polar orbit on board the EOS
flagship Terra on December 18, 1999, and two additional
CERES instruments (FM3 and FM4) were launched on
board EOS Aqua on May 4, 2002. CERES data are
available for January 1998 through the present.
The MISR instrument obtains precisely calibrated images
at nine different viewing angles and four wavelengths
(red, blue, green, near-infrared) to provide radiance,
aerosol, cloud and land surface data. The MISR
instrument also is onboard EOS Terra. MISR data are
available from February 2000 through the present.
CALIPSO data are used to study the vertical structure of
clouds and aerosols. CALIPSO comprises three
instruments, the Cloud-Aerosol LIdar with Orthogonal
Polarization (CALIOP), the Imaging Infrared Radiometer
(IIR), and the Wide Field Camera (WFC). CALIPSO was
launched into a sun-synchronous orbit on April 28, 2006,
where it joined the A-Train constellation of four other
Earth-orbiting satellites: Aqua, Aura, CloudSat and
PARASOL. CALIPSO data are available for May 2006
through the present.
CERES, MISR, and CALIPSO data along with
documentation, read software, and tools for working with
the data may be obtained from the NASA Langley ASDC
at http:// eosweb.larc.nasa.gov.
2C.7
Sensitivity of Simulated MODIS Reflectances to Dust
Optical Properties. KELLEY WELLS, Graeme Stephens,
Sonia Kreidenweis, Colorado State University.
Solutions for inversion schemes used in the retrieval of
aerosol physical properties (amount, size, and shape) from
remotely-sensed reflectances can be strongly driven by
the a priori assumptions made about the aerosol optical
properties (single-scattering albedo and phase function).
Assumptions made about dust aerosol are often especially
inappropriate; dust optical properties are not wellquantified since the particles are irregularly-shaped and
composed of minerals with different indices of refraction.
This leads to uncertainties in aerosol retrievals over dusty
regions, and corresponding uncertainties in dust direct
forcing estimates.
A forward radiative transfer model was developed to
simulate reflectances above an aerosol layer, such as
those that might be detected by the Moderate Resolution
Imaging Spectroradiometer (MODIS). The model
includes a doubling-adding scheme to describe the bulk
scattering (including multiple scattering processes) and
absorption properties of the aerosol, rather than a singlescattering approximation, which is used in the MODIS
aerosol retrieval algorithm. Various physical and optical
parameters are modified in the model to determine the
sensitivity of simulated reflectances to different aerosol
parameters, and results using MODIS aerosol models are
compared to MODIS over-ocean reflectances for a July
2006 North African dust event.
An improved characterization of the uncertainties
attached to aerosol optical depth (AOD) retrieved over
dusty regions will be important not only because it will
provide improved information for aerosol climatology and
direct forcing calculations, but also for the added
information content that will be made available for
retrievals from other satellite-borne instruments. For
instance, AOD from the MODIS instrument aboard the
Aqua satellite can be used to constrain choices of
backscatter-to-extinction ratio used to retrieve aerosol
extinction profiles from lidar measurements from the
Cloud-Aerosol Lidar and Infrared Pathfinder Satellite
Observation (CALIPSO) instrument. A case study
retrieval for the July 2006 event using this combined
information is also shown.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
2C.8
Measurement Of The Optical Properties Of On-Road LightDuty And Heavy-Duty Vehicle Particulate Emissions. AW.
STRAWA, NASA-Ames Research Center; AG. Hallar, Desert
Research Institute; TW. Kirchstetter, MM. Lunden, Lawrence
Berkeley National Laboratory; GA. Ban-Weiss, RA. Harley, JP.
McLaughlin, University of California, Berkeley; AJ. Kean,
California Polytechnic State University; ED. Stevenson, GR.
Kendall, Bay Area Air Quality Management District.
2C.9
Relative Humidity and Wavelength Dependence of Aerosol
Extinction as Measured by Cavity Ring Down Spectrometry
during TeXAQS-GoMACCS 2006: Selection of Case Studies.
PAOLA MASSOLI, Daniel Lack, CIRES Univ. of Colorado and
NOAA ESRL/CSD; Tahllee Baynard, CIRES Univ. of
Colorado and NOAA ESRL/CSD (now at Lockheed Martin
Inc.); Edward Lovejoy, A.R.Ravishankara, NOAA ESRL/CSD;
Patricia Quinn, Tim Bates, NOAA PMEL.
This paper discusses the measurement of climate relevant
physical and optical properties of aerosols emitted from
motor vehicles during summer 2006 at the Caldecott
Tunnel in the San Francisco Bay Area. Measurements
were made in two separate traffic bores: one carrying only
light-duty (LD) vehicles and the other carrying a mix of
LD vehicles and heavy-duty (HD) diesel trucks. A unique
instrument that uses cavity ring-down (CRD) techniques
and a reciprocal nephelometer to simultaneously measure
the aerosol extinction and scattering coefficients,
respectively, facilitated calculation of the aerosol
absorption coefficient and single scattering albedo. These
quantities are important in determining the radiative
forcing of aerosols on climate. In addition, real-time
measurements of ultrafine particle number, black carbon
(BC), CO, CO2, and NOx concentrations and timeintegrated measurements of PM2.5, EC, and OC mass
concentrations were made.
The optical properties that determine the direct effect of
aerosols (i.e., optical depth, single scattering albedo and
asymmetry parameter) vary with the wavelength of
incoming radiation (i.e., Angstrom exponent) and relative
humidity (i.e., fRH). The size distribution and refractive
index can vary significantly if the aerosol is hygroscopic
and is exposed to varying relative humidity. Accurate
estimates of fRH are therefore critical to properly quantify
the aerosol direct effect.
The cavity ring down aerosol extinction spectrometer
(CRD-AES) developed at NOAA ESRL was deployed on
the NOAA RV Ronald.H.Brown during the TEXAQSGoMACCS summer 2006 study in the Gulf of Mexico to
assess the air quality of the Houston area and evaluate the
radiative impact of aerosols on local and regional scales.
The CRD-AES measured the aerosol extinction
coefficient at three wavelengths (355, 532, 1064 nm) and
at different relative humidities for both fine and coarse
aerosol sizes. This work presents selected aerosol types of
different chemical composition and origin (ship
emissions, urban outflow, continental airmasses, Saharan
dust), characterized based on fRH, Angstrom exponent
(from the CRD-AES), and single scattering albedo (from
CRD and Photoacoustic Aerosol Absorption
Spectrometer, PAS). Hydrophobic character is shown by
fresh emissions (such as ship plumes), whereas the
continental outflow exhibits some variability depending
on source, composition and degree of transformation.
Saharan dust properties are clearly affected by the level of
mixing with other aerosol species and by transformation
during the long range transport. These data will be
discussed with an emphasis on the proper treatment of the
variation of the optical properties with relative humidity.
Measured extinction coefficients for individual vehicle
plumes reached as high as 2000 Mm-1 with absorption
coefficients as high as 1800 Mm-1. Single scattering
albedo ranged from near 1.0 for clean air to 0.2 at a
wavelength of 675 nm for some exhaust plumes, which
was similar to that measured by us in pure BC from an
inverted diffusion flame. For the first time, particulate
emission factors are reported as extinction and absorption
cross section per kg_fuel, which is the important
parameter for visibility and climate studies. Measured
values will be compared to those reported in emission
inventories, used in climate models, and measured in
other field campaigns. Mass absorption efficiencies were
typically very low, 7.3 in the LD bore and 5.4 in the
mixed vehicle bore. These low values are consistent with
the fact that in the tunnel we sampled fresh vehicle
exhaust. Correlations between particulate optical
properties and more typical measurements are explored.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
3A.1
Tropopsheric Aerosol Chemistry via Aerosol Mass
Spectrometry. DOUGLAS WORSNOP Aerodyne Research,
University of Helsinki.
A broad overview of size resolved aerosol chemistry in
urban, rural and remote regions is evolving from
deployment of aerosol mass spectrometers (AMS)
throughout the northern hemisphere. Using thermal
vaporization and electron impact ionization as universal
detector of non-refractory inorganic and organic
composition, the accumulation of AMS results represent
a library of mass spectral signatures of aerosol chemistry.
For organics in particular, mass spectral factor analysis
provides a procedure for classifying (and simplifying)
complex mixtures composed of the hundreds or thousands
of individual compounds. Correlations with parallel gas
and aerosol measurements (e.g. GC/MS, HNMR, FTIR;
mostly on collected aerosol) supply additional chemical
information needed to interpret mass spectra. The
challenge is to separate primary and secondary,
anthropogenic and biogenic sources and transformations
of aerosol chemistry and microphysics. A summary of
recent progress, based on results from softer (chemical)
ionization approaches combined with volatility
measurements, will be presentes, comparing physical and
chemical properties for both laboratory and ambient
aerosol, including sampling of biomass burning
experiments.
3A.2
Measurements of the impact of aerosols on climate using online single particle mass spectrometryP. KIMBERLY
PRATHER, Scripps Institution of Oceanography, University of
California, San Diego.
Aerosol chemistry plays a critical but largely uncertain
role in affecting climate. Measurements of the mixing
state of individual particles as a function of size are
critical to advancing our understanding of the role of
aerosols in climate. The direct effect is determined by the
distribution of light absorbing and reflecting chemical
components amongst individual aerosol particles (i.e.
mixing state). The size and chemistry of particles also
impact cloud formation and the indirect effect; recent
studies debate whether size or chemistry is most
important. On-line single particle mass spectrometry has
been used in a number of field campaigns to acquire a
picture of the mixing state of dust, soot, sea salt, organic
carbon and other aerosols with secondary species such as
sulfate, nitrate, oxalic acid, and ammonium. This
presentation will discuss how single particle mass
spectrometry can provide new insights into the direct
effect of aerosols through measurements of the optical
properties of aerosols as a function of size and mixing
state. Results from field studies motivated a thorough set
of lab studies designed to investigate increases in CCN
activity of dust particles after heterogeneous reaction with
acidic gases. These experiments are designed to
investigate the ability of ATOFMS to quantitatively
assess the amount of secondary species associated with
the reacted dust particles, and how these chemical
changes alter the CCN activation properties of the mineral
dust. In this presentation, an overview will be presented
of synergistic lab and field investigations using single
particle mass spectrometry to understand the impacts of
aerosols on the direct and indirect effects of climate.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
3A.3
Examining the Relationship between El Nino, Biomass
Burning, and Aerosol Levels in the Southern United States.
BRET ANDERSON, Erik Snyder, U.S. Environmental
Protection Agency; Jay R. Turner, Washington University in St.
Louis.
In 2003, the National Aeronautical and Space
Administration (NASA) reported on a study that reviewed
data observed by NASA's TOMS satellite to quantify the
amount of aerosol from biomass burning from 1979 2000. In its study, NASA found that the highest levels of
aerosol of the 20 year study period were associated with
the increase of wildfire activity due to the El Nino/
Southern Oscillation (ENSO) of 1997-98. NASA's
MOPITT program reported that the the wildfire activity
and corresponding pollution associated with the 2006
ENSO event was the highest observed since the 1997-98
ENSO event.
Recently, the USEPA examined the relationship of ENSO
events on air quality in the Southern US. Analyzing
IMPROVE monitoring data for Big Bend National Park
for the years 1988 - 2004 using positive matrix
factorization (PMF), Anderson (2006) noted a persistent
biomass burning signal in the IMPROVE time series.
This occured annually between late April and late May,
corresponding to the annual fires in Central America and
Mexico. Air mass history analysis indicated these as a
likely source region for this biomass burning signal.
Anderson also noted that during certain years, the strength
of the PMF resolved biomass burning profile signal was a
factor of two to four times greater than the normalized
strength of the biomass signal. The two highest periods
corresponded with the ENSO events of 1997-98 and 2002
-03, with other elevated years corresponding to the
occurence of ENSO, corroborating previous NASA
studies. In this study, we examine the occurence of
ENSO events between 1988 - 2004, its impact upon
regional climate characteristics which lead to increased
wildfire activity, and the impacts of enhanced wildfire
activity on fine particulate levels in the Southern US,
focusing upon the Big Bend IMPROVE site in Texas.
3A.4
Extratropical waves drive boreal wildfire impact frequency
and regional air quality dynamics. KEITH BEIN, Yongjing
Zhao, Anthony Wexler, University of California Davis; Murray
Johnston, University of Delaware; Natalie Pekney, National
Energy Technology Laboratory; Cliff Davidson, Carnegie
Mellon University; Greg Evans, University of Toronto.
Source attribution analyses involving a combination of
Rapid Single-ultrafine-particle Mass Spectrometry data
(RSMS), satellite imagery and HYSPLIT trajectories have
been performed to identify a total of eight separate
wildfire events - five occurring in the boreal forest of
Canada and three in the western U.S. - that resulted in
significantly elevated levels of pollutants during the
months of June and July, 2002, at the Pittsburgh
Supersite. These results were corroborated by various
concurrent PM and gas measurements. In combination,
these data explicitly illustrate the impact of large scale
wildfires and reveal a larger structure in the nature of
pollution episodes in the Pittsburgh air shed and the
Northeastern US characterized by alternating periods of
stagnation and cleansing. In total, eight different
wildfires bounded by seven successive stagnation events
were observed. These receptor site dynamics were
correlated to the structure and propagation of extratropical
waves through analyses of 500hPa geopotential height
fields. Results revealed a connection between boreal fire
activity, southeast subsiding transport of the emissions,
alternating periods of stagnation and cleansing at the
receptor and the development and propagation of
amplified trough-ridge-trough configurations, where the
latter has been posited to drive the overall sequence of
events.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
3A.5
Mineral Dust Simulation in a Global Aerosol Microphysics
Model and Evaluation with Remote Sensing Data. YUNHA
LEE, Peter J. Adams, Carnegie Mellon University.
Mineral dust aerosol is climatically important because it
has significant direct forcing and also modifies the aerosol
size distribution in dusty regions. Global simulations of
aerosol climate effects require realistic treatment of dust
aerosols to understand long-range dust transport and
radiative forcing. A dust aerosol simulation is developed
for the TwO-Moment Aerosol Sectional (TOMAS)
aerosol microphysics model, which runs in the Goddard
Institute for Space Studies General Circulation Model
(GISS GCM) II-prime. A one-year dust simulation is
performed in conjunction with other important and
previously implemented tropospheric aerosols such as
sulfate, sea-salt, and carbonaceous aerosols. Dust
emissions are specified using the dust source function
given by Ginoux et al. [2001], the emissions
parameterization of Gillete and Passi [1988], and the
threshold friction velocity of Marticorena and Bergametti
[1995]. Dust emissions depend strongly on the surface
wind speeds; therefore we evaluate the GISS wind fields
against reanalysis fields from the National Center for
Environmental Prediction. Sensitivity simulations are
performed to assess the importance of biases in the GISS
wind fields. Dust simulations are evaluated using
observation data such as long term measurement of dust
concentrations performed by University of Miami. The
overall aerosol simulation is evaluated by comparing
model predicted aerosol optical depth (AOD) against
observations from MODIS, MISR, and AERONET. We
have developed an AOD calculation in the global model
in which aerosol species except hydrophobic elemental
carbon are assumed to be internally mixed. Water uptake
by sulfate, sea-salt, and organic aerosols is accounted for
and optical properties are calculated based on Mie theory.
3A.6
Effects of Photochemsitry and Convection on the UT/LS
Aerosol Nucleation: Observations. DAVID R. BENSON, LiHao Young, William M. Montanaro, Shan-Hu Lee, Kenst State
University; Heikki Junninen, Markku Kulmala, University of
Helsinki; Teresa L. Campos, David C. Rogers, Jorgen Jensen,
National Center for Atmospheric Research.
Nucleation is an important step in the chain reactions that
lead to cloud formation, but the nucleation mechanisms
are poorly understood. Recent studies show that new
particle formation is very active in the upper troposphere
and lower stratosphere (UT/LS). And, these results lead
to a new question: when does new particle formation not
occur? Here, we show how photochemistry, surface area
and convection affect new particle formation, using the
measured aerosol size distributions during the NSF/
NCAR GV Progressive Science Missions in December
2005. Three days of sunrise and sunset experiments were
made at the latitudes from 18 degrees N to 52 degrees N
and altitudes up to 14 km. This is the first time that
intensive nighttime aerosol measurements were made in
the UT/LS. Aerosol size distributions with diameters
from 4 to 2000 nm were obtained, along with other trace
gas species including water vapor, ozone, and carbon
monoxide. Surprisingly high concentrations of ultrafine
particles were seen continuously during the day and
nighttime with high aerosol growth rates, indicating
unknown sources of aerosol precursors and the particle
formation at night. Also, for air masses that had new
particle formation events were closely associated with
convection. On the other hand, for the cases where no
new particle formation events were observed, air masses
did not experience a vertical motion and there were also
high surface area densities. Latitude dependence of new
particle formation is also discussed, by comparing with
previous studies.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
4A.1
Global Contribution of Nucleation and Primary Particle
Emissions to CN and CCN. JEFFREY R. PIERCE, Peter
Adams, Carnegie Mellon University.
The relative contributions to the tropospheric burdens of
CN and CCN from nucleation versus primary emissions
are very uncertain. To evaluate these contributions, we
perform global aerosol microphysical simulations to
explore how uncertainties in nucleation mechanisms,
nucleation rates, primary emissions amount, and primary
emissions size distributions affect the CN and CCN
concentrations. The global model used is the GISS IIprime general circulation model with the size-resolved
aerosol microphysics module, TOMAS (1-2). Various
binary, ternary and ion-induced nucleation theories are
tested in the model to determine how they affect the
spatial and temporal distribution of CN and CCN (3-5).
For example, a robust feature of global aerosol
microphysics models is that binary nucleation
parameterizations predict high nucleation rates in the
upper troposphere, but the impact of these particles on
boundary layer CCN concentrations requires
investigation. A pseudo steady-state approximation is
applied to gas phase sulfuric acid allowing for longer time
steps during the calculation of nucleation and
condensation and a decrease the computation time. The
sensitivity of CN and CCN concentrations to the fraction
of sulfur emitted as \primary\ sulfate particles and the
emission size of biomass burning aerosol are explored.
Also investigated is the effect of sub-grid scale
coagulation on the concentration of CN and CCN.
4A.2
Linking Pacific Storms to Asian Pollution Aerosols. RENYI
ZHANG, Guohui Li, Jiwen Fan, Texas A&M University; Dong
L. Wu, Jet Propulsion Laboratory, California Institute of
Technology; Mario J. Molina, University of California.
Indirect radiative forcing of atmospheric aerosols by
modification of cloud processes poses the largest
uncertainty in climate prediction. In this talk, we present a
trend of increasing deep convective clouds over the
Pacific in winter from long-term satellite cloud
measurements (1984-2005). Simulations using a cloudresolving Weather Research and Forecast model reveal
that the enhanced deep convective clouds are reproduced
when accounting for the aerosol effect from the Asian
pollution outflow, which leads to intensified storms. We
suggest that the wintertime Pacific is highly vulnerable to
the aerosol-cloud interaction because of favorable cloud
dynamical and microphysical conditions from the
coupling between the Pacific storm track and Asian
pollution outflow. The intensified Pacific storm track is
climatically significant and represents possibly the first
detected climate signal of the aerosol-cloud interaction
associated with anthropogenic pollution. In addition to
radiative forcing on climate, intensification of the Pacific
storm track likely impacts the global general circulation
due to its fundamental role in meridional heat transport
and forcing of stationary waves.
(1) Hansen, J. et al., Mon. Weather Rev., 111, (1983).
(2) Adams, P. J., and J. H. Seinfeld., J. Geophys. Res.,
107, (2002).
(3) Vehkamaki, H., et al., J. Geophys. Res., 107, (2002).
(4) Napari, I., et al., J. Geophys. Res., 107, (2002).
(5) Modgil, M. S., et al., J. Geophys. Res., 110, (2005).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
4A.3
GCM Assessment of Aerosol-Cloud Interactions: The
Importance of Entrainment on Indirect Forcing and
Autoconversion. ATHANASIOS NENES, Donifan Barahona,
Georgia Institute of Technology; Peter J. Adams, Carnegie
Mellon University; John H. Seinfeld, California Institute of
Technology.
The goal of this study is to assess the importance of
entrainment on global cloud droplet number, indirect
forcing and autoconversion rates. This is accomplished by
using a state-of-the-art GCM, the NASA Goddard
Institute for Space Studies GCM Model II', coupled with
a) the TOMAS two-moment sectional aerosol
microphysics module (Adams and Seinfeld, 2002) and, b)
a comprehensive parameterization of cloud droplet
formation that explicitly treats film-forming organics and
entrainment (Barahona and Nenes, 2007) is used to
simulate aerosol-cloud interactions. Autoconversion rates
are computed using the physically-based parameterization
of Khairoutdinov and Kogan, (2000). The aerosol module
includes primary emissions, chemical production of
sulfate in clear air and clouds, new particle formation, dry
deposition, wet scavenging and hygroscopic water uptake.
Emissions include SO2 (fossil fuel and natural), organic
carbon (OC) and sea salt. In-situ observations from the
CIRPAS Twin Otter airborne platform obtained from a
variety of field campaigns (CRYSTAL-FACE, CSTRIPE,
ICARTT, MASE, GoMACCS) are used to constrain incloud vertical velocity, entrainment rate and droplet
growth kinetic parameters. The GCM radiative transfer
routine is used for calculations of indirect forcing. For the
first time, the sensitivity of indirect forcing and
autoconversion to droplet growth kinetics, in-cloud
dynamics and entraining is consistently explored within
the framework of a GCM.
4A.4
Cloud Condensation Nuclei Sizes. JAMES G. HUDSON,
Subhashree Mishra, Desert Research Insitute.
Cloud condensation nuclei (CCN) are characterized by
critical supersaturation (Sc), a function of size and
chemistry. Small variability of size-Sc measurements has
been cited as evidence that CCN can be deduced from
particle size measurements alone. This would have
advantages since size is easier to measure than chemistry
or CCN. However, we present size-Sc measurements
with a greater range of variability.
CCN size is determined by passing an aerosol through
a differential mobility analyzer (DMA) and then to a CCN
spectrometer (i.e., Hudson, 1989), which provides a mean
Sc for each particle size that is measured. Recent surface
and airborne measurements have confirmed Hudson and
Da (1996) that CCN are significantly smaller in cleaner
air masses where they behave like NaCl or ammonium
sulfate. CCN are two to four times larger in polluted air
masses.
Dusek et al. (2006) found only very large CCN and a
small size-Sc range. This led them to conclude that CCN
can be determined solely based on particle size. The much
larger range of CCN sizes indicates that this conclusion
might only be valid in polluted air masses. Dusek et al.
(2006) also said that, although there might be different
size-Sc ranges in different air masses, if there were
limited variability in size-Sc within each air mass, then it
still might be possible to deduce CCN from size
measurements, if size-Sc relationships are determined for
each air mass. However, this would require not only
measurements of such, but also an analysis similar to
Dusek et al. (2006) for each air mass. Mixed air masses
that are important for indirect aerosol effect research
would present a problem.
Dusek et al., (2006), Science, 312, 1375-1378.
Hudson, (1989), J. Atmos. & Ocean. Techn., 6, 1055
-1065.
Hudson and Da, (1996), J. Geophys. Res., 101, 4435
-4442.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
4A.5
Variations in Cloud Drop Number Concentrations with
Changes in Aerosol Hygroscopicity. Markus Petters, Trude
Eidhammer, SONIA KREIDENWEIS, Colorado State
University.
Cloud droplet number concentrations formed in adiabatic
updrafts are dependent on the initial thermodynamic
conditions, the ascent velocity of the air, and the belowcloud aerosol size distribution and chemical composition.
We have developed a single-parameter model of aerosol
hygrosopicity that can be used to characterize the cloud
condensation nucleus activity of a homogeneous particle
population. We use a parcel model, initialized with a
single lognormal input aerosol and run through a large
number of simulations, to establish the sensitivity of
simulated cloud droplet number concentrations to
variations in hygroscopicity. The results are used to
bracket the required accuracy in observations of aerosol
chemical composition and mean particle size for
application of such data to studies of aerosol indirect
effects.
4A.6
Aerosol Residual Water Content, CCN Activity and
Hygroscopicity of Mixed Aerosols. TIMOTHY RAYMOND,
Mark Zimmerman, Bucknell University.
Significant research in the past two decades has focused
on individual particulate constituents and their
interactions with water. Previously, the interaction
between water and the inorganic fraction of atmospheric
aerosols has been well characterized, and the role of
organics in aerosols is becoming more fully understood.
It still remains to investigate the water interactions of
particles containing numerous inorganic and organic
constituents to develop a more realistic simulation of the
complex nature of ambient particulates and to discover
the best way to simplify natural aerosols for modeling
studies.
In this work, we will present results of CCN studies along
with scanning electron microscope (SEM) and atomic
force microscope (AFM) imagery of complex aerosols.
These results are compared to single-component aerosols
of the complex aerosol constituent species obtained from
the literature and also from our experiments. A novel
method for determining residual water content of aerosol
particles has been developed.
The results indicate that most chemically complex
ambient aerosols would be expected to display similarly
CCN activity. This implies that even under low RH
conditions, most ambient aerosols would be expected to
contain water. This has been confirmed by SEM and
AFM investigations. Chemically complex aerosol
particles including multiple organic constituents have
been shown to retain water below 2% RH and to
demonstrate identical CCN activity compared to the same
aerosols generated at over 25% RH.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
5A.1
Cloud Processing of Atmospheric Organic Matter: New
Insights from LC/MS. JEFFREY L. COLLETT JR., Lynn. R.
Mazzoleni, Amy P. Sullivan, and Xinhua Shen, Colorado State
University.
Clouds and fogs act as important processors of
atmospheric organic matter. Carbonaceous aerosol
particles can serve as cloud condensation nuclei while
soluble volatile organic compounds dissolve into cloud
droplets. Chemical reactions inside cloud drops can
convert volatile organic compounds into less volatile
products that may be left behind as new particulate mass
when droplets evaporate. Surface-based fogs and
precipitating clouds can also remove scavenged
particulate matter from the atmosphere via occult or wet
deposition. The organic composition of clouds and fogs
is complex, reflecting the complex organic composition of
precursor aerosol particles, the composition of dissolved
volatiles, and products of aqueous phase reaction.
Understanding this composition, however, is critical to
improving our knowledge about cloud processing of
atmospheric organic matter. We report here on new
measurements of the organic composition of radiation
fogs made by liquid chromatography coupled with timeof-flight mass spectrometry (LC/ToF-MS). The accurate
mass capability of the Agilent ToF mass spectrometer
used in the experiments facilitates definition of likely
chemical formulae for observed molecular ion masses.
Analyses of the fog samples reveal the presence of high
molecular weight compounds (up to 1200 Da), a
prevalence of nitrogen- and sulfur-containing organics,
and changes in organic composition over the course of a
fog episode. Implications of these observations for our
understanding of aerosol chemistry will be discussed.
5A.2
The chemical composition of intercepted clouds in northern
Arizona during North American monsoon season. JAMES
HUTCHINGS, Jennifer Triplett, Heide McIlwraith, Pierre
Herckes, Arizona State University; Marin Robinson, Northern
Arizona University.
Clouds play an important role in the transport and
transformation of atmospheric pollutants from the gas and
particle phases. Although many laboratory and some
observational studies addressed acid rain formation in the
1980s and early 1990s, only a few studies have monitored
cloud water composition in recent years. Our current
knowledge of cloud chemistry and composition beyond
acidity and major ions is rather poor. Little observational
data exist on trace metals and organic species in clouds in
the U.S., and observations of cloud composition in the
Southwest are particularly rare.
The chemical composition of clouds in northern Arizona
was investigated during the North American Monsoon
seasons in 2005, 2006 and 2007. Intercepted clouds were
sampled on the top of Mt. Elden (9299 ft) near Flagstaff
(AZ) with an automated Caltech Active Strand
Cloudwater Collector (CASCC). A variety of chemical
species were determined including major ions, trace
metals, total organic carbon and volatile organic
compounds (VOCs). The pH of clouds was consistently
high (~6), likely the result of neutralization of acidity by
dust components as suggested by high calcium values.
Total organic carbon concentrations were high (9ppmC
on average) compared to typical pristine areas. While
VOCs like toluene, ethylbenzene and xylenes were
detected, indicating that these species are readily
scavenged by clouds, however they accounted for less
than 0.1% of the organic matter present. Clouds showed a
large inter-event variability which can be explained by air
mass history. Finally, we will compare our results to other
Areas in the US as this was the first study on cloud
chemistry in the Desert Southwest.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
5A.3
Chemistry of Organic Substances in Atmospheric Fog and
Cloud Waters: Insights from High Resolution Mass
Spectrometry. QI ZHANG, Yele Sun, University at Albany,
SUNY; Lynn Rinehart, Jeff Collett, Colorado State University.
Understanding the effects of fogs and clouds on the
microphysical and chemical processing of aerosol
particles requires detailed information on their chemical
composition. While inorganic compounds have been
studied extensively in fog and cloud waters, little is
known about the organic constituents, mainly due to
analytical difficulties. We have recently developed a
technique for characterizing organic compounds in
atmospheric aqueous phases using an Aerodyn HighResolution Time-of-Flight Aerosol Mass Spectrometer
(HR-ToF-AMS). This technique allows the quantification
of total organic mass and characterization of chemically
meaningful organic classes in fog and cloud waters. Fog
samples collected from Fresno, CA during winter 2006
and cloud samples collected from the Whiteface
Mountain, NY in summer 2006 were analyzed. Our
analyses indicate that organic substances account for ~ 10
- 40% of the total mass of dissolved materials in fog and
cloud waters. The chemical characteristics of major
organic classes such as hydrocarbon like, oxygenated,
nitrogen-containing species are evaluated based on high
resolution mass spectrometry. Organic nitrogen (ON)
compounds contribute a significant portion, ~10 - 20%, to
the total organic mass in the Fresno fog, but their
contribution to the WFM cloud organics is < 5%. We
also detect very low levels of other heteroatom organics
(e.g., phosphorus-, sulfur-, and halogen-containing) in
some Fresno fog samples. One possible source for these
compounds is pesticides and herbicides. The elemental
ratios of C : N : O are estimated for the bulk organics and
for the individual classes. These characteristics and the
overall mass spectral patterns of fog and cloud organics
are compared to those of ambient organic aerosol, from
which we will discuss some insights into the chemical
processes of organic aerosol in clouds and fogs.
5A.4
Cloud-Processing and Aerosol Optical Properties at a
Polluted Continental Site. ELISABETH ANDREWS,
University of Colorado and NOAA/GMD;John Ogren, NOAA/
GMD; James Allan, Keith Bower, Hugh Coe, Ben Corris,
Michael Flynn, Dantong Liu, William Morgan, Paul Williams,
University of Manchester.
The optical properties of aerosol particles are one of the
controlling factors in determining direct aerosol radiative
forcing. These optical properties depend on the chemical
composition and size distribution of the aerosol particles,
which can change due to various processes during the
particles' lifetime in the atmosphere. Here we present
results from a study investigating how cloud-processing
of atmospheric aerosol changed aerosol properties at a
polluted continental site. Aerosol physical, chemical and
optical properties were measured continuously at Holme
Moss, UK in late 2006.
While the aerosol light extinction (extinction = absorption
+ scattering) measured at Holme Moss was similar to that
measured at other rural continental sites, the single
scattering albedo (SSA) at Holme Moss was significantly
lower. The Holme Moss aerosol was very absorbing median SSA was around 0.82 (rural continental values of
SSA in the US tend to be 0.92-0.95). The aerosol
absorption can likely be attributed to urban, industrial and
diesel emissions upstream of the sampling site.
During the three week study, there were six cloud events
which provided ample opportunity to study how this very
polluted aerosol changed during cloud processing.
During cloud events the SSA of the interstitial aerosol
(the aerosol not in the cloud drops), was even lower than
that observed during clear periods. Indicators of particle
size showed that the interstitial aerosol was also smaller
than the typical ambient aerosol. Measurements made
downstream of a counterflow virtual impactor, a special
inlet which sampled only cloud droplets, showed that the
aerosol scavenged by cloud drops was larger in diameter
and less absorbing than both the interstitial aerosol and
the ambient aerosol observed during cloud free
conditions. Both of these observations are consistent with
the notion that larger, scattering aerosol is preferentially
scavenged by cloud droplets due to its more hygroscopic
nature.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
5A.5
Interaction of Saharan Dust with Liquid and Ice Clouds.
CYNTHIA TWOHY, Oregon State University; Andrew
Heymsfield, Aaron Bansemer, National Center for Atmospheric
Research; Bruce Anderson, NASA Langley Research Center.
The liberation of mineral dust from Africa has increased
in magnitude over past decades. The impact of this dust
on tropical convection is potentially large. Dust is known
to be an effective ice nucleus, and may also act as a cloud
condensation nucleus if coated with soluble material.
Studies have shown that the Saharan dust layer is
anticorrelated with tropical cyclone and hurricane activity
in the Atlantic. Whether this effect is caused by
dynamical and radiative effects, and/or dust nucleation
impacts on latent heat release and vertical transport is not
known.
In the NASA African Monsoon Multidisciplinary
Activities (NAMMA) experiment, aerosol particle
physiochemical characteristics and cloud size
distributions were measured aboard the NASA/University
of North Dakota DC-8 aircraft in summer of 2006. Both
low-level warm clouds, deep convection, and anvil cirrus
outflow from mesoscale systems impacted by various
amounts of dust were sampled. Ambient aerosol and
cloud residual particles were collected with a counterflow
virtual impactor (CVI) to assess the percentage and size
of dust particles actually incorporated into these clouds.
Analysis of microphysical properties of small
stratocumulus over the ocean revealed droplet number
concentrations ranging from 100 cm-3 to as high as 900
cm-3. In most cases, concentrations were higher than
expected for clean marine clouds, despite low liquid water
contents. This implies that a substantial fraction of dust
particles are acting as cloud condensation nuclei in the
region. Correlations between aerosol number and
properties and droplet number will be presented.
Samples of ambient aerosol and residual nuclei are
being analyzed at Oregon State University by
transmission electron microscopy. Particles are identified
as crustal dust, salts, sulfate, soot, organics, or mixtures of
these types. These results will be combined with cloud
microphysical measurements to determine primary
nucleating agents of eastern Atlantic storm systems that
may spawn hurricanes.
5A.6
A Further Analysis of the Phase Transitions in Mixed Phase
Cloud During the CLACE Series of Aerosol-Cloud
Interaction Experiments at the Jungfraujoch High Alpine
Research Station, Switzerland. KEITH N. BOWER, Ian
Crawford, Tom Choularton, Martin Gallagher, Paul Connolly,
Hugh Coe, Michael Flynn, Jonny Crosier, University of
Manchester; Ernest Weingartner, Urs. Baltensperger, Rami
Alfarra, Paul Scherrer Institut, Switzerland; and Bart Verheggen,
ETH, Switzerland.
A series of Cloud-Aerosol Characterisation Experiments
(CLACE) have been performed at the Jungfraujoch
mountain top site (3580masl) in the Swiss Alps under
different meteorological conditions and season, to
investigate the relationship between clouds and the
aerosol population upon which they form, both in warm
(CLACE-2, July 2002) and at continuously sub-zero
temperatures (CLACE-3/4/5/6 in Feb/March 2004/5/6/7),
respectively.
A suite of instrumentation was deployed to measure
aerosol properties as well as the microphysics of the
clouds. Internally, a switching total and interstitial (i/s)
sampling inlet system enabled sequential measurements
of the total sub-micron aerosol population (dry cloud
particle and i/s residuals) and i/s particles separately.
Measurements of dry aerosol size distribution (SMPS and
OPC), composition (Q and ToF AMS) and hygroscopicity
(H-TDMA) amongst others were made. Externally, high
frequency cloud microphysics measurements were made
from a raised platform and tower mounted rotating cross
arm to direct cloud probes into wind. Ice crystal habit and
size distributions were measured by a Spec Inc. Cloud
Particle Imager (CPI) and cloud droplet size distributions
and liquid water content were measured using a Forward
Scattering Spectrometer Probe (FSSP-100). Gerber PVMs
also measured cloud liquid water content. Formvar
replicas of ice crystals were collected for later analysis by
ESEM.
Results indicate aerosol composition dominated by
sulphate and organics (the latter showing a high degree of
photochemical ageing, reduced in winter). Generally, a
single hygroscopic growth mode is seen, which together
with composition data suggests an internally mixed
aerosol composition. Growth factors are lower in winter.
In wintertime, rapid transitions from wholly supercooled
to fully glaciated clouds on timescales of a second or less
are often observed. Cloud Ice Mass Fraction (IMF) has
been calculated and its relationship to a number of
parameters - including aerosol properties, examined.
These results will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
7A.1
Cloud activating properties of aerosol observed during the
Marine Stratus/Stratocumulus Experiment (MASE). JIAN
WANG, Yin-Nan Lee, Peter Daum, Brookhaven National
Laboratory; Liz Alexander, Pacific Northwest National
Laboratory; John Jayne, Aerodyne Research Inc.
7A.2
Study of the nucleation of cloud droplets on ambient
aerosols in stratiform and convective cloud. W. RICHARD
LEAITCH, Wanmin Gong, Desiree Tom-Sauntry, Katherine
Hayden, Anne Marie Macdonald, Kurt Anluaf, Shao-Meng Li,
Walter Strapp, Mohammed Wasey, Environment Canada.
The microphysics, CCN concentrations, and chemical
composition of marine aerosol were characterized on
board the Department of Energy Gulfstream-1 aircraft
during the Marine Stratus/Stratocumulus Experiment
(MASE) conducted over the coastal waters between Point
Reyes National Seashore and Monterey Bay, California,
in July 2005. Aerosol size distribution ranging from 15 to
500 nm was measured by a Scanning Mobility Particle
Sizer every 1 minute. Aerosol components, including
sea-salt- (sodium, chloride, magnesium, methansulfonate)
and terrestrial/pollution-derived (ammonium, sulfate,
nitrate, organics, potassium, and calcium) were measured
using the particle-into-liquid sampler-ion chromatography
technique and an Aerodyne AMS at a time resolution of 4
min and 30 s, respectively, both covering the size range of
~0.08 to 1.5 micrometers. CCN concentrations at 0.08%
and 0.2% supersaturations were determined at a 1-s time
resolution using CCN counters (DMT Inc). Aerosols
sampled during these flights ranged from relatively clean
marine aerosol to aerosols that were substantially
influenced by anthropogenic emissions. Closure analyses
are carried out by comparing the CCN concentrations
calculated from the measured size distribution and
chemical composition using modified Kohler theory to
simultaneous measurements. The agreements between
the calculated and measured CCN concentrations are
compared for different airmasses, and the effects of
organic species on aerosol cloud activation properties are
discussed. The CCN concentrations are also derived
using various simplifications of the measured aerosol
chemical composition, and compared to simultaneous
measurements. As the atmospheric aerosol often consists
of numerical species that can not be individually
simulated in global or regional models, analyses
employing various simplification of chemical
composition provide insights into the essential (or
minimum) information of particle chemical composition
that needs to be represented in these models to adequately
predict the CCN concentration and cloud microphysics.
Measurements of aerosol particles and clouds from two
studies are used with an adiabatic parcel model to
examine the factors constraining the closure of aerosols
and cloud droplet number concentrations. One study was
conducted in stratiform cloud over the western Altantic in
the fall of 2003 and the other was conducted in towering
cumulus over the Great Lakes region in 2004 during
ICARTT. It is particularly important to understand two
components: the relevant updraft speed and the growth
rates of the particles/droplets that are controlled by the
mass accommodation, the particle chemistry and the
particle size distribution. Measurements of CCN seldom
reflect the particle growth rates.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
7A.3
Aerosol hygroscopicity and CCN distributions at Gosan and
Seoul, Korea, measured in Summer and Autumn 2006.
SEONG SOO YUM, J. H. Kim, S.-C. Lee, K. Y. Song, S. B.
Shim, Yonsei University; James G. Hudson, Desert Research
Institute; Kang H. Ahn, Hanyang University.
A Hygroscopicity Tandem Differential Mobility Analyzer
(H-TDMA) system was set up to measure hygroscopic
growth properties of atmospheric aerosols at the relatively
clean coastal environment of Gosan, Jeju Island, and
highly populated city of Seoul, Korea. Measurement
periods were in August for Gosan and in October for
Seoul. Simultaneously CCN distributions were measured
by a Droplet Measurement Technologies (DMT) CCN
Counter and by the DRI (Desert Research Institute) CCN
Spectrometers. Submicron aerosol size distributions and
total particle concentrations were also measured by an
SMPS and two CPCs with different size cuts (3 and 10
nm). Measured hygroscopic growth factors were on
average 1.37, 1.56, 1.61 and 1.64 at Gosan and 1.25, 1.32,
1.33 and 1.43 in Seoul for the four mobility diameters, 50,
100, 150 and 200 nm, respectively. Measured CN (< 10
nm) and CCN (1% supersaturation) concentrations were,
on average, 4785 cm-3 and 2140 cm-3 for Gosan and
15765 cm-3 and 3527 cm-3 for Seoul. On average CCN
to CN ratios were 0.64 and 0.25 for Gosan and Seoul,
respectively. There was a very good agreement between
the two CCN instruments (DMT and DRI): relative errors
were 5.3% and 10.2% at 0.2% and 1% supersaturation,
respectively. H-TDMA measurement and aerosol size
distribution measurement results were applied to predict
CCN distributions. This method used more chemical
information than previous assumption of pure inorganic
salts, i.e., ammonium sulfate or sodium chloride and
produced a better agreement with the measured CCN
distributions.
7A.4
Analysis of Cloud Condensation Nuclei using a Pumped
Counterflow Virtual Impactor and Aerosol Mass
Spectrometer. JAY SLOWIK, Jonathan Abbatt, University of
Toronto; Richard Leaitch, Environment Canada.
We present a new method of determing the size and
composition of CCN-active aerosol particles. A
continuous-flow thermal-gradient diffusion chamber
(TGDC), pumped counterflow virtual impactor (PCVI),
and Aerodyne time-of-flight mass spectrometer (AMS)
are operated in series. Ambient particles are sampled into
the TGDC, where a constant supersaturation is
maintained, and CCN-active particles grow to about 2-3
microns. The output flow from the TGDC is directed
into the PCVI, where a counterflow of dry N2 gas
opposes the particle-laden flow, creating a region of zero
velocity. This stagnation plane can only be traversed by
particles with sufficient momentum, which depends on
their size. Particles that have activated in the TGDC cross
the stagnation plane and are entrained in the PCVI output
flow, while the unactivated particles are diverted to a
pump. Because the input gas is replaced by the
counterflow gas with better than 99% efficiency at the
stagnation plane, the output flow consists almost entirely
of dry N2 and water evaporates from the activated
particles. In this way, the system yields an ensemble of
CCN-active particles whose chemical composition and
size are analyzed using the AMS. Preliminary
experiments on urban aerosol in downtown Toronto
identified an external mixture of CCN-active particles
consisting almost entirely of ammonium nitrate and
ammonium sulfate, with CCN-inactive particles of the
same size consisting of a mixture of ammonium nitrate,
ammonium sulfate, and organics. We will discuss results
from the first field deployment of the TGDC-PCVI-AMS
system, to be conducted from mid-May to mid-June 2007
in Egbert, Ontario, a semirural site ~80 km north of
Toronto influenced both by clean air masses from the
north and emissions from the city.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
7A.5
Measurements of the Rate of Cloud Droplet Formation on
Atmospheric Particles. CHRIS RUEHL, Patrick Chuang,
Univeristy of California, Santa Cruz; Athanasios Nenes, Georgia
Institute of Technology.
The influence of aerosols on cloud properties is an
important modulator of the climate system, and remains
one of the most uncertain components of the
anthropogenic influence on the radiative budget of the
atmosphere. Traditional Kohler theory can predict the
ability of an atmospheric particle of known size and
composition to act as a cloud condensation nucleus
(CCN) at equilibrium. However, it is not known to what
extent particles exist in the atmosphere that may be
prevented from acting as CCN by kinetic limitations. We
measured the rate of cloud droplet formation on
atmospheric particles sampled at four sites across the
United States during the summer of 2006: Great Smoky
Mountain National Park, TN; Bondville, IL; Houston,
TX; and Lamont, OK. We parameterized droplet growth
rates with the mass accommodation coefficient (alpha),
and report values of alpha measured in the field
normalized to alpha measured for lab-generated
ammonium sulfate (AS) particles (i.e., alpha' = alpha/
alphaAS ). On 7 out of 16 days during which these
measurements were made, >20% of the particles observed
during at least one scan had alpha'<10-0.5, and ~4% had
alpha'<10-1. On the other 9 days, all ambient particles
formed cloud droplets at approximately the same rate as
AS particles (i.e., <10% had alpha'<10-0.5, and <1% had
alpha'<10-1). The highest observed proportions of lowalpha' particles were ~50% with alpha'<10-0.5 and ~10%
with alpha'<10-0.5, in Illinois. Day to day variability was
greatest in Tennessee and Illinois, and low-alpha' particles
were observed on days when NOAA HYSPLIT backtrajectories suggested that air was arriving from aloft.
These results suggest that for some air masses, accurate
quantification of CCN concentrations may need to
account for kinetic limitations.
8A.1
The Ability of Fresh and Aged Monoterpene Secondary
Organic Aerosol to Act as Cloud Condensation Nuclei.
GABRIELLA ENGELHART, Spyros Pandis, Carnegie Mellon
University; Spyros Pandis, University of Patras, Greece; Akua
Asa-Awuku, Athanasios Nenes, Georgia Institute of
Technology.
The ability of secondary organic aerosol (SOA) particles
formed during the ozonolysis of alpha-pinene and other
monoterpenes to act as cloud condensation nuclei (CCN)
was investigated using a static CCN counter and a
cylindrical continuous-flow streamwise thermal gradient
CCN counter developed by Droplet Measurement
Technologies (DMT). Secondary organic aerosol (SOA)
was produced from the reaction of alpha-pinene and
monoterpene mixtures (alpha-pinene, beta-pinene,
3
limonene and 3-carene) with ozone in a 12 m ,
temperature-controlled smog chamber. The initial
monoterpene concentrations were in the 10-30 ppb range
and an excess of ozone was used. The CCN
concentration, activation diameter and droplet growth
kinetic information were monitored as a function of
supersaturation for several hours and their changes with
age were quantified.
Both fresh and aged monoterpene SOA are quite active as
CCN. The initial concentrations of ozone and
monoterpene precursor do not appear to affect the activity
of the resulting SOA. However, reactions of the hydroxyl
radicals produced during the monoterpene ozonolysis lead
to further oxidation of the SOA material and an
improvement of their CCN properties with time. The
DMT CCN counter measured a decrease in CCN
activation diameter for alpha-pinene SOA of
-1
approximately 3 nm hr at 0.33% supersaturation. The
activation diameters of alpha-pinene and mixed
monoterpene SOA were consistent (within 10% or so)
with the predictions of classical Kohler theory assuming
that all the material was soluble in water.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
8A.2
Synthetic Biomass Aerosol Activation in Static and
Continuous-flow CCN Instruments. JEFFERSON R.
SNIDER, University of Wyoming; Heike Wex, Leibniz Institute
for Tropospheric Research, Leipzig, Germany; Adam
Kristensson, University of Copenhagen; Diana Rose, Max
Planck Institude for Chemistry, Mainz, Germany.
Four CCN instruments were used to sample nearly
monodisperse aerosol prepared at the Leipzig Aerosol
Cloud Interaction Simulator (LACIS) facility. Included
were two Wyoming static diffusion CCN instruments, and
the continuous-flow DMT and LACIS instruments. The
aerosols were composed of ammonium sulfate,
ammonium sulfate and soot, levoglucosan and soot, and a
mixture of ammonium sulfate, levoglucosan and soot. A
spark discharge was used to prepare the soot particles the mobility diameter of the prepared particles was
evaluated using electrostatic classifiers operated up and
downstream of the ammonium sulfate or levoglucosan
coating ovens. The goal of the work was two-fold. First,
to study aerosols somewhat characteristic of those
produced by biomass combustion followed by aging
within the atmosphere. Of importance is the extent to
which these particles function as cloud droplet nuclei.
Second, the collection of CCN instruments provided the
opportunity to compare instrument-based values of the
activation supersaturation.
The Wyoming and the DMT instruments scanned
supersaturation over a range of values that produced
negligible to complete activation. Utilized in the data
analysis were measurements of total particle
concentration derived from a condensation particle
counter. The total concentration values were used to
normalize the measurements of CCN concentration, thus
forming a supersaturation-dependent activation fraction.
Fits of activated fraction versus supersaturation were used
to derive a 50% activation supersaturation (i.e., the
aerosols critical supersaturation). The comparison of the
critical supersaturations is quite encouraging - with few
exceptions the instruments produced values which agreed
within measurement error. This was the case for
ammonium aerosols prepared between 35 and 95 nm, and
for coated-soot aerosols prepared at 84 nm. Critical
supersaturations derived from measurements made by the
LACIS instrument were also in good agreement with the
other two techniques.
8A.3
Cloud Condensation Nucleus (CCN) Behavior of Organic
Aerosol Particles Generated by Atomization of Water and
Methanol Solutions. TRACEY A. RISSMAN*, Varuntida
Varutbangkul**, Jason D. Surratt, Richard C. Flagan, John H.
Seinfeld, California Institute of Technology; David O. Topping,
Gordon McFiggans, The University of Manchester (*Currently
with DuPont, **Currently with Boston Consulting Group).
Cloud condensation nucleus (CCN) experiments were
carried out for malonic acid, succinic acid, oxalacetic
acid, DL-malic acid, glutaric acid, DL-glutamic acid
monohydrate, and adipic acid, using both water and
methanol as atomization solvents, at three operating
supersaturations (0.11%, 0.21%, and 0.32%) in the
Caltech three-column CCN instrument (CCNC3).
Predictions of CCN behavior for five of these compounds
were made using the Aerosol Diameter Dependent
Equilibrium Model (ADDEM). The experiments
presented here expose important considerations associated
with the laboratory measurement of the CCN behavior of
organic compounds. Choice of atomization solvent results
in significant differences in apparent CCN activation for
some of the compounds studied, which could result from
residual solvent, particle morphology differences, and
chemical reactions between the particle and gas phases.
Also, significant changes in aerosol size distribution
occurred after classification in a differential mobility
analyzer (DMA) for malonic acid and glutaric acid,
preventing confident interpretation of experimental data
for these two compounds. Filter analysis of adipic acid
atomized from methanol solution indicates that gasparticle phase reactions may have taken place after
atomization and before methanol was removed from the
sample gas stream. Careful consideration of these
experimental issues is necessary for successful design and
interpretation of laboratory CCN measurements.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
SYMPOSIUM: Aerosols, Clouds & Climate
2007 AAAR Annual Conference Abstracts
8A.4
The Impact of Surface Ocean Organics on Surface Tension,
CCN Activity, and Droplet Growth Kinetics of Marine
Aerosol. RICHARD MOORE, Ellery Ingall, Athanasios Nenes,
Georgia Institute of Technology.
8A.5
Studying the properties and vapor processing of organic
coated water droplets using Molecular Dynamics
Simulation. PURNENDU CHAKRABORTY, Michael
Zachariah
Oceanic surface waters contain substantial amounts of
organic surfactants that are transferred to the aerosol
phase during the process of sea spray generation. It is
known that these organics depress surface tension, and
such a depression could significantly alter the ability of
marine aerosol to act as cloud condensation nuclei (CCN).
While the CCN properties of the inorganic marine aerosol
fraction have been studied extensively, the contribution of
the organic fraction is less well-understood. It would be
advantageous, then, to be able to isolate the organic
species present in marine aerosol in order to characterize
their aggregate thermodynamic properties (e.g., molar
volume, surface tension, and water uptake coefficient)
and better understand the role of organics in marine CCN.
Atmospheric aerosols play a very important role in
atmospheric processes and have a major influence on the
global climate. Recently, it has been shown that fatty
acids reside on surfaces of sea-salt and continental
aerosols.
In this study, we report results of a Molecular Dynamics
(MD) study on the unique properties of fatty acid coated
water droplets. In particular we have found that for
particles preferring an inverted micelle structure, the
lower chain-chain interaction, with increasing radial
distance from the water-fatty acid interface, results in a
negative internal radial pressure profile for the organic
layer. Using a simple geometric model, we have
illustrated that this negative pressure is a manifestation of
the curved surface. As a result, the particle seems to
behave in a manner consistent with a
A recently-developed technique using electrodialysis and
reverse osmosis to remove the electrolyte salts and
concentrate the dissolved organic matter present in
seawater provides the means to accomplish this goal.
Samples were taken in July 2006 in the Atlantic Ocean
1
surface waters offshore of Georgia . Organics present in
the concentrated sample likely include surface-active
species such as short-chain fatty acids, proteinaceous
material, and humic substances among others. Because
marine aerosol are formed from sea spray, which involves
the generation of droplets from wave breaking, the
chemical species found in surface waters are expected to
be the same as those in marine aerosol.
In the present study, we examine, using Kohler Theory
2,3
Analysis , the surface tension, CCN activity, and droplet
growth kinetics of laboratory-generated particles
containing mixtures of these concentrated organics and
inorganic salts. Previous studies have observed sizedependent enrichment of organic surfactants in marine
2,3
aerosol , and we simulate this enrichment by varying the
ratio of organics to salts in the laboratory-generated
particles. The implications of organic enrichment on
droplet activation and growth will be discussed.
1
Vetter, T.A., et al., in review.
Asa-Awuku, A., et al., in review.
3
Padro, C., et al., in review.
4
Oppo, C., et al., Mar. Chem., 63, 1999.
5
O Dowd, C., et al., Nature, 431, 2004.
2
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
SYMPOSIUM: Innovation in Medicinal Nanopoarticles
2D.1
The Effect of Drug Physico-Chemistry on Pulmonary
Absorption Pharmacokinetics in Dogs. KATHLEEN SIMIS,
Peter Lloyd, Ron Hale, Alexza Pharmaceuticals.
2D.2
Development of AERx Essence for Delivery of Novel
Inhalation Formulations. DEBBIE YIM, Eric Johannson,
David Cipolla, Aradigm Corporation.
Dronabinol (THC, delta-9-tetrahydrocannabinol) is the
primary active compound in marijuana (Cannabis sp.) and
has garnered increasing attention in the medical
community as a result of its complex and widespread
systemic effects. The medical indications that have been
reported for THC (and other cannabinoids) are numerous
and most notably include appetite stimulation in patients
with AIDS, nausea and vomiting associated with
chemotherapy, and neuropathic pain and spasticity
associated with multiple sclerosis (1). Alexza's Staccato
condensation aerosol generation system is an excellent
drug delivery platform for THC. The Staccato system is a
breath actuated inhaler that generates an aerosol with an
appropriate particle size range for deep lung delivery and
rapid systemic absorption of the drug.
Aradigm has developed a family of devices for use with
the AERx strip. The first generation device, currently
being used for the delivery of insulin, is a batterypowered, hand-held, electromechanical device designed
for extremely precise systemic drug delivery. This device
has been used previously to efficiently deliver a variety of
solution formulations to the lung. Aradigm has recently
developed an all-mechanical, second-generation device
platform called Essence. This device is intended for both
systemic and precision topical delivery applications. Key
advances have enabled Essence to offer very similar
aerosol performance in a light, palm-size device.
Preliminary in vitro performance of the AERx Essence
platform using a simple cromolyn solution formulation
has been shown to be reproducible with emitted aerosol
from an individual device averaging 60.0% +/- 2.1%
(n=40 strips) and 56.8% +/- 4.6% across fifty devices
(n=8-10 strips per device). The fine particle fraction less
than 4.95 microns is 90% resulting in a predicted fine
particle dose (ED x FPF) of approximately 51-54%.
Essence also has the capability to effectively deliver more
sophisticated suspension and liposomal formulations.
Suspensions of ketoprophen and indomethacin were
tested at a concentration of 60 mg/mL using both the
electronic AERx with micron-sized nozzles and AERx
Essence with sub-micron-sized nozzles. The emitted
doses (EDs) were in the range of 49 to 60 % of the loaded
dose. Liposomal ciprofloxacin solutions at 50 mg/mL
were also aerosolized in Essence with approximately 42%
mean emitted dose. Following aerosolization, the
liposomes retained >95% encapsulation. This work
demonstrates that nano-suspension formulations can be
successfully delivered using the AERx Essence allmechanical system with sub-micron-sized nozzles.
Dronabinol is a moisture and light sensitive viscous liquid
with poor shelf-life stability. A thermally-labile solid
prodrug of THC has been identified that meets chemical
and physical shelf-life stability requirements. When
heated under optimum conditions using Alexza's Staccato
system, the prodrug is converted to THC (~90%) and
vaporized to form a pharmaceutically active aerosol
containing both THC and intact prodrug.
Delivery of the aerosol to dogs via inhalation resulted in
rapid systemic absorption and high bioavailability of THC
with demonstrable and significant differences (3-fold) in
time to maximum plasma concentration (Tmax) between
THC and its prodrug. These differences may be
attributable to differences in the physicochemical
properties of the two molecules.
This study with thermally-labile prodrugs further
demonstrated the utility of Alexza's Staccato system in
pulmonary delivery of pharmaceutical condensation
aerosols and provided insight into the physicochemical
mechanisms that govern drug transport from pulmonary
to systemic circulation.
1. Janet E. Joy, Stanley J. Watson, Jr., and John A.
Benson, Jr., Editors, 1999.Marijuana and Medicine:
Assessing the Science Base, Washington, D.C.: National
Academy Press
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
2D.3
Electromechanical Properties Analysis of Four Pressurized
Metered Dose Inhalers Using Laser Doppler Velocimetry.
MOHAMMED ALI, Rama Reddy, and Malay Mazumder,
University of Arkansas at Little Rock.
A novel method for analyzing electromechanical
properties (e.g., size, electrostatic charge, polarity) of
therapeutic aerosols produced by four different
commercially available pressurized metered dose inhalers
(pMDIs), including Albuterol (TM), Atrovent (TM), Qvar
(TM), and, Ventolin (TM) is presented. In recent,
influence of electrostatic charge on particle deposition in
the respiratory airways has attracted much attention,
which requires precise quantification from analytical
perspective. Experimental studies using Electrical Low
Pressure Impactor (ELPI) reported the net charge (q) and
aerodynamic diameter (da) of the pMDIs. However, the
ELPI has a limitation of providing the net charge of all
particles deposited on its impactor plate, not for each
particle in real time. To resolve this issue, we report the
application of an Electronic Single Particle Aerodynamic
Relaxation Time (ESPART) analyzer, which operates on
the principle of Laser Doppler velocimetry to measure
simultaneously da and q (magnitude and polarity) on a
single particle basis and in real time. Its draws aerosols
from an aerosol sampling chamber (ASC). The chamber's
inside walls were lined with a grounded wire mesh. The
pMDI devices were actuated at the inlet of valve holding
chamber, which had the other end connected to the ASC.
Prior to each run ASC was cleaned and evacuated (50 mb)
to simulate the inhalation of an aerosol bolus @ 30 L/m
for 8 s. Aerosol particles from all drug delivery devices
were found to not only have different size distributions
but also varied in their polarities. The drug aerosols cloud
emitted by Albuterol and Ventolin were determined to be
electropositive while Atrovent and Qvar were
electronegative. Count and mass distributions were
reproducible for all pMDIs. These findings can be
explained by variation in the drug propellant surfactants,
metal surfaces of delivery devices, and drug/carrier
homogeneities. In conclusion, the ESPART provided
more detailed charge information about the pMDI aerosol
particles.
SYMPOSIUM: Innovation in Medicinal Nanopoarticles
9B.1
Inhaled Liquid Vaccines: Implications for Devices and
Delivery. JAMES FINK, Nektar Therapeutics.
Liquid vaccines approved for subcutaneous
administration have been successfully administered by
inhalation. In Mexico, 4 million children were vaccinated
by aerosol inhalation using a compressor driven jet
nebulizer over a 12year period, with similar or better
protection than subcutaneous injection. The World
Health Organization in collaboration with the Gates
Foundation, has launched an initiative to license one or
more inhaled vaccines for use in the third world during
this decade. Several other vaccines have been identified
as likely candidates for aerosol administration, in a range
of environments. Device design is an integral component
of any successful program for liquid aerosol vaccines.
Beyond elimination of needles, aerosol administration of
liquid vaccines allows use of proven safe and effective
vaccines with a minimum of reformulation, and the
potential of providing 2 - 10 fold more vaccinations from
the same volume of vaccine. To realize these benefits,
liquid aerosol systems must be efficient, portable,
inexpensive, and easily used in either the clinic or the
field.
The age of subjects vaccinated, the need to reduce second
hand exposure greatly impact design requirements for
aerosol vaccine delivery. In many cases, vaccines are
administered to infants, children and adults. While simple
mouthpieces can work for subjects greater than 3 years of
age, the infant requires a mask. Reducing second hand
exposure is of some importance when aerosolizing live
virus vaccines. Masks capable to firmly seal to contain
aerosol, and filters to collect exhaled aerosol can add cost
and complexity to the device.
In the clinic, vaccination may be limited to a few subjects
per day, while in mass campaigns several hundred
vaccinations may occur in a few hours. To meet cost
targets, liquid aerosol generators may be used to safely
administer aerosol for multiple vaccinations. Challenges
and examples of device strategies will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
9B.2
The Staccato System for Thermal Aerosols and its Clinical
Evaluation. DAN MYERS, Pravin Soni, Jim Cassella, Ramesh
Damani, Reynaldo Quintana, Martin Wensley, Pete Lloyd,
Patrik Munzar, Krishna Sharma, Amy Lu, Ron Hale, Alexza
Pharmaceuticals; Josh Rabinowitz, Princeton University.
Pharmaceutical aerosols for systemic delivery via the lung
can be generated by rapid, controlled heating of a thin
film of pure drug. This method of excipient-free aerosol
delivery is applicable to a large number of clinicallyrelevant pharmaceutical compounds. Alexza incorporates
this concept into its proprietary Staccato (R) system for
aerosol drug delivery. Alexza is currently developing two
versions of the Staccato system. One is a single dose,
fully disposable device, which produces aerosols by using
an exothermic chemical heating source to vaporize a drug
film coated on a stainless steel substrate. The other
device is a multi-dose configuration, which uses electrical
power to resistively heat thin stainless steel foil
substrates. Both Staccato systems are portable, easy to
use, and create consistent aerosols of high emitted dose,
very low levels of thermal degradants, and ideal particle
size for systemic delivery over a broad range of patient
use conditions. The Staccato system is breath actuated
and delivers the aerosol into the lungs in less than 1
second.
Phase 1 clinical testing of four different pipeline products
has been completed, showing good tolerability, rapid
absorption into systemic circulation, and high
bioavailability of the drugs. Initial Phase 2 trials have
also been conducted with two products. Staccato
Prochlorperazine showed efficacy in treatment of
migraine headache as early as 15 minutes after drug
administration, while Staccato Loxapine was effective in
treatment of agitation in schizophrenic patients as early as
20 minutes after dosing. Based on these early clinical
observations, the simple design and optimal aerosol
characteristics of the Staccato system eliminate much of
the patient variability inherent in metered dose and dry
powder inhalers, and lead to rapid delivery and absorption
of the drug into the bloodstream. The fast onset of
pharmacological action of the Staccato system could
provide great benefit for many therapeutic classes.
SYMPOSIUM: Innovation in Medicinal Nanopoarticles
9B.3
Development of Inhalable Nanoparticles. RAIMAR
LOEBENBERG, Warren H Finlay, University of Alberta;
Wilson H Roa, Cross Cancer Institute; Elmar J Prenner,
University of Calgary.
Nano-technology can be considered a new frontier in
biomedical sciences. Delivery systems in the nanometer
range are very promising drug carriers due to their ability
to overcome many limitations associated with
conventional drug delivery systems including multi drug
resistance in cancer treatment. The presentation will
review briefly some historical aspects of nanoparticle
based drug delivery technology. Examples of different
applications of nanoparticles in drug delivery will be
given.
The lungs are getting more and more attention as possible
absorption organ for molecules which are difficult to
formulate for the oral or intravenous route of
administration. Pulmonary drug delivery has the potential
to overcome many obstacles of the oral route of
administration and to treat lung specific diseases locally
or to absorb molecules for systemic delivery. This is
especially important for the chemotherapeutic treatment
of lung cancers. Advances in dry powder inhalers and the
development of suitable carriers for nano-medical drug
delivery systems enable the application of nano-medical
treatment strategies to the pulmonary route of
administration.
The talk will show the strategies which were used for the
development of inhalable nanoparticles. One aspect was
to add an active drug release mechanism to the carrier
particles to improve the dispersion of nanoparticles when
the carrier dissolves. Furthermore, nano-medical
strategies have to consider nanotoxicological aspects of
any nano-based delivery system and the lung surfaces.
Any interaction between the nano-medical device and the
lungs has to be carefully assessed. Strategies to evaluate
nanotoxicological aspects between nanoparticles and the
lungs surfactants will be discussed. The cytotoxicity and
cellular uptake of doxorubicin loaded nanoparticles using
different lung cancer cell lines was investigated and will
be discussed. The talk will show preliminary in vivo data
of the toxicity of inhalable nanoparticles using an in vivo
mouse model.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
9B.4
Targeted Delivery of High Aspect Ratio Particles in Small
Airway Bifurcations. ANDREW R. MARTIN, Warren H.
Finlay, University of Alberta.
The deposition of elongated, high aspect ratio particles in
the respiratory tract has been studied at length owing to
the health risk posed by inhaled mineral and synthetic
fibers. These particles have smaller aerodynamic
diameters than do compact particles of equivalent mass,
and as such are better able to penetrate the upper airways
and reach the lung. However, in the peripheral regions of
the lung, where airway diameters are small, the
interception mechanism can enhance deposition
efficiencies for elongated particles above those expected
for mass-equivalent compact particles. These
considerations combine to make elongated particles
candidates for broadly targeted aerosol drug delivery to
the peripheral lung. More speculatively, the ability to
noninvasively control the deposition of elongated
particles in order to increase doses received at specific
sites within the lung would allow for localized targeting
to those sites.
With these goals in mind, we have recently been
investigating the deposition of elongated drug particles in
small, bifurcating airways. Deposition efficiencies have
been measured in physical airway models, and
comparison made to mathematical models predicting
deposition of such particles due to impaction,
sedimentation, diffusion, and interception. In addition,
initial in vitro experiments demonstrating that
noninvasive external control over particle deposition can
be achieved in small airway bifurcations will be
presented.
Acknowledgements: the financial support of the Natural
Sciences and Engineering Research Council of Canada,
the Alberta Ingenuity Fund, and the Killam Fund is
greatly appreciated.
SYMPOSIUM: Innovation in Medicinal Nanopoarticles
9B.5
Leucine Shells on Spray-dried Medicinal Microparticles.
Christopher I. Grainger; King
Microparticles for pulmonary or nasal delivery must have
properties that provide physical and chemical stability of
the dosage form and lead to adequate powder and aerosol
behavior. Particles must be designed to possess correct
aerodynamic diameters and adequate dispersibility to
facilitate device emptying and delivery to the intended
target.
Leucine has been successfully used as a dispersibility
enhancer, both as a particulate additive and as an
excipient in a homogeneous powder when manufactured
by spray-drying. Microparticles with leucine shells have
been developed for therapeutic and vaccination purposes.
However, it has not been described how leucine shells
form on evaporating droplets and how such particles can
be designed.
This Paper presents experimental and theoretical work
describing the particle formation process of leucinecontaining microparticles made by spray-drying. It
presents characteristic times describing the shell
formation process in a multi-component droplet. These
characteristic times can then be used to select appropriate
process and formulation variables to achieve the desired
particle morphology.
The experimental tool used in this study was a modified
bench-top spray-drier. The atomizer of the dryer was
replaced with an aerosol generator to allow drying of
monodisperse droplets of known diameter. The
aerodynamic diameter of the dry particles was measured
in-situ using a time-of-flight technique. Particles
consisting of immunoglobulin and leucine in various
ratios were dried under different conditions that lead to
particles with a leucine shell, an immunoglobulin shell, or
a shell of mixed composition.
The results are explained in the context of a simplified
analytical model based on a steady-state evaporation
approximation; and particle engineering guidelines for the
design of leucine shells in general are derived.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
SYMPOSIUM: Innovation in Medicinal Nanopoarticles
9B.6
Drying Behavior of Polymer Solution Droplets during the
Production of Microparticles for Sustained Drug Release.
WILLARD R. FOSS, Amgen, Inc.
Active pharmaceutical ingredients can be encapsulated
into microparticles with the biodegradable polymer poly
(lactide-co-glycolide) (PLGA) for sustained release
delivery by parenteral administration. The microparticles
must be small enough to pass through a moderate gauge
hypodermic needle but large enough to minimize initial
burst release of drug caused by high particle surface area.
The microparticles should have a size range between
about 10 and 100 microns.
Particles of this size can be effectively produced by spray
drying of PLGA solutions with dissolved or suspended
drug. High volatility solvents are chosen to speed the
evaporation. However, polymer solution droplets form a
low permeability skin when dried rapidly, trapping
solvent in the particle core and resulting in an undesirable
hollow morphology. Subsequent evaporation is limited
by the diffusion of solvent through the dry skin.
In this work, we assess the drying kinetics of polymer
solution droplets to predict behavior in spray dryers, using
PLGA as an example. A combination of experimentation
and modeling is used. Dilute polymer solution droplets
dry according to the diameter-squared law during the
early stages of drying. In this region, drying kinetics are
accurately modeled with analytical solutions of pure
solvent droplet evaporation and confirmed with
measurements in a laminar flow drying column.
At the latter stages of drying, the polymer concentrates
near the surface of the droplet due to high Peclet number
effects. The development and progression of the skin is
modeled with accurate concentration and temperaturedependent polymer diffusion data. Measurement of the
polymer-solvent diffusion coefficients is discussed as well
as their application to modeling of polymer skin
formation within a single droplet. The effects of gas
temperature and vapor composition on the skin formation
and final particle morphology are reviewed and
experimental confirmation of the model is shown.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.1
Speciation of Ultrafine Particulate Matter Formed via
Ozonolysis of Household Volatile Organic Compounds.
KARA HUFF HARTZ, Meagan Hatfield, and Hardik Amin,
Southern Illinois University.
Laboratory-generated secondary organic aerosol (SOA) is
typically created by exposing a single organic precursor to
an oxidant. However, in the atmosphere, oxidation occurs
in the presence thousands of other organic compounds. In
this research, SOA is generated by the ozonolysis reaction
chamber by oxidizing household air fresheners, which act
as a surrogate for realistic organic precursor mixtures.
SOA is generated in a 5.5 cubic meter Teflon smog by
exposing volatilized air fresheners to ozone. The selected
air fresheners contain monoterpenes, which oxidize
forming less volatile products that partition into the
condensed phase. The aerosol mass concentration is
measured by scanning mobility particle size spectrometry.
The aerosol mass fraction is calculated by dividing the
wall-loss corrected aerosol mass concentration by the
concentration of monoterpenes introduced into the
chamber (determined by gas chromatography). Samples
of particulate matter are removed from the chamber by
filters and the filters are solvent extracted. The extracts
are concentrated, the analytes are converted to their
trimethyl silyl ester and ether products, and the products
are analyzed by gas chromatography with mass
spectrometric detection. The aerosol products are
identified by analysis of their characteristic ion fragments
and determined using authentic (when available) and
surrogate standards. The product distributions and
aerosol mass concentrations are compared to SOA formed
from the ozonolysis of a single monoterpene precursor
under similar conditions. Because the organic precursors
used are mixtures instead of single species, the results of
these experiments provide an SOA species distribution
that more closely mimics the complexity that is found in
atmosphere.
2E.2
CMAQ predictions of in-cloud secondary organic aerosol
(SOA) in the Eastern U.S. ANNMARIE G. CARLTON, Rohit
Mathur, Shawn J. Roselle, National Oceanic and Atmospheric
Administration (In partnership with the U.S. Environmental
Protection Agency).
Mounting experimental evidence suggests secondary
organic aerosol (SOA) is formed in the atmosphere
through aqueous phase reactions in clouds. An in-cloud
SOA parameterization developed from laboratory
experiments is applied to a regional-scale chemical
transport model for the first time using the U.S.
Environmental Protection Agency's (EPA) Community
Multiscale Air Quality (CMAQ) model. Initial
simulations of the Eastern U.S. using CMAQ suggest
substantial in-cloud SOA production from glyoxal and
methylglyoxal, common atmospheric constituents
primarily formed in the atmosphere during the oxidation
of biogenic and anthropogenic precursors. Air quality
implications are discussed regarding increased PM2.5
mass concentrations in the surface layer (0 - 33 m). Long
range pollution transport and potential climate effects are
discussed relative to increased PM concentration
predictions for the free troposphere (FT) (~ 100 mb).
Predicted aerosol concentrations are compared with
ground-based observations from the speciation trends
network (STN) and with aerosol measurements made
aloft during the International Consortium for Atmospheric
Research on Transport and Transformation (ICARTT)
study. Increased aerosol concentrations suggest that and
in-cloud SOA formation mechanism provides one
possible explanation for the under prediction of organic
aerosol by CMAQ.
The research presented here was performed, in part, under
the Memorandum of Understanding between the U.S.
Environmental Protection Agency (EPA) and the U.S.
Department of Commmerce's National Oceanic and
Atmospheric Administration (NOAA) and under
agreement number DW13921548. This work constitutes
a contribution to the NOAA Air Quality Program.
Although it has been reviewed by EPA and NOAA and
approved for publication, it does not necessarily reflect
their policies or views.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.3
Formaldehyde and Glyoxal in Ambient Particulate Matter:
A Discussion on Their Chemical Identities. JIAN ZHEN YU,
Ho Sai Simon Ip, Xiaohui Hilda Huang, Hong Kong University
of Science & Technology.
2E.4
Efficient SOA Formation from Heterogeneous Oxidation of
Organic Surfaces by OH Radicals. KEVIN R. WILSON,
Jared D. Smith, Musahid Ahmed, Stephen R. Leone, Erin
Mysak, Lawrence Berkeley National Laboratory.
Formaldehyde and glyoxal are among the more abundant
carbonyl species in the ambient environment as a result of
common emission sources as well as being an
atmospheric oxidation product of many anthropogenic
and biogenic hydrocarbon precursors. Despite their high
volatility, they have been measured to exist in ambient
particulate matter at a concentration ranging from a few
-3
tenths to tens ng m . The level of detected formaldehyde
and glyoxal in the aerosol phase far exceeds what is
anticipated based on the vapor pressure of these
compounds in their free form. The detection of aerosol
phase formaldehyde and glyoxal has invariably involved
the use of water as the extraction solvent and subsequent
derivatization with a hydrazine or a hydroxylamine before
GC or HPLC analysis. We here present evidence to
indicate that the sulfate ester of the hydrated
formaldehyde and both the sulfate and the sulfite ester of
glyoxal could undergo hydrolysis and be detected as if
they were free formaldehyde and glyoxal. Model
simulation of the formation of the sulfate or sulfite esters
in clouds followed by evaporation of cloud water could
explain a significant fraction of detected formaldehyde
and glyoxal in the aerosol phase. The existence of
formaldehyde and glyoxal in their sulfate/sulfite form
resolves the myth of their unexpected presence in particle
phase.
Currently, there is much interest in the formation rates
and mechanisms of secondary organic aerosols (SOA)
from ozone reactions with both biogenic and
anthropogenic precursors. However, with the exception
of isoprene (1), little work has been done to understand
SOA formation from OH radical reactions with other
volatile organic compounds. Using a coated flow tube
reactor, rapid secondary organic aerosol (SOA) formation
is observed when an organic film (such as stearic acid) is
exposed to OH radicals. In addition to films, we have also
observed that OH oxidation of submicron organic
particles also leads to similar SOA formation. These
results suggest an entirely new, and very efficient,
formation mechanism of SOA via OH radical oxidation of
organic surfaces. Analysis of these SOA particles, via
VUV photoionization mass spectrometry, suggests that
these particles are chemically complex and perhaps
oligomeric in nature. We suggest a potential mechanism
for this process in which gas phase products, such as
semi-volatile aldehydes and carboxylic acids, evolve from
the oxidation of the organic films. Subsequent reactions
of these volatile products with OH in the gas phase results
in efficient SOA formation. This mechanism is supported
by the observation that OH radical reactions with gas
phase hexanal and nonanal leads to strong SOA
formation. These results provide a direct link between
volatile organic compounds produced by particle
oxidation and SOA formation.
References
1. M. Claeys et al., Science 303, 1173 (2004).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.5
Heterogeneous processing of organic carbonyls on
submicron aerosol particles. ALEXEI KHALIZOV, Huaxin
Xue, Jun Zhao, Renyi Zhang, Texas A&M University.
The aerosol loading of the troposphere has increased
continuously from pre-industrial times to the present day.
Elevated concentrations of aerosols have several distinct
impacts, including effects on human health and the Earth's
energy balance. A major difficulty in assessing these
impacts arises from inadequate knowledge of the
formation and growth of secondary organic aerosols
(SOA). Low- and semi-volatile organic compounds
(SVOC) produced upon photo-oxidation of anthropogenic
and biogenic volatile organics have been shown to
contribute to the formation of SOA in the atmosphere.
However, the chemical and physical processes leading to
SOA formation and growth are still poorly understood.
One of the major uncertainties is due to a very limited
knowledge about the chemical reactions taking place
within aerosol particles. These reactions, which may be
acid-catalyzed, have been shown to produce compounds
of higher molecular weight and correspondingly lower
vapor pressure, increasing the partitioning of SVOC to
condensed phase. When these reactions are neglected, the
contribution of semi- and intermediate-volatility organic
compounds to SOA formation is significantly
underestimated. We report the laboratory measurements
of heterogeneous reactions between semi- and
intermediate volatility organic compounds and preexisting aerosol particles. Monodisperse aerosols of
different initial chemical composition and acidity were
exposed to ppb-to-ppm (part per million to part per
billion) concentrations of several organic carbonyls and
di-carbonyls. Aerosol growth and heterogeneous
processing were quantified by concurrent measurements
of changes in the particle size, the particle chemical
composition, and the gaseous organic precursor
concentrations. The implication of our results on the
growth and composition of secondary organic aerosols
under different atmospheric conditions will be discussed.
2E.6
First-Order Sensitivity and Uncertainty Analysis of the
MAGIC Model Using NaCl aerosols. PAUL NISSENSON,
Jennie Thomas, Barbara Finlayson-Pitts, Donald Dabdub,
University of California, Irvine.
A first-order sensitivity and uncertainty analysis is
conducted on the Model of Aqueous, Gaseous and
Interfacial Chemistry (MAGIC). Uncertainty ranges are
established for 197 input parameters and are used in
conjunction with latin hypercube sampling and multiple
linear regression to determine (1) the correlation between
each input parameter and the model output and (2) the
contribution of each input parameter to the uncertainty in
the model output. In this study, the output parameter of
interest is the peak concentration of molecular chlorine,
[Cl2(g)]peak, since MAGIC has previously demonstrated
the importance of an interfacial reaction between OH(g)
and Cl-(aq,surf) in the production of Cl2(g). The
sensitivity analysis reveals that the interface reaction rate
is the input parameter most strongly correlated with [Cl2
(g)]peak. The uncertainty analysis shows that the interface
reaction rate also is responsible for most of the
uncertainty in MAGIC's ability to precisely calculate [Cl2
(g)]peak. Results from this study establish priorities for
refining estimates of input parameters.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.7
Modeling Secondary Organic Aerosol from the Ozonolysis of
Monoterpenes in the Presence of Inorganic Aerosols.
NORTHCROSS AMANDA, Jang Myoseon, University of
North Carolina.
2E.8
Thermodynamic Modeling of Atmospheric Inorganic
Aerosols. ANDREY MARTYNENKO, Fang-Yi Cheng, Jiwen
W. He, University of Houston; John H. Seinfeld, California
Institute of Technology.
Atmospheric aerosol models traditionally have
separated modules for organic and inorganic aerosols and
assume that aerosol is additive for each compartment.
However this simplified approach does not consider the
interaction of the organic and inorganic species on the
aerosol system, or the additional mass introduced by the
aerosol phase reactons. Previous studies have suggested
that inorganic acid such as sulfuric acid has been shown
to catalyze heterogeneous reactions of the organic
components in atmospheric aerosols increasing aerosol
mass. Organic compounds can also react with sulfuric
acid to form organic sulfates.
This study demonstrates an aerosol model which
includes the influence of inorganic and organic
compounds interacting within an aerosol. The new model
approach was evaluated using secondary organic aerosols
(SOA) created from the ozonolysis of a series of
monoterpenes in the presence of inorganic aerosols. The
model combined three individual models: a gas phase
model using the master chemical mechanism (MCM)1; a
heterogeneous aerosol model2; and a modified version of
the SOGRAM model3 for the organic aerosol mass
formed solely due to thermodynamic partitioning. The
near explicit product distribution from the modeled gas
phase ozonolysis reactions are lumped based on vapor
pressure and their ability to participate in particle phase
reactions. The lump species are then partitioned to SOA
using the partitioning model and the heterogeneous
aerosol mass is estimated based on the heterogeneous
reaction rate constant for each lumped group. This model
is able to dynamically track SOA mass formed in indoor
Teflon chamber.
A variety of thermodynamic models have been developed
to predict inorganic gas-aerosol equilibrium. The models
can be distinguished based on three general features:
(1) the method of computing activity coefficients of the
aerosol-phase species,
(2) the method of computing the aerosol water content,
and
(3) the numerical technique that is used to determine the
equilibrium state.
1. Jang, M.; Czoschke, N.M.; Northcross, A.L.; Cao, G.;
Shaof, D. (2006) SOA formation from partitioning and
heterogeneous reactions: Model study in the presence of
inorganic species. Environ. Sci. Technol. 40, 3013-3022.
2. Saunders, SM., Jenkin, ME, Derwent, RG, and Pilling,
MJ (2003). Protocol for the development of the master
chemical mechanism, MCM v3 (part A): tropospheric
degradation of non-aromatic volatile organic compounds.
Atmos. Chem.& Phys. 3: 161-80.
3. Schell, B, Ackermann, IJ, Hass, H, Binkowski, FS, and
Ebel, A. (2001). Modeling the formation of secondary
organic aerosol within a comprehensive air quality model
system. J. of Geophys. Res. 106: 28275-28.
In this talk, the current state of the art of thermodynamic
equilibrium models is reviewed, and a new inorganic
atmospheric aerosol phase equilibrium model (UHAERO)
is presented. UHAERO incorporates two mole fraction
based multicomponent activity coefficient models,
namely the PSC an the ExUNIQUAC model. UHAEROPSC is benchmarked against predictions obtained with the
Aerosol Inorganic Model (AIM) and proved to be
identical in terms of accuracy over the entire range of
compositions and relative humidity for the sulfate/nitrate/
ammonium/water system.
Based on the widespread application of ISSOROPIA, the
examination of the model performance of UHAERO
against that of ISSOROPIA over an extended
composition, temperature, and RH domain is detailed. It
is observed that the overall speed of UHAERO is
comparable to that of ISSOROPIA. The accuracy in the
prediction of thermodynamic properties of electrolyte
solutions such as pH value is quantified. The potential
error in using the ZSR mixing rule to estimate the aerosol
water content at low RH is assessed. The model success
or failure to accurately predict deliquescence point
depression in the multi-phase aerosol growth is compared.
At the end, the performance of UHAERO modules in
conjunction with chemical transport models such as
CMAQ is assessed for a PM episode. The issues related to
the effect of organic compounds on aerosol
hygroscopicity as well as the dynamic approach vs the
equilibrium approach for predicting the mass transfer of
semi-volatile species between the gas and aerosol phases
are addressed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.9
Understanding the Chemical Interactions between Gases
and Aerosols. CHAO WEI, Geogery R. Carmichael, University
of Iowa.
Chemical interactions between gases and particles are
important to regional chemical transport models (CTMs)
because they can influence photochemical oxidant cycle
and regional distribution of aerosols. Current CTMs have
limitations to deal with the chemical interactions. A new
approach to solve the growth equation of aerosol due to
condensation/evaporation process is introduced and
verified in a box-model study. It can get accurate results
with less size sections, which will improve the
computational efficiency for 3-D models. The effects of
the complex of heterogeneous chemistry on atmospheric
chemistry are also investigated in a box model study. For
example, both saturation effect and coating on aerosol can
influence the heterogeneous uptakes of gases on aerosols
in certain conditions. A new aerosol model is developed
by incorporating the important results in our studies. By
using data from field observations the new aerosol model
is evaluated in a back trajectory study.
2E.10
Role of Cloud Processing in Organic Acid Aerosol
Formation: A Review of Field Measurements. ARMIN
SOROOSHIAN, Miao-Ling Lu, Fred J. Brechtel, Richard C.
Flagan, John H. Seinfeld, California Institute of Technology;
Graham Feingold, NOAA; Haflidi Jonsson, Naval Postgraduate
School.
Organic acids are ubiquitous in atmospheric aerosols.
The presence of organic acids in aerosols may alter
hygroscopic behavior and consequently, the radiative
impact of particles. There is growing evidence, based on
laboratory and ambient measurements, that secondary
organic aerosol (SOA), similar to sulfate, is formed by
aqueous-phase reactions. Due to the water-soluble nature
of organic acids, these species constitute a significant
fraction of SOA mass. Airborne particle-into-liquid
sampler (PILS) measurements made on the Center for
Interdisciplinary Remotely-Piloted Aircraft Studies
(CIRPAS) Twin Otter are presented from three separate
field campaigns representing urban and marine
atmospheres: International Consortium for Atmospheric
Research on Transport and Transformation (ICARTT
2004), Marine Stratus/Stratocumulus Experiment (MASE
2005), Gulf of Mexico Atmospheric Composition and
Climate Study (GoMACCS 2006). A strong correlation
exists between sulfate and oxalic acid (most abundant
dicarboxylic acid) in ambient aerosols, including
evaporated droplet residual particles in clouds, suggesting
that cloud processing is a key formation mechanism for
particulate oxalic acid. Predictions from a chemical cloud
parcel model considering the aqueous-phase production of
dicarboxylic acids and sulfate show good agreement for
the relative magnitude of sulfate and oxalate growth for
clouds influenced by power plant plumes and those that
are not. Enhanced organic acid aerosol layers were
observed above cloudtops in both MASE and GoMACCS.
As derived from large eddy simulations of stratocumulus
under the conditions of MASE, both Lagrangian
trajectory analysis and diurnal cloudtop evolution provide
evidence that a significant fraction of the aerosol mass
concentration above cloud can be accounted for by
evaporated droplet residual particles. The evolution of
organic acids with increasing altitude in cloud provides
evidence for the multi-step nature of oxalic acid
production; model predictions are consistent with the
observed oxalate:glyoxylate ratio as a function of altitude
in GoMACCS cumuli.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.11
A Kinetic Study of the Heterogeneous Reaction of
Deliquesced NaCl Particles with gaseous HNO3. YONG
LIU, Pacific Northwest National Laboratory; Jeremy P. Cain,
Hai Wang, University of Southern California; Alexander Laskin,
Pacific Northwest National Laboratory.
The kinetics of heterogeneous reaction of deliquesced
sodium chloride particles (NaClaq ) with HNO3 gas and its
uptake on NaClaq were measured using a novel Particleon-Substrate Stagnation Flow Reactor (PS-SFR) approach
under conditions, including particle size (0.5-1.5 micrometer), relative humidity (RH=20-80%), nitric
concentration (2-20 ppb) and reaction time (5-720 min),
directly relevant to the atmospheric chemistry of sea salt
particles. The technique utilizes the exposure of substrate
deposited aerosol particles to reactive gases followed by
chemical analysis of the particles using computer
controlled scanning electron microscopy with energy
dispersive analysis of X-ray (CCSEM/EDX) capability.
Reactor design and experimental conditions were guided
by computational fluid dynamics calculation results to
ensure uniformity of the diffusion flux to all particles
undergoing reaction. As we discuss in this presentation,
fundamental reaction kinetics data can be obtained from
these experiments after a theoretical kinetic-diffusion
analysis of effects of gaseous reactant transport from the
bulk gas to the substrate surface. The apparent, pseudo
first order rate constants for the heterogeneous reaction
were obtained based on the changes in the chloride
concentration of individual particles upon reaction with
HNO3. The net reaction uptake coefficient was found to
be 0.11 with an uncertainty factor of 3. Additional
experiments examined the variations of HNO3 uptake on
pure NaCl, a sea salt-like mixture of NaCl and MgCl 2
(Mg-to-Cl molar ratio of 0.114) and real sea salt particles
as a function of relative humidity. The uptake of HNO3
on pure NaCl particles was also examined as a function of
particle size under a constant relative humidity of 80%.
Application of a resistance model of reaction kinetics and
reactant diffusion over a single particle suggests that, over
the range of particle size studied, the uptake is largely
controlled by gaseous reactant diffusion from the free
stream to the particle surface.
2E.12
A Kinetic Study of the Heterogeneous Reaction of CaCO3
Particles with gaseous HNO3. YONG LIU, Pacific Northwest
National Laboratory; Elizabeth R. Gibson, University of Iowa;
Jeremy P. Cain, University of Sourthern California; Vicki H.
Grassian, University of Iowa; Hai Wang, University of Southern
California; Alexander Laskin, Pacific Northwest National
Laboratory.
Mineral dust aerosol, a complex mixture of particles and particle
aggregates of varying composition and mineralogy, has the
largest aerosol burden in terms of mass in the atmosphere.
Recent studies have shown that CaCO3, a particular reactive
component, can be partially or even entirely converted to Ca
(NO3)2 during the processes of aging, reaction and
transportation via the following pathway: CaCO3(s) + 2HNO3
(g)=<Ca(NO3)2(s,aq) + CO2(g) + H2O. Its physiochemical
properties are altered by this reaction. In the present work, the
heterogeneous reaction kinetics of CaCO3(s) particles ( ~ 0.8
micro-meter) with HNO3 gas and its uptake on CaCO3(s) were
measured using a novel Particle-on-Substrate Stagnation Flow
Reactor (PS-SFR) approach under conditions, including relative
humidity (RH=10-80%), nitric acid concentration (6-25 ppb)
and reaction time (5-300 min), directly relevant to the
atmospheric chemistry of mineral dust particles. The technique
utilizes the exposure of substrate deposited aerosol particles to
reactive gases followed by chemical analysis of the particles
using computer-controlled scanning electron microscopy with
energy-dispersive analysis of X-ray (CCSEM/EDX) capability.
Reactor design and experimental conditions were guided by
computational fluid dynamics calculation results to ensure
uniformity of the diffusion flux to all particles undergoing
reaction. As we discuss in this presentation, fundamental
reaction kinetics data can be obtained from these experiments
after a theoretical kinetic-diffusion analysis of effects of gaseous
reactant transport from the bulk gas to the substrate surface.
Such effects arise from the close proximity of the reacting
particles mounted on the substrate, which may result in kinetic
rates being diffusion controlled. We show that these effects may
be quantified and kinetic rates can be obtained where the
transport limitations associated with the substrate experiments
are decoupled from the reaction kinetics. The apparent, pseudo
first-order rate constant for the heterogeneous reaction was
obtained based on the changes in the O/Ca ratio of individual
particles upon reaction with HNO3. The intrinsic, second-order
-15
rate constant was obtained as kII = 2.5 x 10 cm3 molecule-1 s
-1 in the limit of zero particle loading and by assuming that the
substrate is inert to HNO3. Net reaction uptake coefficient was
found to be 0.08 with an uncertainty factor of 3 at RH=40%, and
has a monotonic dependence with the increasing RH.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.13
Phase Sequence Law. Michael Anisimov, ANATOLIY
BAKLANOV, and Vladimir Akimov. Institute of Chemical
Kinetics and Combustion, Siberian Division of the Russian
Academy of Sciences. Novosibirsk, Russia.
More than century ago, Ostwald [1] manifested a rule,
which, according to his believing, is a general principle
for any process in nature. Ostwald's rule of stages gives
the sequence of different phase formation. The rule has
been tested several times during Ostwald's life and later.
The clear evidences of the out of order phase formation(s)
usually is explained by the fast (non-detected) kinetics of
the \true\ phase(s). That circumstance makes actual the
direct prove of the Ostwald's rule using a simple enough
model. The model needs to be applicable for the broad
spectrum of the barrier processes, which are joining by
that rule. Schmelzer et al. [2] generalized this rule of
stages for nucleation. Seemingly their formulation is
presenting the one more boundary case for the common
nucleation phenomena.
In the present research the dynamics of phase
transformations is considered for the case of vapor multichannel nucleation using the idea by Anisimov and Hopke
[3] for the semiempirical nucleation rate surface design.
General dynamics of multi-channel nucleation, associated
with these phase transitions, is derived. The current
generalization of the kinetic law for the formation of
phases is proved on the basis of common results on
nucleation [4, 5]. Current consideration reduces problem
immediately to simplified models of nucleation. The
present consideration uses the qualitative prove on the
base of several plausible assumptions.
2E.14
Influence of Aerosol Acidity on Secondary Organic Aerosol
Formation from Isoprene and Alpha-Pinene. JOHN H.
OFFENBERG, Michael Lewandowski, Tadeusz E. Kleindienst,
Edward O. Edney, US EPA / NERL; Mohammed Jaoui, Alion
Science and Technology; Jason D. Surratt, John H. Seinfeld,
California Institute of Technology.
The effect of particle-phase acidity on secondary organic
aerosol (SOA) formation from isoprene and alpha-pinene
were each investigated in the laboratory. Steady-state
SOA was generated by irradiating hydrocarbon / NOx
3
mixtures in a 14.5 m reaction chamber with the relative
humidity held constant at 30%. The acidity of the mixed
ammonium sulfate and sulfuric acid seed aerosol was
varied systematically for each precursor hydrocarbon. The
observed enhancement in secondary organic carbon
(SOC) concentration is highly correlated with increasing
aerosol acidity for both systems. Under the conditions of
these chamber studies, SOC increases by 0.34% per nmol
-3
-3
[H+] m for isoprene, and 0.04% per nmol [H+] m for
alpha-pinene. Aerosol mass concentrations for the 2methyltetrols, which serve as tracers for isoprene SOA in
ambient aerosols, increased significantly with enhanced
aerosol acidity.
Disclaimer: This work has been funded wholly or in part
by the United States Environmental Protection Agency
under Contract 68-D-00-206 to Alion Science and
Technology and through Cooperative Agreement CR
-831194001 to California Institute of Technology.
Although this work was reviewed by EPA and approved
for publication, it may not necessarily reflect official
Agency policy.
[1] Ostwald, W. Phys. Chem. Stoechiom.
Verwandtschaftsl. 22, 289 (1897).
[2] Schmelzer, J. W. P., Schmelzer, J., Jr., Gudzow, I. S.
J. Chem. Phys., 112(8), 3820 (2000).
[3] Anisimov, M.P. and Hopke, P.K. J. Phys. Chem. B. V.
105(47), 11817 (2001).
[4] Anisimov M. P. J. Aerosol Sci., 21(Suppl.1.), 23
(1990).
[5] Anisimov, M. P., Hopke, P. K., Rasmussen, et al. J.
Chem. Phys., 109(4), 1435 (1998)
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
2E.15
Laboratory Evidences of SOA Formation by Acid-Catalyzed
Heterogeneous Reactions of Toluene Oxidation Products.
GANG CAO, Myoseon Jang, The University of North Carolina
at Chapel Hill.
2E.16
Variation of Secondary Organic Aerosol Formation with
Temperature from Cyclohexene and alpha-Pinene
Ozonolysis. BETHANY WARREN, David R. Cocker III,
University of California-Riverside.
Secondary organic aerosol (SOA) was generated from oxidation
of toluene by OH radicals from photolysis of H2O2 in an indoor
Teflon film chamber under non acidic and acidic conditions. In
this study, we hypothesize that particle acidity influences
compositions and distribution of secondary organic products.
The gas and particle phase products were tentatively identified
by a gas chromatograph-ion trap mass spectrometer (GC-ITMS).
The peak areas of the organic products from the GC-ITMS were
normalized by the internal standards and used for the
comparison studies between non acidic and acidic conditions.
For the gas phase products in the presence of acidic aerosol,
more ring-retaining compounds and alcohols were found while
less multifunctional carbonyls were detected. This result
suggests that multifunctional carbonyls contribute SOA
formation through acid-catalyzed heterogeneous reactions
creating high molecular weight compounds in the particle phase.
As the oligomerization progresses via heterogeneous reactions
in the particle, the average molecular weight of SOA and the
activity coefficients of organic compounds increase, which
affects the partitioning coefficient of organic compounds. The
partitioning of organic compounds appears to be less favorable.
Thus, the relative concentrations of ring-retaining compounds
and alcohols in the gas phase are higher with acidic aerosol.
The changes of SOA were also tracked with the partitioning
behavior of deutrated alkanes (d40-nonadecane and d42eicosane). The partitioning coefficients of deutrated alkanes
were significantly higher with the neutral seed. FTIR was used
to observe the difference in SOA products generated at different
acidity of aerosols: neutral and acidic aerosols. The FTIR
spectra showed that compared to neutral aerosols, with acidic
aerosols, the C=O stretching band associated with conversion of
carbonyl compounds was significantly reduced, while the
intensity of C-O-C stretching band corresponding to
heterogeneous products (e.g., acetals and organic sulfates)
increased. In addition, a colorimetric analysis integrated with
the reflectance UV-Vis suggested the conversion of sulfuric acid
into organic sulfate. The laboratory analyses of toluene SOA
using GC-ITMS, FTIR, partitioning of deutrated alkanes, and
colorimetry support our hypothesis that acid-catalyzed
heterogeneous reactions significantly influence yields and
chemical properties of SOA.
Temperature is a key parameter for secondary organic
aerosol (SOA) formation. Up until the past few years,
most environmental chamber experiments were conducted
either indoors at fixed temperature or outdoors with
diurnal temperature patterns. Recently presented
temperature data by Takekawa et al. (2001) and Pathak et
al. (2007) has shown that temperature will effect the
amount of SOA formed, favoring the cooler temperatures
for ozonolysis and photo-oxidation systems.
This work utilizes the UC Riverside/CE-CERT
environmental chamber facility, which was designed for
study of SOA formation under well controlled
environmental conditions. Aerosol formation potential
for the cyclohexene/ozone system was obtained at fixed
temperatures of 278K, 300K, and 318K. A near ten fold
increase in SOA formation is noted at 278K compared
with 318K. The change in aerosol production could not
be wholly described by vapor pressure dependence on
temperature; therefore, additional experiments were
performed by cycling temperatures within the three set
parametres after the maximum SOA mass had been
reached. During these cycles the original aerosol mass
was never fully recovered, showing that there is a
temperature path dependence for this SOA system. This is
indicative of a kinetic effect in the gas and/or particle
phase chemistry for this system. By combining the
thermodynamic and kinetic theory and experimental
effects, a model will be used to explain this system.
Additional alpha-pinene experiments were performed at
the given temperature set points along with temperature
cycling. SOA is found to have a fourfold increase from
318K to 278K and the temperature cycling produced
similar results where the initial SOA mass could not be
recovered. These experiments are directly compared with
the Pathak et al. data set presented in 2006. The same
thermodynamic and kinetic model can also be used to
describe this system.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
3E.1
Secondary Organic Aerosol Formation from Photochemical
Transformations of Modern Diesel Vehicle Emissions.
BARBARA ZIELINSKA, Shar Samy, Desert Research Institute;
Jacob McDonald, Jean-Clare Seagrave, Lovelace Respiratory
Research Institute; Monica Vazquez, Klaus Wirtz, Fundacion
Centro de Estudios Ambientales del Mediterraneo.
The overall objective of this study was to investigate
photochemical transformations of diesel emissions in the
atmosphere. The specific aims were: (1) to characterize
the gas- and particle- phase products of atmospheric
transformations of diesel emissions; and (2) to explore the
changes in biological activity of diesel exhaust before and
after the atmospheric transformations take place. The
project was executed with the aid of the EUPHORE
simulation chamber in Valencia, Spain, which is currently
one of the largest and the best equipped outdoor
simulation chamber in the world, allowing investigation
of atmospheric transformation processes under realistic
ambient conditions. Diesel exhaust was generated on-site
using a light-duty modern diesel engine and a
dynamometer, equipped with Horiba continuous gas
analyzer. The experiments were carried out in January
2005 (winter campaign) and in May 2005 and 2006
(summer campaigns). The test matrix examined the
effects of aging and NO3 radical reactions in the dark and
photooxidation and OH radical reactions in the light on
the composition of diesel exhaust. Particle size, number
and volume concentrations were monitored with Scanning
Mobility Particle Sizer (SMPS) and particle mass was
continuously monitored using TEOM. NOx and NOy
species were monitored using chemiluminescence and
open-path Fourier Transfer Infrared (FTIR) instruments.
The excess NOx was removed from the diesel exhaust by
using a NOx denuder. The semi-volatile and particleassociated organic compounds were collected from the
chamber at the end of the exposures, using an XAD
coated annular denuder followed by a filter and XAD
cartridge. Since the diesel engine was equipped with an
oxidation catalyst and emissions of volatile hydrocarbons
are low, we investigated the effect of adding the aromatic
VOC, such as toluene to diesel exhaust under low NOx
conditions (~20 and 100 ppb). These experiments resulted
in a substantial secondary organic aerosol formation.
3E.2
Formation of Secondary Organic Aerosol from Reactions of
Cyclic and Branched Alkanes with OH Radicals in the
Presence of NOx. YONG B. LIM, Paul J. Ziemann, University
of California, Riverside.
Alkanes are the largest anthropogenic source of
hydrocarbons to the atmosphere, and consist of hundreds
of compounds with linear, branched, and cyclic structures
and a large range of molecular weights. Their atmospheric
oxidation reactions are initiated almost exclusively by OH
radicals, and in the presence of NOx can lead to a variety
of products, some of which can form secondary organic
aerosol (SOA). It has recently been suggested, for
instance [Robinson et al., Science, 315, 1259 (2007)], that
intermediate volatility alkanes may be a large, previously
underestimated source of SOA. In addition to being
atmospherically relevant compounds for study, alkanes
provide an ideal system for exploring basic chemical
reaction mechanisms involved in SOA formation (i.e.,
hydrogen abstraction, isomerization, and decomposition)
that apply to many hydrocarbon reactions. We previously
reported results of a study of SOA formation from
reactions of linear alkanes, and here we extend that work
to reactions of cyclic and branched alkanes with OH
radicals in the presence of NOx. Reactions were
performed in an ~6000 liter environmental chamber using
methyl nitrite/NO photolysis to generate OH radicals.
Particle composition was analyzed using a thermal
desorption particle beam mass spectrometer, and SOA
yields were measured using an SMPS for particle mass
and a GC-FID for alkane analysis. Reaction mechanisms
of cyclic alkanes appear to be similar to those of linear
alkanes, leading to SOA products including alkyl nitrates,
hydroxynitrates, cyclic hemiacetals, and substituted
tetrahydrofurans. Reactions of branched alkanes are
sensitive to alkane molecular structure, such that
pathways leading to condensable or volatile products can
be restricted or enhanced. SOA yields for cyclic alkanes
were higher than those for linear alkanes with the same
carbon number, while yields from branched alkanes were
substantially lower. This was apparently due to an
enhancement of alkoxy radical decomposition pathways
with chain branching.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
3E.3
Predicting Secondary Organic Aerosol Formation from
Aromatics: m-Xylene Case Study. BETHANY WARREN,
David R. Cocker III, University of California-Riverside and CECERT, University of California-Riverside; Chen Song, currently
at Pacific National Laboratories.
Aromatic hydrocarbons are complex and important
precursors for secondary organic aerosol (SOA)
formation. Recent studies have highlighted the
importance of environmental conditions on the total
amount of SOA formation, such as HC:NOx ratios (e.g.,
Johnson et al (2005), Song et al. (2005)). m-Xylene/NOx
photo-oxidation experiments were conducted under a
wide variety of environmental conditions using the UC
Riverside/CE-CERT environmental chamber. The
experimental parameters tested include light intensity,
light source (argon-arc vs. black lights), HC:NOx ratio
including ultra-low NOx experiments and no NOx
experiments, and the addition of non-aerosol and aerosol
forming hydrocarbons.
Using a data set of over 100 separate m-xylene photooxidation experiments, several key parameters of SOA
formation are noted. For example, increases in aerosol
formation were noted for decreasing NOx conditions and
increasing light intensity for the same amount of reacted
hydrocarbon. Addition of CO, propene and other light
hydrocarbons significantly altered the aerosol formation
in the m-xylene/NOx photo-oxidation system. Addition
of aerosol forming precursors (such as alpha-pinene) was
proven to be a non-additive process. The impact of these
variables on SOA formation is explained using several
key chemical pathways added to the APRC-99 gas-phase
chemical mechanism.
3E.4
Products and Mechanism of Secondary Organic Aerosol
Formation from the Reaction of OH Radicals with Linear
Alkenes. AIKO MATSUNAGA, Paul Ziemann, University of
California, Riverside.
Secondary organic aerosol (SOA) is formed by the
oxidation of volatile organic compounds by OH radicals,
NO3 radicals, or O 3. Alkenes, such as monoterpenes,
comprise a significant fraction of the atmospheric
hydrocarbon burden, and a large number of studies have
focused on SOA formation from their reactions with O3.
Much less is known about alkene reactions with OH
radicals, which is the major atmospheric oxidant. In this
study, we investigated the products and mechanisms of
the reactions of C8-C17 terminal linear alkenes with OH
radicals in the presence of NOx. The study was carried in
a ~6000 liter environmental chamber equipped with UV
lights, and OH radicals were generated by photolysis of a
methyl nitrite/NO mixture. Particle composition was
analyzed using a thermal desorption particle beam mass
spectrometer (TDPBMS) and SOA yields were
determined from SMPS measurements of SOA mass and
GC-FID measurements of the alkene. The TDPBMS and
an HPLC with an UV detector were both used to quantify
multifunctional nitrate products. The major SOA products
of the reactions are hydroxynitrates, dihydroxynitrates,
cyclic hemiacetals, dihydrofurans, and dihydroxyaldehyde
dimers. The cyclic hemiacetals are formed by
isomerization of dihydroxycarbonyls that do not
otherwise form aerosol. They can then dehydrate to
dihydrofurans. When corrected for gas-particle
partitioning, total molar yields measured for
hydroxynitrates and dihydroxynitrates were ~9% and
~10%, respectively. SOA yields increased from ~3% for
C8 up to a plateau of ~50% at C11. Increases in
contributions of hydroxynitrates and dihydroxynitrates
with carbon number due to enhanced gas-to-particle
partitioning accounted for ~50% of this increase in SOA
mass, while the remainder was due to enhanced dimer
formation. The results of this study add substantial new
insights into the products and mechanisms of alkene-OH
reactions and SOA formation, including the role of
oligomerization, which should be applicable to more
complex systems.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
3E.5
Organic Tracers Formed Under Acidic Conditions from
Isoprene Photooxidation. M. JAOUI, Alion Science and
Technology; T.E. Kleindienst, J.H. Offenberg, M.
Lewandowski, E.O. Edney; National Exposure Research
Laboratory, U.S. Environmental Protection Agency.
3E.6
Evaluating the Effects of Gas-Particle Partitioning and
Aging of Primary Organic Emissions using the Chemical
Transport Model PMCAMx. MANISH K. SHRIVASTAVA,
Timothy E. Lane, Neil M. Donahue, Spyros N. Pandis, Allen L.
Robinson, Carnegie Mellon University.
Recent studies reveal that isoprene, one of the most
strongly emitted biogenic hydrocarbon into the
atmosphere, lead to the formation of ambient secondary
organic aerosol (SOA). Particle acidity has been also
linked to SOA enhancement based on experiments
involving the photooxidation of isoprene. To investigate
the formation of organic tracers formed under acidic
conditions, a series of chamber experiments were
conducted in a 14.5 m3 smog chamber operated in
dynamic mode, in which isoprene/NOx mixtures were
irradiated either in the presence of SO2 or acidic seed
aerosol. In addition, PM2.5 samples collected during 2003
in Research Triangle Park, North Carolina were analyzed
in order to determine the contributions to ambient aerosol
of isoprene products formed under acidic conditions.
Chamber SOA and PM2.5 samples were derivatized using
bis(trimethylsilyl)trifluoro acetamide, then analyzed with
GC-MS.
GC-MS analysis of smog chamber experiments show the
presence of sulfuric acid and two newly detected organic
compounds with MW 253, and 313 Da (BSTFA
derivative) formed only under acidic conditions. These
compounds were detected also at high concentrations in
ambient PM2.5, indicating the impact of acidity on the
regional aerosol burden at least in the southeastern United
States. These organic compounds could serve as tracer
compounds for isoprene under acidic conditions.
This talk describes the effects of gas-particle partitioning
and photochemical aging of primary emissions on organic
aerosol concentrations in the Eastern United States using
a research version of the Chemical Transport Model
PMCAMx. The model treats primary organic aerosol
(POA) emissions using the basis-set approach, which
represents the POA emissions using a set of volatility bins
that span a basis set of effective saturation concentrations.
PMCAMx then calculates the amount of POA using the
simulated concentrations of the condensable emissions
and partitioning theory, thus the amount of POA varies
with atmospheric conditions. The effects of
photochemical aging are represented using gas-phase
reactions with the OH radical, which reduce the effective
saturation concentration of gas-phase organics causing a
larger fraction of organics to partition to the particle
phase. Inputs for the model are based on laboratory data.
The volatility distribution of the POA has been derived
from gas-particle portioning data for diesel exhaust and
woodsmoke. The SOA created by the aging mechanism
is consistent with experimental data of photochemical
aging of diesel exhaust in the smog chamber.
Disclaimer: This work has been funded wholly or in part
by the United States Environmental Protection Agency
under Contract 68-D00-206 to Alion Science and
Technology. Although this work was reviewed by EPA
and approved for publication, it may not necessarily
reflect official Agency policy.
Simulations were performed for July 2001 and January
2002. Comparison of the new results with those from the
traditional version of PMCAMx shows a dramatically
altered picture for ambient organic aerosol. Much of the
primary organic aerosol emissions evaporate, reducing
concentrations in urban areas. Photochemical aging of
these emissions creates substantial, regional SOA,
consistent with recent field measurements indicating SOA
dominance even in urban areas. Model predictions are
compared with data from the STN and IMPROVE
networks. We also evaluate the sensitivity of the model
predictions to the volatility distribution of emissions,
reaction rates with OH, reduction of volatility due to
aging, and the emissions of intermediate volatility
compounds.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
5E.1
Photochemical Aging of Organic Aerosol Particles. JIAHUA XING, Adam P. Bateman, Stephen A. Mang, Sergey A.
Nizkorodov, University of California Irvine.
5E.2
Changes in condensed-phase reactivity of organic
compounds with solvent composition. AMY M. SAGE, Neil
M. Donahue, Carnegie Mellon University.
Secondary Organic Aerosol (SOA) particles are produced
in the atmosphere as a result of oxidation of volatile
organic compounds (VOC). Primary Organic Aerosol
(POA) particles are directly emitted in the atmosphere by
their sources. Oxidation of VOC by NO3 and O3 plays an
important role in night-time generation of SOA and
chemical processing of POA. After the SOA and POA
particles are produced, they are further processed by daytime heterogeneous chemistry and by direct
photochemical processes. The goal of this study is to
understand the photochemical aging processes during
exposure of model SOA and POA particles to sunlight.
The photochemistry of model aerosol particles is
investigated with laboratory-based approaches relying on
cavity ring-down spectroscopy (CRDS). SOA particles
are generated by dark oxidation of Limonene and alphaPinene with NO3 and/or O3. Particles are then irradiated
with wavelength-tunable radiation in the actinic region
(<290 nm). The resulting gas-phase and particle-phase
photolysis products are studied by their CRDS spectra and
mass-spectrometry, respectively. Detail reaction
mechanisms and their implications for photochemical
aging of aerosol particles will be discussed.
Rates for heterogeneous and condensed-phase oxidation
of organic species are important inputs for both climate
and source-receptor models. However, the reactivity of
organic compounds in solution changes markedly with
solvent composition, so that rate constants measured in
single-component systems cannot easily be extrapolated
to atmospherically-relevant mixtures. And, ambient
particles present a nearly limitless array of solvent
combinations, the breadth of which we cannot begin to
cover with discrete laboratory measurements. Taken
together, these facts underline the necessity for
experiments that systematically explore trends in
reactivity, with results that are generalizable beyond the
reactants considered.
To this end, we have studied ozone-alkene chemistry in
particles dominated by both participating and nonparticipating solvents. We observe changes in effective
oxidation rate constants as a function of particle
composition and age for reactive species in several simple
mixtures. Even in initially pure oleic acid particles, the
observed reaction rate drops by a factor of two as the
particles' composition becomes a mixture dominated by
the ozonolysis products. By combining single phase and
mixed-phase relative rate constant measurements, we can
constrain the influence of confounding transport processes
in these particles, and quantify effects of reactant-solvent
interactions on condensed-phase reaction rates. We use a
compound-specific uptake coefficient to relate these rate
constants to gas-phase uptake measurements and quantify
secondary chemistry occurring in the particles.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
5E.3
Laboratory Investigation of Photochemical Oxidation of
Organic Molecular Markers used for Source
Apportionment. EMILY A WEITKAMP, Amy M. Sage,
Andrew T. Lambe, Neil M. Donahue, and Allen L. Robinson,
Carnegie Mellon University; Kara E. Huff Hartz, Southern
Illinois University.
Molecular markers are reduced organic compounds that
are used as tracers for primary sources of organic carbon
such as diesel engines and wood combustion. Current
receptor models assume that organic molecular markers
are chemically stable, that is, they do not react with
oxidants in the atmosphere at a rate that is relevant to the
lifetime of the particles. This paper presents smog
chamber measurements of condensed-phase oxidation of
organic molecular markers for meat cooking and motor
vehicle exhaust exposed to oxidants (ozone and OH)
under atmospherically relevant conditions in real aerosol
systems. The changes in composition of the aerosol were
determined by GC-MS of solvent extracted filter samples.
Kinetic rate information was then determined using a
relative rate approach. Several real and model aerosol
mixtures for meat cooking and motor vehicle exhaust
were studied and the effects of relative humidity and
secondary organic aerosol coatings were explored.
Results show that oxidation rates of molecular markers in
realistic aerosols (motor oil or hamburger grease) are fast
enough to be a concern for modelling areas with regional
transport. For example, our most conservative reaction
rate for cholesterol, a molecular marker for meat cooking
emissions, indicate a half-life of two days at modest
ambient ozone concentrations of 30ppb. Experiments also
reveal substantial oxidation of steranes in motor oil
exposed to typical summertime levels of OH. Data for
other molecular markers including palmitoleic acid, oleic
acid and hopanes will also be presented. Oxidation is
non-linear; therefore it represents a substantial
complication to linear source apportionment techniques
such as the Chemical Mass Balance model.
5E.4
Extremely Rapid Volatilization and Oligomer Formation via
OH Radical Initiated Oxidation of Organic Aerosols.
JARED D. SMITH, Erin Mysak, Stephen R. Leone, Musahid
Ahmed, and Kevin R. Wilson, Lawrence Berkeley National
Laboratory.
Ambient aerosols are known to play a significant role in a
variety of atmospheric processes such as direct and
indirect effects on radiative forcing. Chemical
composition can be an important factor in determining the
magnitude of these effects (optical density,
hygroscopicity, etc.) (1,2). However, a major fraction (80
- 90%) of organic aerosols can not be resolved on a
molecular level. Recent identification of high mass
oligomeric species as a major component in laboratory
and ambient organic aerosols has received much attention
due to the possibility that these species may account for
much of the unknown organic mass in ambient aerosols
(3, 4). Although, a few mechanisms have been proposed,
the origin and formation processes of these compounds
remain largely unknown. Using VUV photoionization
aerosol mass spectrometry we provide strong evidence for
a previously unidentified mechanism of extremely rapid
oligomer formation, via OH radical initiated oxidation of
organic aerosols. This process appears capable of
converting a sizable fraction of an organic particle to
higher mass oligomers within only a few hours of
exposure to OH radicals at typical atmospheric
concentrations. Furthermore, we have found that rapid
volatilization, followed by oligomerization, is also
important for specific reaction systems ( i.e. n-alkane
particles), and can lead to the loss of a large fraction (>
60%) of a particle within 15 minutes of exposure to
atmospheric OH. We propose that such a rapid processing
(oligomerization and volatilization) is possible due to a
radical chain reaction which quickly propagates
throughout the entire particle and is only initiated by the
surface OH reaction.
1. J. H. Seinfeld, S. N. Pandis, Atmospheric Chemistry
and Physics (Wiley, New York, 1998).
2. Y. J. Kaufman, I. Koren, Science 313, 655 (2006)
3. M. Kalberer et al., Science 303, 1659 (2004).
4. V. Samburova et al., J Geophys Res-Atmos 110,
D23210 (2005).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
5E.5
AFT-FTIR Investigation of the Heterogeneous Chemical
Reactions of Multi-component Aerosols and Ozone. CINDY
DEFOREST HAUSER, Stephanie Scott, DJ Singleterry,
Davidson College.
5E.6
A New Mini-flow-reactor for Aging Aerosols Without Wall
Effects. Xin Yang, Fudan University; Shanghai, China; Martin
J. Iedema, Hashim Ali, JAMES P COWIN, Pacific Northwest
National Laboratory.
The importance of aerosols in a wide range of
atmospheric processes is now well established, and has
led to concerted research efforts to better characterize
them. Characterization is complicated by heterogeneous
reactions of aerosols with gas-phase oxidants such as
ozone, halogen atoms, nitrogen trioxide and OH. Not
only is knowledge of the composition of the aerosol
important, but the rate of heterogeneous reactions and
factors affecting the rate determines the relevance of these
reactions in the atmosphere. Recent studies have further
shown that aerosols of more than one component and/or
phase show different reaction rates and mechanisms than
single-component aerosols. Methods to study these
reactions using entrained aerosols, in lieu of bulk liquid
measurements, are under continuous development.
Kinetics measurements of long time (< day) aging of
aerosols with trace gases require huge aerosol chambers
(e.g. <10m) to minimize "wall effects". More convenient,
supported-aerosol methods can require extremely
-5
unreactive substrates --gammas<<10 averaged over the
substrate-- and less than picogram/cm2 particle loadings,
to eliminate gross transport-limited trace gas depletion.
These limitations are avoided in a new mini-flow-reactor
(.003 m), with 102 -107 supported particles, and
subsequent sensitive (single) particle analysis. The flow
(550 cm/s) creates a 2 micro-meter "zone of isolation"
around each particle: which requires only that gammas <
0.1 locally (within 2 micro-meter). Measured with it was
the reactive uptake of gaseous HNO3 with deliquesced
NaCl particles. This mini-flow reactor is well-suited for
aging both lab and field collected aerosols, and
subsequent analysis with single particle analysis methods
and infrared microscopy .
Here, we will present the results of our studies of the
products and reaction times of laboratory-generated multicomponent organic aerosols with ozone using an
entrained aerosol flow cell followed by analysis via
Fourier transform infrared spectroscopy (AFT-FTIR). In
this method, a high number density mixed organic aerosol
is generated using homogeneous nucleation, mixed with a
stream of argon to maintain laminar flow and carried into
the flow cell. Ozone is introduced at one of five positions
along the cell, thus changing the reaction time of the gas
with the aerosols, or at one position with a range of
concentrations. Particles exiting the flow tube are heated
and vaporized in two stages. The evaporated particles and
equilibrium vapor then flow through a long pass cell,
heated to prevent recondensation, for analysis by FT-IR.
Comparison of the initial and reacted aerosol spectra
provides information about intermediates and final
products formed. Changes in spectral features as a
function of reaction time or concentration additionally
provide insight regarding the rate of the reaction.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
7D.1
Probing Hygroscopic Properties of Atmospheric Particles
Using Complementary Methods of Micro FTIR
Spectroscopy and Micro Analyses. Yong Liu, Pacific
Northwest National Laboratory; Zhiwei Yang, University of
Delaware; Yuri Desyaterik, Paul L. Gassman, Pacific Northwest
National Laboratory; Hai Wang, University of Southern
California; Alexander Laskin, Pacific Northwest National
Laboratory.
The phase and hygroscopic properties of atmospheric
particles play crucial roles in affecting global radiative
forcing and atmospheric composition by changes of
scattering, absorption and reactivity of aerosols. In this
work, deliquescence, efflorescence and hygroscopicity of
aerosols were investigated using FTIR microscopy and
assisted with ESEM and TOF-SIMS observations.
Substrate deposited submicron particles of atmospheric
importance, specifically NaCl, sea salt, NaNO3, (NH4)
2SO4, Ca(NO3)2 were used in the present case study to
demonstrate the feasibility of the applied approach.
Additionally, we report application of this approach to
study hygroscopic properties of few organosulfur particles
such as CH3SO3Na, CH3S(O)ONa, and CH3SO2CH3
which are important products of dimethylsulfide (DMS)
oxidation in marine boundary layer over the areas with
cold ocean surfaces. The deliquescence relative
humidities (DRH) and efflorescence relative humidities
(ERH) of aforementioned particles were determined by
integration of absorbance of H2O bands from FTIR
spectroscopy data. Water to solute molar ratios of
hydrating particles were quantified by comparing
integrated absorbance of H2O band with corresponding
-2
infrared active bands (e.g. NO3-, SO4 ) based on
integrated cross sections, if available, or measured by
Attenuated Total Reflectance Infrared Spectroscopy
(ATR/IR). Our DRH, ERH and hydration growth data are
in good agreement with previous levitation
measurements, demonstrating a promising new approach
to aerosol hydration studies. As shown in this
presentation, a distinct advantage of the substrate
deposited particle approach is its ability to employ
multiple microprobe analytical techniques over the same
sample to obtain complementary qualitative and
quantitative data.
7D.2
Probing the photochemistry of monoterpene-derived
secondary organic aerosols with chemical ionization mass
spectrometry. XIANG PAN, Joelle S. Underwood, and Sergey
A. Nizkorodov, University of California, Irvine.
Aerosol particles have a major impact on atmospheric
chemistry, climate, and human health. A significant
fraction of organic aerosol particles are formed as
secondary organic aerosol (SOA) by condensation of
partially-oxidized volatile organic compounds (VOC).
For example, monoterpenes have been shown to form
SOA in impressively large yields. Once such SOA
particles are formed, they age via physical
transformations and heterogeneous atmospheric
chemistry, often with profound effects on the physical and
chemical properties of the particles. The primary goal of
this research is to study the photochemical aging of
monoterpene-derived aerosol particles. Artificial SOA
particles are generated in the lab by reacting limonene or
alpha-pinene with O3 in a Teflon reaction chamber. The
particles are collected on filters and irradiated with
wavelength-tunable radiation in the actinic region
(lambda<290 nm). The resulting gas-phase photolysis
products are studied using chemical ionization mass
spectrometry (CIMS). The ability to monitor products
across a wide mass range using CIMS provides detailed
information about the photochemical processes taking
place within the aerosol particles. Reaction mechanisms
and their implications for photochemical aging of organic
aerosols, as well as the effects of relative humidity and
the presence of NOx on the photochemistry of SOA
particles, will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
7D.3
Measuring Particle Acidity in the Atmospheric Aerosol
Using a Colorimetric Analysis. MYOSEON JANG, Gang
Cao, Amanda L. Northcross, Jared Paul, The University of
North Carolina at Chapel Hill.
BACKGROUND. Our studies suggest that the
transformation of atmospheric organics via heterogeneous
acid-catalyzed reactions is important mechanisms for
SOA formation. However, the data interpretation of acid
effects on both laboratory SOA and ambient aerosol has
been restricted due to the limitation in analytical methods
for particle acidity.
APPROACH. We demonstrate a new operational
method using a colorimetric analysis integrated with a
reflectance UV-Visible spectrometer to measure the
proton concentration ([H+]) in a particle filter sample.
Colorimetric analysis measured changes at realistic
particle acidities over time due to short sampling time, no
workup procedure after particle collection on a filter, and
almost in situ measurement using a reflectance UVVisible spectrometer. The feasibility of colorimetry was
demonstrated both for SOA produced from ozonolysis of
alpha-pinene in acidic aerosol using indoor Teflon film
chambers and for ambient aerosols collected on the roof
of the McGavran-Greenberg Hall at UNC-Chapel Hill.
RESULTS. This method significantly reduces
sampling time to less than one minute for SOA created in
the chamber and to 30 minutes for ambient particles. The
SOA studies suggest that [H+] in aerosol dynamically
changes due to organic sulfate formation. The results
show that heterogeneous reactions in the presence of an
acid catalyst rapidly progress within a short time (minute
scale). The study applied to ambient particles indicated
that particle acidity changes as a function of time during a
day. The uncertainty of colorimetry is due to the
calculation of [H+] by a thermodynamic model. The
accuracy of calculation of [H+] affects the quality of the
calibration curve made for absorbance at 550 nm vs. [H
+]. Colorimetric analysis in conjunction with the
conventional methods using a pH meter or ion
chromatography intermediated through aqueous media
will significantly improve our ability to evaluate the
relation between organic carbon (OC) and particle acidity
in field data.
7D.4
FTIR Spectroscopy of Surficial Ozonolysis Reactions.
SCOTT A. EPSTEIN, Greg T. Drozd, Neil M. Donahue,
Carnegie Mellon University.
Ozone is an important reactant in the troposphere because
it is reactive with many organic species, especially
alkenes. Ozonolysis products often have lower vapor
pressures than the initial reactants and therefore are a
dominant source of secondary organic aerosol (SOA) in
the atmosphere. Key intermediates in ozonolysis are the
primary ozonide (POZ) and the Criegee intermediate (CI).
However, the POZ and the CI decompose too rapidly to
be observed in solution. In 1972, Heiklen et al. studied
simple alkene/ozone reactions on a cold surface using
offline infrared spectroscopy. (J. Amer. Chem. Soc. 94,
4856-4864). A similar apparatus has been built to
perform ozonolysis reactions on a liquid nitrogen cooled
window under vacuum. However, in the current study,
real-time Fourier Transform Infrared Spectroscopy
(FTIR) is used. Preliminary proof-of-concept
experiments with ozone and 2,3-dimethyl-2-butene yield
FTIR spectra indicating the presence of reactants,
intermediate species, and products on the cold window
surface. Temperature Programmed Reaction
Spectroscopy of intermediates isolated on the cold
window surface will constrain key decomposition barrier
heights. This in turn will reveal factors controlling
reaction product distributions for substituted compounds
typically found in the atmosphere.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
7D.5
A New Chamber Design for Aerosol Evolution Studies in the
Ambient Environment. CRYSTAL REED, Don Collins,
Texas A&M University.
Atmospheric aerosols impact climate, both directly
through scattering and absorption and indirectly through
cloud processes. Because of the complexity of the
atmospheric aerosol, understanding such impacts can be
challenging. In order to better understand the effects of
aerosols on the environment, we must study the processes
that transform aerosols, such as new particle formation,
condensation / dissolution, coagulation, and cloud
processing. Numerous studies have utilized laboratory
smog chambers to predict the behavior of the atmospheric
aerosol. Comparing such results to the behavior of the
ambient aerosol, however, is difficult given the
complexity of the atmosphere and the inability to replicate
such conditions. Captured air chambers, while more
representative of the ambient environment, result in a
decay in gas and particle concentrations with time.
Therefore, processes that have timescales of days will not
be accurately represented. Recently, a new chamber was
designed to study the evolution of a monodisperse aerosol
in actual ambient conditions over intervals of one or more
days.
9D.1
Integrated Raoult's Law and Henry's Law Approach for
Multiphase Organic Aerosol Partitioning. FRANK
BOWMAN, Karen Eskelson, Bonnie Fort, University of North
Dakota.
Organic gas-aerosol partitioning has been described in
atmospheric models using Raoult's Law (typically for a
primarily organic aerosol phase) and/or Henry's Law
(typically for an aqueous aerosol phase). In both cases,
activity coefficients are needed to account for nonideal
behavior in the aerosol solution. Saturation vapor
pressures, Henry's law constants, and activity coefficients
for organic aerosol compounds must typically be
estimated. As a result, thermodynamic inconsistencies
can arise when using Raoult's and Henry's Law
approaches together due to errors in estimation methods.
Comparison of known and estimated parameters for
representative semivolatile organics suggest overall
uncertainties of 2-3 orders of magnitude, resulting in
dramatic shifts in predicted gas-aerosol partitioning for
many compounds. Results also indicate that activity
coefficient estimates are the largest source of uncertainty.
By appropriately scaling estimated activity coefficient
values, Raoult's Law and Henry's Law can be integrated
in a thermodynamically consistent manner, despite
uncertainties in estimation methods.
The Ambient Aerosol Chamber for Evolution Studies
(AACES) is a roughly cubical chamber constructed of a
rigid Acrylite OP-4 acrylic outer shell, which transmits
UV radiation both in the UV-B and UV-A ranges. FEP
Teflon lines the inside of the chamber on all sides and the
top, while expanded-PTFE (ePTFE) Teflon is used on the
bottom of the chamber. The fibrous structure of the
ePTFE acts as a barrier to particulates, while allowing gas
molecules to move virtually unimpeded from one side of
the chamber to the other, creating an initial environment
inside the chamber that is free of particles and
continuously mimics the ambient air. Particles of known
size and composition can be then introduced, and
processes of aerosol transformation studied over much
longer periods of time than previously accomplished.
AACES construction and preliminary test results will be
discussed
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
9D.2
Simulating the Partitioning of Semivolatile Inorganic
Aerosol during the MILAGRO 2006 Campaign. CHRISTOS
FOUNTOUKIS, Athanasios Nenes, Amy Sullivan, Rodney
Weber, Georgia Institute of Technology; Timothy Vanreken,
National Center for Atmospheric Research; Marc Fischer,
Lawrence Berkeley National Laboratory; Edith Matias, Mireya
Moya; Universidad Nacional Autonoma de Mexico; Delphine
Farmer, Ronald Cohen, University of California Berkeley.
ABSTRACT
High time resolution measurements of aerosol and gasphase constituents coupled with the ISORROPIA-II
thermodynamic equilibrium model (which explicitly
treats K+/Ca2+/Mg2+/NH4+/Na+/SO42-/HSO4-/NO3-/
Cl-/H2O aerosol) are used to study the partitioning of
semivolatile inorganic species and phase state of Mexico
City aerosol sampled at the T1 site during the MILAGRO
2006 campaign. In the ammonia-rich environment of
Mexico City, nitrate and chloride primarily partition in
the aerosol phase; PM2.5 is insensitive to changes in
ammonia but is to acidic semivolatile species. When
below 50% RH, predictions improve substantially if the
aerosol follows a deliquescent behavior. Overall,
predictions agree very well with measurements of
ammonium, nitrate, chloride and gas phase ammonia.
Treating crustal species as "equivalent sodium" (rather
than explicitly) in the thermodynamic equilibrium
calculations has an important impact on predicted aerosol
water uptake, nitrate and ammonium. This suggests that
comprehensive thermodynamic calculations are required
to predict the partitioning and phase state of aerosols in
the presence of dust.
9D.3
Evaluation of New Approaches to Modeling Organic
Particulate Matter in CAMx. Bonyoung Koo, GREG
YARWOOD, Ralph Morris, ENVIRON International
Corporation; Kirk Baker, Lake Michigan Air Directors
Consortium.
Of constituents of atmospheric particulate matter (PM),
organic compounds have been one of the major
uncertainties. Modeling studies conducted for
southeastern US have shown that air quality models
exhibit poor organic PM performance with significant
underprediction in summer months.
Recently, two attempts to improve model representation
and performance for organics were made. The first one
focused on biogenic secondary organic aerosol (SOA)
formation and tried to incorporate potentially important
processes which have been missing in the traditional
modeling of SOA: (1) SOA formation from isoprene, (2)
SOA formation from sesquiterpenes, and (3)
Polymerization of SOA. Based on result from field
studies and laboratory experiments, Morris et al. (2006)
have modified the SOA algorithm in CMAQ to add these
processes. We have updated this approach and
implemented in CAMx. The comparison of model
performance between the updated and traditional SOA
schemes is presented.
Another approach recently proposed by Robinson et al.
(2007) introduced two amendments to the current
framework of primary organic aerosol (POA). In their
revised framework, POA emissions, which were
traditionally treated as non-volatile, are distributed over a
"basis set" based on their volatility, then allowed to
evaporate, oxidize (age), and recondense onto the particle
phase over time. This new concept is also implemented in
CAMx and its impact on the model performance of
organic PM is discussed.
References
Morris, R. E., B. Koo, A. Guenther, G. Yarwood, D.
McNally, T. W. Tesche, G. Tonnesen, J. Boylan, and P.
Brewer (2006) Model sensitivity evaluation for organic
carbon using two multi-pollutant air quality models that
simulate regional haze in the southeastern United States.
Atmos. Environ., 40, 4960-4972.
Robinson, A. L., N. M. Donahue, M. K. Shrivastava, E.
A. Weitkamp, A. M. Sage, A. P. Grieshop, T. E. Lane, J.
R. Pierce, and S. N. Pandis (2007) Rethinking organic
aerosols: semivolatile emissions and photochemical
aging. Science, 315, 1259-1262.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
9D.4
Effects of Uncertainties in the Thermodynamic Properties of
Organic Aerosol Components in an Air Quality Model.
SIMON L. CLEGG, University of East Anglia, Norwich, U.K.;
Michael J. Kleeman, University of California, Davis; Robert J.
Griffin, University of New Hampshire; John H. Seinfeld,
California Institute of Technology.
Air quality models that generate the concentrations of
semi-volatile and other condensable organic compounds
using an explicit reaction mechanism require estimates of
the physical and thermodynamic properties of the
compounds that affect gas/aerosol partitioning: vapour
pressure (as a subcooled liquid), and activity coefficients
(in the liquid aerosol phase(s)). One example is the model
of Griffin, Kleeman and co-workers (e.g., Griffin et al., J.
Geophys. Res. 110, D5, art. no. 05304, 2005). Here,
aerosol particles consist of an aqueous phase containing
inorganic electrolytes and soluble organic compounds,
and a hydrophobic phase containing mainly primary
hydrocarbon material. Thirty eight reaction products are
grouped into ten semi-volatile surrogate species which
partition between the gas phase and both phases in the
aerosol. Activity coefficients of dissolved organic species
are calculated using UNIFAC.
We have examined the likely uncertainties in the vapour
pressures of the semi-volatile compounds and their effects
on partitioning over a range of atmospheric relative
humidities. Uncertainties appear to be an order of
magnitude or greater, and this factor increases when the
fact that each surrogate compound represents a range of
reaction products (for which vapour presssures can be
independently estimated) is taken into account.
Dependencies of organic compound partitioning on the
treatment of inorganic electrolytes in the air quality
model, and the performance of this component of the
model, were determined by analysing the results of a
trajectory calculation using an extended version of the
Aerosol Inorganics Model of Clegg and Wexler (http://
www.uea.ac.uk/~e770/aim.html). Simplifications were
identified where substantial efficiency gains could be
made. The implications of the results, and this method of
analysis, for the development of aerosol models will be
discussed.
9D.5
Describing Volatility Evolution and Reversible Partitioning
Using the Volatility Basis Set. NEIL M. DONAHUE, Allen L.
Robinson, Carnegie Mellon University.
Atmospheric organic aerosols are an incredibly
dynamic mixture of many thousands of individual
compounds spanning a wide range of volatility.
However, a substantial body of evidence shows that a
large fraction of these compounds should be regarded as
semi volatile, meaning that they have a reasonable chance
of being found in the gas phase at some point during their
residence in the atmosphere. The atmosphere is a
dangerous place for a large, reduced organic compound to
be in the gas phase, and the resulting oxidation chemistry
will without question alter the organic volatility
distribution, ensuring that organic aerosol will remain an
extremely dynamic mixture throughout their lifetime.
Some compounds undergo condensed-phase reactions,
forming macromolecular products (oligomers), but the
overall effect of this on the organic volatility distribution
remains unclear.
By distributing organic material over a volatility basis set
we can describe the equilibrium partitioning of organics
over a single, internally mixed phase. While there is
some reason to believe that such a simple situation may
obtain in the background atmosphere, it is demonstrably
not true in urban areas where, for example, fresh, reduced,
vehicular emissions occupy a mode distinct from more
aged, oxidized material in the accumulation mode. This
talk will explore how many properties need to be
considered to adequately describe the life cycle of organic
aerosols from sources to sinks, including volatility,
oxidation state, and the size distribution, drawing on both
laboratory and field observations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
9D.6
Considering Compound Complexity and Aging in Models of
Organic Particulate Matter (OPM) Formation. JAMES
PANKOW, Oregon Health & Science University; Kelley
Barsanti, James Smith, National Center for Atmospheric
Research.
Modeling organic particulate matter (OPM) formation by
absorptive gas/particle (G/P) partitioning requires
knowledge of the identities and quantities of the
condensable compounds. Because detailed compoundspecific information is lacking for atmospheric aerosols,
OPM modeling has proceeded based on simplified
representations of the partitioning compounds. When
considering secondary OPM formation, foremost among
these representations is the two-product (2p) model of
Odum et al. (1996), which assumes that the range of
condensable compounds produced from each parent
hydrocarbon (HC) can be lumped into a pair of surrogate
products, 1 and 2. Application of the 2p parameters when
considering N different parent HCs is referred to here as
the N-2p model. Use of the N-2p model in regional and
global models has led to significant underprediction of
OPM levels. While some portion of this problem may be
due to missing parent HCs, some measure may be due to
failure of the N-2p approach to adequately represent the
complexity and time-dependence in the range of
condensable products that can form from a given parent
HC. Indeed, in its original form, 2p modeling of chamber
data with a given parent HC yields a narrow and static set
of Kp values: such a set cannot be expected to be
adequate for aging ambient OPM. Recent reports from
multiple laboratories indicate that products both more and
less volatile than lumped products 1 and 2 can form, and
will be themselves subject to alteration with time. A
range of reaction types have been identified that can lead
to essentially non-volatile polymeric (P) material.
Therefore, higher order secondary OPM models are
needed. A model that considers n initial products as well
as P material from a particular parent HC might then be
denoted a n(t)p(t)+P(t) model, where (t) denotes the
possibility of time dependence in the number of products
and their properties (e.g., increasing polarity due to
continued oxidation). Methodologies are discussed for
handling the resulting complexities in OPM modeling.
11D.1
Monte Carlo Simulations of Porous Film Deposition by
Electrohydrodynamic Atomization. CHRISTOPHER J.
HOGAN JR., Pratim Biswas, Washington University in St.
Louis.
In film deposition processes, control over the deposited
film morphology is critical, as film morphology affects
film properties and performance. Here, sequential Monte
Carlo simulations have been performed to predict the
morphology of films produced by electrohydrodynamic
atomization (EHDA) of nanoparticle sols. These
simulations show, for the first time, the effect of EHDA
process parameters, i.e. droplet size, droplet size
polydispersity, nanoparticle size, and deposition time, on
the thickness, porosity, surface roughness, and
characteristic feature size of deposited films.
Morphological parameters determined from simulations
were compared to parameters measured in EHDA
deposition experiments with a ZnO sol in ethanol.
Excellent agreement was found between experiments and
simulations. The simulation technique developed here is
a useful tool in the design of film deposition systems,
such as multi-jet EHDA systems for multi-component
film deposition with control over the film thickness,
roughness, and porosity.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.2
Multiple Scattering Measurements using Multistatic Lidar
in Aerosol Research Chamber. JIN H. PARK, C. R. Philbrick,
The Pennsylvania State University; Roy Gilles, Defence
Research and Development Canada Valcartier.
Multiple scattering is an important factor in treating the penetration of
radiation through an optically thick medium. A multistatic lidar system,
which was developed by Novitsky (2002), has the ability to evaluate the
multiple scattering effects in a dense medium by measuring polarization
ratio of the scattering phase function at different scattering angles. A
combination of multistatic imaging lidar techniques and polarization
measurements can eliminate many of the device non-linearities
(Novitsky, 2002), improve spatial resolution near the transmitter, and
reduce the dynamic range needed of the detector and electronics (Barnes
et al., 2003).
Experiments were conducted in a cooperative research project by PSU
researchers and researchers of Defence Research and Development
Canada Valcartier (DRDC) facility. The laser and two cameras were
separated by a 1.95 m from the laser beam path and each camera was in
line with the direction of the beam propagation inside the chamber. A
polarization cube and a 90 degree polarization rotator were used to
separate laser beam into two polarized components. The multistatic lidar
measurements were made in a 22-m long aerosol chamber which has 2.4
m times 2.4 m cross-section. The inside of the chamber is coated with
optical-black paint to avoid reflecting light from the wall. The aerosol
substitute is fog oil disseminated by a MDG Super Max 5000 fog-oil
generator. Size distribution of fog oil was measured using a particle size
spectrometer. The cameras used are commercial CCD cameras and have
field of view of 48 degrees at each direction. Measurements were made
in back and forward scattering directions.
The scattering-angle-specific polarization ratio is the scattered intensity
from incident parallel polarization divided by the intensity from incident
perpendicular polarization. Polarization ratio of single scattering is also
included in the results, which was calculated using a log-normal
distribution with geometric mean radius of 54.25 nm and a geometric
standard deviation of sigma g = 1.71. The fog particles had refractive
index of n = 1.51+i0 and were assumed spherical. In the results, some
background and CCD readout noise were introduced in each pixel,
which is summed in the beam-containing regions (Barnes et al., 2003).
Multiple scattering makes scattered radiation more depolarized as the
scattering angle increases from 0 and 180 degree respectively. Using the
multistatic lidar and the measurements of polarization ratio, multiple
scattering can be distinguished from single scattering over different
angular ranges.
____________________
Barnes, John E., Bronner, Sebastian, Beck, Robert, and Parikh, N. C.
(2003). Boundary layer scattering measurements with a charged-coupled
device camera lidar, Appl. Opt., 42, 2647-2652.
Novitsky, E. J. (2002). Multistatic Lidar Profile Measurement of Lower
Tropospheric Aerosol and Particulate, Ph. D Thesis, The Pennsylvania
State University.
Park, Jin H. and Philbrick, C. R. (2006) Multiple Scattering
Measurements Using Multistatic Lidar, 2006 International Aerosol
Conference, 1, 703-704.
11D.3
Photosynthesis in suspended bacterial aerosol droplet and
capsules in morphology dependent resonance conditions.
MIKHAIL JOURAVLEV, Tel-Aviv University, Israel.
Photosynthesis in the single suspended plant chloroplasts,
algae and bacteria in the microcapsulate form has the
specific features of photosynthetic processes occur in
aerosol droplets with light active inclusions. The
photosynthesis active inclusions are surrounded by the
low rate evaporative water layer of water aerosol droplet
or polymer membrane of encapsulate form. Due to
morphology dependent resonance effect there is strong
localization of the solar light intensity in bioaerosol
droplet volume. The photosynthesis model in aerosol
droplets based on C3 metabolites path and it represents by
in form of simplified two chains reactions: -light
reaction, and , - reaction of Calvin Cycle. Where:
substances of biochemical reaction described by ATP
(S*-substance) and ADP (S-substance) compounds are
involved in the scheme of light dependent reaction and
simplified carbon reduction cycle described by the two
substances reaction of RuBP (X-substance) and PGA (Ysubstance) [1]. The morphology dependent resonance
conditions of spheroidal geometry of water aerosol
droplet provide the concentration of electromagnetic
energy in light harvesting antennae of PCI and PCII
complexes localized in grana and thylakoid membranes. It
is effect strongly dependent on resonance condition of
wavelength of absorption chlorophyll molecules and
wavelength of electromagnetic modes of aerosol droplets.
The CO2 compensation point, limiting and enhanced
factors of photosynthesis in aerosol droplet are depends
on the input resonance conditions with the absorption line
of chlorophyll and light concentration volume as well the
thermal and gas exchange on the aerosol droplet surface.
The effective rate of O2 production and CO2 uptake with
diffusion of gases is depend on light intensity in
electromagnetic modes and influenced by the threedimensional droplet structure. The photosynthesis in
bioaerosol droplets and micro-size capsular forms are new
directions of investigation in aerosol science.
Reference:
1. P., Hari, et.al. (2000-2003) From atmosphere to sugars:
journey of CO2 molecule into three -dimensional matrix
of a leaf. Project of Forest Ecology Department.
University of Helsinki. 2000-2003. p.11. (Unpublished)
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.4
Surface Scattering for Charge Detection of Aerosol Droplets.
MIKHAIL JOURAVLEV, Tel-Aviv University, Israel.
11D.5
Controlled Multiscale Interaction of Aerosols. OLEG KIM,
Patrick Dunn, University of Notre Dame.
This report is aimed to develop the theory and
mathematical modeling of the nonlinear optical properties
of charged aerosol particles. The numerous optical
nonlinear effects accompany the propagation of the highpower electromagnetic energy inside the surface mode of
aerosol droplets. The high concentration of the
electromagnetic energy in the volume of the surface layer
under input resonance condition yields the nonlinear
scattering on droplets and provides the nonlinear shift of
eigenfrequencies of electromagnetic modes of droplets
[1]. The shift of eigenfrequencies depends on the
nonlinearity of substance of aerosol droplets as well on
the charge on surface of aerosol droplet. The shift of
eigenmodes in scattering particle wave amplitudes
provide possibility the charge detection of aerosol
droplets. The shift of each eigenmodes in the Mie
scattering is significantly more than the shift from
different nonlinear effects excluding the optical discharge
on the surface of aerosol droplets. The theory of coupled
modes is used to describe the dynamics of the selfinteraction of the single modes and the integral
coefficients of the self-interaction surface modes were
introduced.
As expect, the coupling coefficient of mode interaction
shift of eigenfrequency and increased with the density of
the charge on the surface of aerosol droplets. The think
layer of charge on the surface of the droplet is changing
the refractive index of the surface layer and the
eigenfrequency of the fiber. The boundary condition of
the evanescent and conventional end pumping used for
the excitation of the droplet modes and charge was
considered. The thin charge shell around the droplets
enhances the optical reflection of the electromagnetic
power both in to volume of the droplets and outside and
reduces the dielectric loss in addition it increases the
coupling coefficients of the self interaction modes. The
scattering emission of the charge coating is nonuniform to
image the light output pattern on a surface containing the
charged nanoparticles. This effect could be used for the
creation of the novel optical devices for the finding of the
environmental pollution and for the remote sensing
analysis. This result are significant for the potential used
of the compact scattering tools for identifying the
ultrafine particles by scattering amplitudes on the surface
of the droplets.
Reference:
In many industrial, environmental and biological processes, the
production and control of droplets and particles of micrometer
or even nanometer size with a narrow size distribution are of
interest. Aerosols of compound particles are of particular
importance for encapsulation of food additives and targeting
drug delivery, among other technological areas. The production
of some pharmaceuticals and "smart" materials, for example,
requires the precise control of a particle's components, which
can consist of both small and large aerosols. Nano-sized dieselengine exhaust or tobacco-smoke-material aerosols captured by
larger ambient aerosols are other examples.
The mechanisms involved in the capture of an aerosol of one
size by another of different size are length and time scale
dependent. When the length scales are micro-sized or greater,
impaction, interception and impaction are the primary deposition
mechanisms in a continuum regime. For smaller scale lengths,
diffusion becomes an important deposition mechanism, as well
as any phoretic mechanisms. The regime also could be
transitional or molecular, depending upon the governing
Knudsen number. The collection efficiency of smaller aerosols
by a larger aerosol is reduced as the difference between their
length scales increases. For such situations, electrical charge can
be used to control the rate of collision between the smaller and
larger particles. The characteristic coagulation time can be
decreased by using particles of opposite charge sign.
This presentation focuses on exploratory studies to produce
an aerosol whose components have different length scales.
Digital high-speed and still microphotography is used to view
the process. Two example cases are considered. In the first case,
small (micro-sized) electrosprayed droplets are deposited inflight onto large (mm-sized) solid aerosols of opposite charge. In
the second case, small (micro-sized), solid aerosols are collected
in-flight on an oppositely charged, larger (mm-sized), liquid
droplet. Copper and silver-glass microparticles are used for
producing solid aerosols, and water and ethanol are used for
liquid droplets. The formation of agglomerates is observed as
the liquid droplet evaporates. Estimations of the governing
parameters that control the growth of aerosol of different sizes
are made. The capture efficiency function is used to describe the
rate of aerosol interaction. This efficiency is related to the
Knudsen, Stokes and electric numbers. The number density of
the aerosol cloud of smaller aerosols required for the formation
of a mono-mer, di-mer or tri-mer in a low Reynolds number
flow are obtained for given particle concentrations, sizes,
relative velocities, and larger aerosol transit time through the
region of interaction. The effect of turbulence on particles
interaction also is considered.
1.V. Boutou, C. Favre, L. Woester,J.-P.Wolf. Opt.Lett.
Vol. 30.No7.p.759
The research described in this abstract was supported by Philip
Morris USA Inc. and Philip Morris International.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.6
Problems And Achievments In A Vapor-Gas Nucleation
Research. MICHAEL P. ANISIMOV, Institute of Chemical
Kinetics and Combustion, Siberian Division of the Russian
Academy of Sciences. Novosibirsk, Russia.
Nucleation is common phenomenon for the broad
spectrum of systems with phase transitions of the first
order. Critical embryo of new phase is treated as small
ball (droplet) with uniform density in the Classical
Nucleation Theory (CNT) approximation [1]. It is
assumed that CNT has the best fit to the experimental data
even CNT is distinguished among other \universal
theories\ by the internal consistency and simplicity of the
axiomatic statements only. The experimental data are
crossing usually the CNT predictions only [2].
Thermodynamics consideration comes to the formal
corrections to make consistent the next theory version to
the next set of experimental data. It is the ordinary event
when statistical mechanics is applied unreasonable for
nucleation phenomenon [1]. Nucleation experiment
accuracy has risen considerably during the last half of
century, but it does not lead to consistent experimental
results. Data from the different experimental methods are
not yet consistent each other [3, 4]. One can say that data
interpretation is realised up today in one channel
approximation [5] mostly.
An idea of semiempirical design of the nucleation rate
surfaces over diagram of phase equilibria is generated
over 10 years ago [6]. Design of semiempirical
presentations of nucleation rate theories permits to create
some basic set of nucleation theories, which can be used
for binary and higher dimension nucleation theories.
1. B. Senger, Schaaf, H. Reiss et al. J. Chem. Phys. 1999,
110, 6421.
2. R. Strey, P.Wagner, and T. Schmeling. J.Chem.Phys.
1985, 84, 2325-2335.
3. M.P. Anisimov, K. Hameri, and M. Kulmala. J.
Aerosol Sci., 25(1), 23, 1994.
4. D. Brus, A. Hyvarinen, V. Zdimal, H. Lihavainen. J.
Chem.Phys. 2005, 122, 214506.
5. L. Anisimova, P. K. Hopke, J. Terry, J. Chem. Phys.
2001, 114(22), 9852.
6. M.P. Anisimov, P.K. Hopke, D. Rasmussen, et al. J.
Chem. Phys. 1998, 109(4), 1435.
11D.7
Supercritical Vapor-Gas Binary Solution Nucleation.
MICHAEL P. ANISIMOV, Vladimir F. Podgornyii, Institute of
Chemical Kinetics and Combustion, Siberian Division of the
Russian Academy of Sciences. Novosibirsk, Russia; Philip
Hopke, Clarkson University.
Nucleation is the first step in nanomaterial production.
However, the theory of nucleation is not sufficiently well
developed. In this presentation, the nucleation of a vaporgas binary solution is discussed with respect of
nanomaterial production. Supercritical solutions of some
substance in gas are widely used for nanomaterial
generation even if the theory of that process is not fully
understood. Some researchers believe that there are no
effects of the nature and pressure of the carrier gas on the
nucleation rate [1, 2]. There have been a limited number
of publications of the role of carrier gas and the effect of
its pressure on vapor nucleation rates. Heist and coworkers [3-4] have reported, for example, effects of
pressure and nature of the (so called) non-condensable
carrier gas on the rate of nucleation for series of shortchain alcohols. In these experiments measurements have
made with a thermal diffusion cloud chamber using
pressures up to 0.4 MPa of H2, He, N2, and Ar as the
carrier gas. They observed a strong effect of both pressure
and gas nature.
The discrepancies between results of measurements made
with different experimental systems [5] should be noted
as the largest problem for nucleation science for present
time. A theory of nucleation from any supercritical
solutions cannot developed practically. We will show a
gap between equilibrium and spinodal states for binary
systems. The topology of nucleation rate surfaces will be
illustrated for some selected cases following the idea
formulated by Anisimov et al. [6].
1. Strey, R. and P.E. Wagner J. Phys. Chem. 86,1013
-1015 (1982).
2. Wedekind, J., Iland, K., Wagner, P., and Strey, R. In
Nucleation and Atmospheric Aerosols 2004. Eds. M.
Kasahara&M.Kulmala. Kyoto University Press, Kyoto,
49-52 (2004).
3. Heist R.H., J. Ahmed, and M. Janujua. J. Phys.Chem.
98, 4443-4453 (1994).
4. Berttelsmann, A., R.Stuczynski, and R.H. Heist. J.Phys.
Chem. 100, 9762-9773 (1996).
5. Brus, D., A.Hyvarinen, V.Zdimal, H.Lihavainen. J.
Chem.Phys. 122, 214506 (2005).
6. Anisimov, M.P., Hopke P.K., Rasmussen D.H., et al. J.
Chem. Phys. 109, 1435-1444 (1998)
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.8
Measurements of hygroscopic properties of ultrafine/nano
particles using the NanoTDMA technique. JAE-SEOK KIM,
Jiyeon Park, Kihong Park, Gwangju Institute of Science and
Technology, Gwangju, Korea.
Compositions of atmospheric ultrafine/nano particles
may vary depending on their sources, or formation and
growth pathways. Ammonium sulfate has been believed
to be one of major candidates for constituents of
nanoparticles formed by gas-to-particle conversion
process. These particles are believed to form by ternary
nucleation of sulfuric acid, water, and ammonia in the
ambient atmosphere, and grow by condensation and/or
coagulation. Elemental and organic carbons originated
from combustion sources can also be constituents of
nanoparticles. Hygroscopic properties of such small
nanoparticles will vary among sources, and their
measurements will provide useful insights into their
chemical composition. However, the hygroscopic
behaviors (e.g., growth factor (GF) and deliquescence
relative humidity (DRH)) of nanoparticles might be
different from those of large particles. Also, internally
mixed nanoparticles will show more complex hygroscopic
behaviors. In this study, we produce many types of
nanoparticles of NaCl, (NH4)2SO4, H2SO4, organic acid,
soot, and their internal mixtures by various generation
methods such as atomizer, furnace reactor, and laser
ablation reactor, and the hygroscopic properties of sizeselected nanoparticles are measured with the newly
developed NanoTDMA technique. Preliminary results for
hygroscopic properties of NaCl nanoparticles showed that
for particles of 10, 15, and 20 nm, their GF decreased and
DRH increased compared to those of 30 nm particles,
independent of generation methods. The NaCl particles
produced by the furnace reactor system (evaporationcondensation method) initially shrunk with increasing RH
below the DRH, while those produced by the atomizer
showed no such shrinkage. TEM images of the furnace
reactor-generated nanoparticles suggest that these
particles are loose aggregates consisting of small NaCl
crystals, probably restructuring into compact aggregates
with increasing RH.
11D.9
A CECD Web-Based Course for Particle Transport,
Deposition and Removal. GOODARZ AHMADI, Stephen
Doheny-Farina, John McLaughlin, Suresh Dhaniyala, Cetin
Cetinkaya, Jeffrey Taylor, Kambiz Nazridoust, David J,
Schmidt, Xinli Jia,and Xiangwei Liu, Clarkson University; Mark
Glauser, Syracuse University; Fa-Gung Fan, Xerox Corporation;
Ahmed Busnaina, Northeastern University.
The primary objective of this combined research and
curriculum development project is to make the fruits of
these new important research findings available to seniors
and first year graduate students in engineering through
developing and offering of sequence of specialized
courses. In these courses, the process of particle
transport, deposition and removal and re-entrainment was
described. An extensive web for the course materials was
developed and the courses were taught simultaneously at
Clarkson University and Syracuse University.
These combined research and curriculum development
(CRCD) courses are composed of four modules. The
models are:
- Fundamental of particle transport, dispersion, deposition
and removal.
- Computational modeling of particle transport, deposition
and removal.
- Experimental study of particle transport, deposition and
removal.
- Industrial applications of particle
References: http://www.clarkson.edu/projects/crcd/
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.10
Bumpy Particle Adhesion and Removal in Turbulent Flows
-. GOODARZ AHMADI, Shiguang Guo, Clarkson University.
The effect of electrostatic and capillary forces on bumpy
particle adhesion and removal in turbulent flows is
studied. The JKR theory is used and the increase of
adhesion by the capillary force is accounted for. The
effects of electrostatic forces and nonlinear hydrodynamic
drag are included in the analyses. The criteria for
incipient rolling and sliding detachments and electrostatic
lifting removal are evaluated. A turbulence burst model is
used for evaluating the peak air velocity near the
substrate. The critical shear velocities for detaching
particles of different sizes under different conditions are
evaluated. The electric field strength needed for
electrostatic removal of particles with different charges is
also estimated. The results are compared with those
obtained in the absence of the capillary force.
Comparisons of the model predictions with the available
experimental data are also presented.
11D.11
Prediction of Deposition Pattern in a Particle Laden
Turbulent Channel Flow by Large Eddy Simulatio. Mazyar
Salmanzadeh, Shahid Bahonar University of Kerman (Iran) and
Clarkson University; Mohammad Rahnama, Shahid Bahonar
University of Kerman (Iran); GOODARZ AHMADI, Clarkson
University.
Large-eddy simulations (LES) of particle transport and
deposition in turbulent channel flow were presented.
Finite volume method was used for Large-eddy
simulation of Navier-Stokes equations for finding
instantaneous filtered fluid velocity field of the
continuous phase in the channel. Selective structure
function model was used to account for the subgrid-scale
Reynolds stresses. The Lagrangian particle tracking
approach was used and the transport and deposition of
particles in the channel were analyzed. The Stokes drag,
lift, Brownian and gravity forces were included in the
particle equation of motion. The Brownian force was
simulated using a white noise stochastic process model. It
was shown that the LES was capable of capturing the
turbulence near wall coherent eddy structures and the
initial location of deposited particles is concentrated
around certain bands.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.12
Characteristics of Aerosol Growth Events at Urban and
Rural Locations in New York. MIN-SUK BAE, James J.
Schwab, Kenneth L. Demerjian, Olga Hogrefe, G. Garland
Lala, Qi Zhang, University at Albany, SUNY; Brian P. Frank,
New York State Department of Environmental Conservation.
11D.13
Method for parameterizing the effect of sub-grid scale
aerosol dynamics on aerosol number concentration emission
rates. JEFFREY R. PIERCE, Peter J. Adams, Carnegie Mellon
University; Georgia Theodoritsi, Spyros N. Pandis, University
of Patras, Greece.
Measurements of particle number concentrations and size
distributions were obtained using both nano- and longtube scanning mobility particle sizers (SMPS) at the rural
site of Pinnacle State Park (PSP) in upstate New York
from July 18 through August 8 in 2004 and the urban site
of Queens College (QC) in New York City from January
9 through February 6 in 2004. Due to the relatively high
background levels of aerosols (and the associated
condensation sink for new particles), condensation and
growth events are much less common in these locations
than reported by other groups for lower background sites.
With co-located measurements (such as ionic species
from PILS-IC, oxygenated/hydrocarbon-like organic
aerosol (OOA / HOA) from an Aerodyne Aerosol Mass
Spectrometer, organic & elemental carbon from a
SUNSET (NIOSH5040) real time carbon aerosol
analyzer, gaseous pollutants, PM2.5 using a tapered
element oscillating microbalance (TEOM), and
metrological data), diurnal variation and correlation
between particle number size distributions and ambient
pollutants at two monitoring sites present a rich data set
for exploration of particle growth events at high average
background sites. Analysis of the characteristics of
particle growth in comparison to other observations allow
classification of the events and illustrate that particle
growth contains different physical behavior related to
different compounds for these sites. The growth rate
associated with each event will be discussed in relation to
physical, chemical, and meteorological conditions.
One of the major challenges in simulating the aerosol
number concentration and number distribution in the
atmosphere is the description of aerosol dynamics near
the sources of the primary particles. These emission \hot
spots\ may be metropolitan areas in global models and
large point sources in global and regional models. Most
models currently simulate the average particle number
concentration in the grid cell, spreading the effect of the
hot spot unrealistically across the cell. However,
coagulation does not conserve particle number and
numerical \dilution\ of the emitted particles in the full grid
cell introduces potentially significant bias in the model
results. Unfortunately, simulation of the rapid dilution of
particles as they disperse away from their source together
with their coagulation, removal, and growth or
evaporation is prohibitively expensive for regional and
global chemical transport models.
In this study, we develop a method for the
parameterization of the sub-grid scale aerosol dynamics.
This method calculates the probability that a given
particle emitted inside the grid cell will survive and be
available for transfer outside the cell. This survival
probability is calculated theoretically as a function of the
emitted particle size, the pre-existing aerosol size
distribution in the grid cell, the meteorological conditions,
and the size of the grid cell. The net number of particles
effectively emitted to the grid cell can then be calculated
by multiplying the size dependent emission rate in the
inventory with this survival probability. The method
simultaneously conserves mass by adding the mass of
particles lost by coagulation to the larger particle sizes.
The approach is grid-size independent and can be used in
models of all scales. Its results are evaluated against the
predictions of a detailed one-dimensional aerosol
dynamics and chemistry model under a variety of
atmospheric conditions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.14
Estimating the contribution of wall loss and condensation/
evaporation to aerosol size evolution in smog chamber
experiments. JEFFREY PIERCE, Gabriella Engelhart, Emily
Weitkamp, Ravikant Pathak, Neil Donahue, Allen Robinson,
Peter Adams, Carnegie Mellon University; Spyros Pandis,
University of Patras, Greece.
Smog-chamber experiments involving aerosol often
require the calculation of the condensation onto the
particles. An important example is the determination of
secondary organic aerosol (SOA) yields from the
oxidation of gas-phase precursors. Complicating the
calculation of the condensation rate are uncertainties in
the wall-loss rates. Wall-loss rates generally depend on
particle size, turbulence in the bag, the size and shape of
the bag and particle charge. In analyzing smog-chamber
data, some or all of the following assumptions are
frequently made regarding the first-order wall-loss rate
constant: a) that it is constant during an experiment; b)
that it is constant between experiments; c) that it is not a
function of particle size. Each of these assumptions may
not be justified in some circumstances.
We present the development and testing of an inverse
model based on the aerosol general dynamic equation to
determine best estimates for the size-dependent
condensation rate and size-dependent wall-loss rate as a
function of time. Size distribution measurements from an
SMPS provide time boundary conditions that constrain
the general dynamic equation. Wall loss is explored using
data from a smog-chamber experiment of ammonium
sulfate aerosol without condensation. The coupled wall
loss/condensation effects are investigated using limonene
oxidation to form SOA. We compare our model to other
wall-loss correction methods and show that the yield of
SOA calculated from the limonene oxidation may vary by
over 20% due to uncertainty in the aerosol wall loss.
11D.15
The Influence of Particle Shape on the VUV Photoelectron
Imaging of Nanoparticles. MATTHEW J. BERG, Christopher
M. Sorensen, Amit Chakrabarti, Kansas State University; Kevin
R. Wilson, Musahid Ahmed, Stephen R. Leone, Lawrence
Berkeley National Laboratory.
The electronic structure of nanoparticles is of
fundamental interest [1]. This electronic structure can be
described by photoelectron emission which can be
initiated by Vacuum Ultraviolet (VUV) light of
wavelength in the 100-nanometer range. An important
element in the photoelectron emission process is the
influence that the particle's shape has on the distribution
of light within the particle. This work presents
simulations of the electromagnetic field that is induced
inside spherical and cubic-shaped particles by the VUV
light, and relates this field to aspects of the particle's
electron emission distribution. The simulation results are
compared to laboratory measurements conducted with the
Advanced Light Source at the Lawrence Berkeley
National Laboratory.
[1] Wilson, K. R., Peterka, D. S., Jimenez-Cruz, M.,
Leone, S. R. and Ahmed, M., ``VUV photoelectron
imagining of biological nanoparticles: Ionization energy
determination of nanophase glycine and phenylalanineglycine-glycine," Phys. Chem. Chem. Phys. v. 8, pp. 1884
-1890.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.16
Bipolar Diffusion Charging Characteristics of Airborne,
Single-Walled Carbon Nanotubes. PRAMOD KULKARNI,
Gregory Deye, Paul Baron, National Institute for Occupational
Safety and Health.
Single-walled carbon nanotubes (SWCNT) have many
promising technological applications. Large-scale
industrial production of SWCNTs has led to increasing
concerns over their health risks. Measurement of size
distribution of airborne SWCNTs is necessary to
understand their health implications. Particularly, for
electrical mobility based measurement, understanding
diffusion charging properties of such nonspherical
particles in bipolar ion field is important. Once
aerosolized from the bulk material phase, individual
SWCNTs tend to form large, open and low-density
agglomerates due to strong interparticle van der Waals
interactions. These agglomerates have nanoscale fibrous
structures that resemble entangled web of individual
nanotubes, and are characterized by large surface area,
much larger than that obtained from their physical
enveloping diameter. It is therefore important to
understand the effect of complex structure and
morphology of these aerosols on their diffusion charging
properties.
11D.17
Motion of a Drop through a Fabric in Presence of
Wettability Gradient. HOJAT NASR, Goodarz Ahmadi, John
B. McLaughlin, Xinli Jia, Clarkson University.
This study is concerned with the numerical simulation
of a moving drop through a fabric due to a wettability
gradient. The wettability gradient is introduced by
varying the contact angle along the staggered fibers. The
unsteady laminar Navier-stokes equation is solved in each
phase using a fixed Eulerian structured grid. The Volume
of Fluid Model (VOF) is used to account for tracking the
gas/liquid interface. A water drop was placed on top of
the fabric with a small initial velocity, and the motion of
the drop through the fabric with different contact angle
distributions was studied.
Bipolar diffusion charging of SWCNT aerosol with
complex morphologies is investigated. Previous studies
(Rogak and Flagan, J. Aerosol Sci., 23, pp 693, 1992)
suggest that the fraction of uncharged particles leaving a
bipolar charger is as sensitive to shape as singly or doubly
charged particles, and can be used to study the effect of
morphology on charging properties. DMA classified
SWCNT aggregates, PSL spheres, and ammonium sulfate
spheres with mobility diameters ranging from 100 to 800
nm were experimentally studied to determine the fraction
of uncharged particles exiting the bipolar charger. Bias
introduced by concentration fluctuations and multiple
charging were also accounted for. The uncharged
fractions are compared with those expected from Fuch's
and Boltzmann charge distributions to infer whether the
DMA inversion procedures developed compact spherical
particles could be used for size distribution measurement
of airborne SWCNTs.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.18
Dependence of Aerosol Scattering on Relative Humidity and
Particulate Composition. WIESJE MOOIWEER, Derek C.
Montague, Yong Cai, Terry Deshler, University of Wyoming.
11D.19
High Speed Aircraft-Particle Interaction: Application to
Aerosol Sampler Design. ARASH MOHARRERI, Suresh
Dhaniyala, Clarkson University.
Scattering extinction has been measured over a range of
relative humidity values for aerosols at both a small urban
environment (Laramie, Wyoming) during both summer
and winter, and in a clean high-altitude mid-continental
site (Elk Mountain, 3320 m) in summer. Measurements
on the initially dry (relative humidity (RH) typically
<30%) aerosol, restricted to PM1 by an upstream cyclone,
were obtained with two sequentially operated Radiance
Research M903 nephelometers separated by a Nafion tube
humidifier. Size distributions were simultaneously
obtained from SMPS, PCASP, APS, and UHSAS
observations, as was the size-resolved chemical
composition of the particulates (5 minute averages), using
an Aerodyne Aerosol Mass Spectrometer. RH, repeatedly
cycled between 25% and 85%, with an approximate 35
minute periodicity, was monitored, together with the
airflow temperature, by Vaisala HMP238 sensors.
Scattering enhancement (f(RH) = bscat(RH)/bscat(RHmin))
and implied particle growth characteristics can be
interpreted in terms of particulate composition, which
often varied significantly on time scales shorter than that
of the humidification cycle. Increasing percentages of
organic material (OM) in the aerosol were observed to
depress scattering enhancement at high RH, resulting in
an inverse correlation between OM % and f(RH), even for
short temporal variations in OM. Similar to previous
reports, values as low as 1.1 were observed when the OM/
(NH4)2SO4 ratio was 12.0, whereas for OM/(NH4)2SO4 =
0.53, f(RH) was typically about 2.2. This inverse
correlation is also reflected by generally higher scattering
enhancements in winter, when the PM1 that contributes
most of the scattering contains lower average organic
relative mass loading (~43%). Particulate organic
material can be separated into hydrocarbon (HOA) and
one or more oxygenated components (OOAI and OOAII)
using Zhang et al.'s (2005) deconvolution algorithm.
Preliminary analyses suggest that the effectiveness of
scattering enhancement suppression by organics at high
RH differs for each of these components.
Aerosol particles are ubiquitous in the atmosphere and
play an important role in local visibility, human health,
and global climate. Their contribution to global climate is
largely through interaction with solar and terrestrial
radiation as well as interaction with the formation and
precipitation efficiency of liquid-water, ice, and mixedphase clouds. Investigating these effects requires physical
and chemical characterization of activated and nonactivated aerosol over a range of cloud systems, often
from aircraft platforms. A particular challenge for such
aerosol-cloud characterization studies is to sample
particles without contamination and at known
enhancement efficiency, accounting for aircraft and
sampler body influence on particle trajectories in the
vicinity of the sampling inlet. In this study, theoretical
analysis of flow and particle motion around a blunt body
is conducted in order to understand particle concentration
variation around a sampler body and to facilitate the
interpretation of data taken from inlets located on bluntbody samplers. Flow fields around blunt-bodies of
different shapes, including cylinders, spheres, and
Joukowski airfoils with different aspect ratios, are
obtained using potential flow simulations. Particle
trajectories and concentrations around a sampler body are
calculated using an Eulerian-Lagrangian method.
Integrated in these simulations are the effect of particle
bounce and splattering from the blunt-body sampler. The
modeling results are used to identify the optimal body
shape and inlet location for effective sampling of
interstitial particles without interference from the
activated cloud droplets.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.20
Estimating Single Scattering Albedo, Asymmetry Parameter
and Aerosol Optical Depth in the Ultraviolet Using an
Operational Retrieval Algorithm for Houston, TX.
CHELSEA CORR, Thomas Taylor, Sonia Kreidenweis, James
Slusser, John Davis, Colorado State University; Barry Lefer,
University of Houston.
Several recent studies have observed significant radiative
effects of aerosols on ultraviolet (UV) flux and irradiance
at the surface. However, aerosol spectral optical
properties in the UV remain largely uncharacterized.
Total and diffuse horizontal irradiance and by subtraction
direct normal irradiances were measured in Houston,
Texas during the Texas Air Quality Study/Gulf of Mexico
Atmospheric Composition and Climate Study (TexAQS/
GOMACCS) in September 2006 using a UV Multi-Filter
Rotating Shadowband Radiometer (UV-MFRSR).
Irradiances were reported at 3-minute time intervals at
seven wavelengths: 300, 305.5, 311.4, 317.6, 325.4,
332.4 and 368 nm which were subsequently used to
calculate both aerosol optical depths (AOD), using Beer's
Law after removing contributions from Rayleigh
scattering and ozone absorption, and the total ozone
column (TOC), via the direct-Sun method (Gao et al.,
2001).
The values of AOD and TOC were used as a priori
estimates in an operational algorithm to estimate a
wavelength-independent asymmetry parameter (g), and
aerosol single scattering albedos (SSA) and AODs for the
seven wavelengths of interest via Bayesian optimal
estimation (Taylor et al., in review). Irradiances were
cloudscreened prior to analysis by the retrieval algorithm.
The retrieved AOD and SSA values were compared with
AOD and SSA estimates from an AERONET Sun
photometer at the site. Retrieved TOC was compared to
TOC estimates from a NOAA/EPA Brewer
photospectrometer also collocated with the UV-MFRSR.
_____________
Gao, W. et al., Direct-Sun column ozone retrieval by the
ultraviolet multi-filter rotating shadowband radiometer
and comparison with those from Brewer and Dobson
spectrophotometers, Applied Optics, 40, 3149-3155,
2001.
Taylor, T.E. et al., An Operational Retrieval Algorithm
for determining aerosol optical properties in the
ultraviolet, Journal of Geophysical Research, in review.
11D.21
Relation between Electrical Mobility, Mass, and Size in the
Nanometer Range of Charged Nanoparticles Generated by
Electrosprays. BON KI KU, National Institute for
Occupational Safety and Health; Juan Fernandez de la Mora,
Yale University; Sven Ude, Germany.
A large number of new data on mobility and mass has
been obtained for clusters of a diversity of materials, with
the aim of determining the relation between electrical
mobility and size for nanoparticles in the size ranging
from 1 to below 10 nanometer. Solutions of different
materials (ionic liquids, tetra-alkyl ammonium salts,
cyclodextrin, bradikinin, etc.) in acetonitrile, ethanol or
formamide (~ 0.005 up to 0.1 mol/l) were electrosprayed,
charge-reduced (to unity) with a Po-210 source (5 mCi),
and their electrical mobilities were measured by a
differential mobility analyzer (DMA) of the Herrmann
type, which has an unusually high resolution and uses an
electrometer as a detector [1]. Mass assignments were
established indirectly by first distinguishing singly and
doubly charged clusters, and then determining the
structure of the ions associated to the various mobility
peaks observed.
Diameters of different nanoparticles based on mass are
represented as a function of mobility Z. For a sphere of
diameter d above 2-3 nm, both m^1/3 and (z/Z)^1/2 are
linear with d, so that a spherical shape should yield a
straight line (z is charge state, and m particle mass). From
a linear fit of m^1/3 and (z/Z)^1/2 for each material at
large enough sizes. This fit at large sizes leads to
excellent fit at all sizes for all species, with the exception
of doubly charged PEG. The resulting particle densities
were compared to measured bulk densities for ionic
liquids leading to good agreement for all but one ionic
liquid.
References
1. B. K. Ku and J. Fernandez de la Mora (2004). J. Phys.
Chem. B, 108(39), 14915.
2. S. Ude, J. Fernandez de la Mora, and B. A. Thomson
(2004). J. Am. Chem. Soc., 126, 12184.
Disclaimer
The findings and conclusions in this abstract have not
been formally disseminated by the National Institute for
Occupational Safety and Health and should not be
construed to represent any agency determination or
policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.22
Experimental Study for Charge Limit of Nanoparticle Using
Condensation and Evaporation Method for Particle
Charging. JOONGHYUK KIM, Youngjoo Choi, Woojin Kim,
Sang Soo Kim, KAIST, Korea.
11D.23
FracMAP: A Graphical Iser-interactive Package for
Performing Simulation and Morphological Analysis of
Fractal-like Aerosol Agglomerates. Rajan K. Chakrabarty,
Mark A. Garro, Hans Moosm, Desert Research Institute
The charge limit of nanoparticle using condensation and
evaporation method for particle charging was studied
experimentally. The research of nanoparticle charging has
been of many interests in the nanoparticle control and
nanostructure synthesis, for example, nanopatterning,
nano electronic device and so on. As the size of
nanoaprticles are decreased, nanoparticle has limit of
charging by diffusion charging, field charging and bipolar
charging used principally in the aerosol technology,
which is due to some factors such as ion evaporation. To
observe this phenomenon, particle charging using the
condensation and evaporation method was used. Silver
nanoparticles were grown into microdroplets through the
condensation in the ethylene glycol as condensable vapor.
The diameter of condensed droplet was about 2
micrometer and grown droplets were charged uniformly
by indirect corona-based charger. Then, ethylene glycol
coating silver particle was evaporated through the thermal
evaporator, and nanoparticles with high charge remained.
The number of charges was measured by differential
mobility analyzer. Since these nanoparticles were charged
at same sized droplets, they should have uniform charges
regardless of their sizes. However, it is observed that the
amount of charge was decreased as the diameter of
nanoparticle was decreased below 20 nm. This result
means the existence of nanoparticle charging limit. This
work will help various nanotechnologies such as
nanostructures synthesis by controlling charged
nanoparticles.
Computer simulation techniques have proven to be quite
useful for studying the structure of fractal-like aerosol
agglomerates produced by the process of growth through
the aggregation of smaller particles (monomers). The
software package FracMAP numerically simulates the
aggregation of monomers into monodispersive
agglomerates in three-dimensional (3-d) space using the
Monte Carlo method. Making use of a highly efficient
algorithm, the code generates all the possible projections
(similar to that of electron-microscopy) of the 3-d
agglomerate onto a two-dimensional (2-d) plane. The
code is coupled to a graphical user interface, thereby
enabling the user to input all the necessary parameters for
the generation of the 3-d agglomerates. These 3-d
agglomerate and their 2-d projections can also be
graphically dispayed for visual inspection. The program
has various subroutines for performing structural and
fractal analysis of the agglomerates in both 2-d and 3-d.
These subroutines are presented as options to the user on
the user-interface screen. Based on the user input, the
program calls for the specific subroutines, executes them,
and returns the structural analysis results to the user.
FracMAP is written in C++, and the program has been
tested to run successfully under UNIX and WINDOWS
operating systems.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.24
Forces Affecting Particle Adhesion to Complex Surfaces.
JONATHAN THORNBURG, Li Han, RTI International; Jacky
Rosati, U.S. EPA NHSRC.
11D.25
Adhesion and Removal Mechanism for Particles in
Turbulent Flows with Electrostatic Effects. XINYU
ZHANG, Goodarz Ahmadi,Clarkson University.
Particle adhesion to complex surfaces affects a wide range
of activities including aerosol dissemination,
decontamination, surface cleaning, precision equipment
performance, and resuspension. However, the
contribution of different adhesion forces is a fundamental
and poorly understood particle interaction issue that is
dependent on the particle size distribution and surface
properties. This research focused on the development of a
new method to better understand this fundamental aerosol
particle property. A method using Atomic Force
Microscopy was developed to measure the contributions
of van der Waals, capillary, and electrostatic forces to the
cumulative adhesion force. An overview of the new
adhesion force measurement method will be presented.
This method was applied to multiple particle-surface
combinations. Johnson-Kendall- Roberts (JKR) theory,
an adhesion contact model, was used to discriminate the
adhesion force from the hysteresis force. Data illustrating
the influence of particle size, composition, and
morphology as well and surface characteristics on the
adhesion force components will be presented.
The electrostatic effect on particle adhesion and
detachment in turbulent flows was studied based on the
newly developed maximum moment criterion. The
corresponding maximum moments for different adhesion
models were used for evaluation of the criteria for
incipient rolling detachments. The structure of turbulent
near wall flows was included in the analysis. The critical
shear velocities for detaching various size particles based
on different models were evaluated and the results in the
presence of electrostatic forces are compared with those
obtained in the absence of the electrical forces. It is found
that the electrostatic effect increases the critical shear
velocity for particle detachment. Comparisons of the
model predictions with the available experimental data
show good agreements.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
11D.26
Volatility Measurements of Secondary Organic Aerosol
Using a Thermodenuder. BYONG-HYOEK LEE, Gabriella J.
Engelhart, Jeffery R. Pierce, Carnegie Mellon University;
Spyros N. Pandis, Carnegie Mellon University and University of
Patras.
An improved thermodenuder is used to measure the
volatility of secondary organic aerosol (SOA) in the
Carnegie Mellon smog chamber. The thermodenuder
avoids complications due to rapid mass and heat transfer
processes by combining thermal stability from 30 to 400
degree celsius and a wide range of user-selected residence
times of the aerosol in the heated zone. The volatility of
SOA produced from the ozonolysis of alpha-pinene, detapinene, and limonene, at low and intermediate RH, and at
low and high NOx conditions was investigated. More than
90% of the alpha-pinene and beta-pinene SOA volume
and approximately 75% of the limonene SOA evaporated
at 70 degree celsius for a residence time of approximately
16 seconds.
The volatility of SOA produced during the alpha-pinene/
NOx photooxidation was investigated in the 25 to 220
degree celsius temperature range. Almost 98% of the
SOA volume evaporated at 75 degree celsius after 15.8
seconds in the thermodenuder heating section. However,
more than 20% of the SOA volume did not vaporize at
150 degree celsius when the residence time was reduced
to 1.6 seconds. The remaining fraction of the aerosol after
passing through the thermodenuder is quite sensitive to
the residence time of the particles in the heated zone. A
multi-component organic aerosol dynamics model using
the volatility basis-set is developed to explain the effects
of the residence time on the SOA volatility
measurements.
11D.27
Size-Resolved Kinetics Measurement of Nickel Nanoparticle
Oxidation by Electrical Mobility Classification. LEI ZHOU,
Ashish Rai, Nicholas Piekiel, Michael R. Zachariah, University
of Maryland.
Nano scaled nickel particles have attracted great research
interest lately for its potential impact as a fuel in energetic
materials. In this work, we applied both tandem
differential mobility analysis (TDMA) and tandem DMAAPM techniques to study the reactivity of size selected
nickel nanoparticle. Nickel nanoparticles were generated
in-situ using gas-phase thermal pyrolysis of nickel
carbonyl. Three particle sizes (50, 70 and 100 nm,
mobility size) were than selected by using a differential
mobility analyzer. These particles were sequentially
oxidized in a flow reactor in air in the temperature range
of 25-1100
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12C.1
Thermal Equilibration of Soot Electrical Charge by Particle
Coagulation. MATTI MARICQ, Ford Motor Company
The coagulation of primary particles leads to the
fractal-like aggregates that typify the morphology of soot.
Using differential mobility analysis it is possible to
investigate the self preserving size distributions produced
by coagulation, and how their shape depends on the
fractal dimension of the aggregates as well as the
underlying collision dynamics, free molecule versus
diffusion controlled. Furthermore, flame and engine
produced soot particles are naturally electrically charged
as a result of combustion. The charge distribution itself
evolves a result of particle coagulation.
The present paper examines soot particle coagulation both
by experiment and model. Overall particle size
distributions are compared to predictions based on the
Smoluchowski equation. This is done for a variety of
flame conditions that enable generating particles with
fractal dimensions in the range of 1.9 - 2.5, and in size
ranging from ~10 nm to ~300 nm, thus spanning the
interesting transition regime between free molecule and
diffusion dynamics. In addition, charge specific size
distributions are measured. Coagulation in the flame
leads to rapid equilibration of the electrical charge to a
Boltzmann distribution at the flame temperature. When
soot particles are sampled from the flame and allowed to
coagulate in a residence chamber, coagulation leads to reequilibration of the charge distribution to room
temperature. That this irreversible process leads to an
equilibrium distribution occurs because, just as the
Boltzmann expression, the coagulation rate constants
scale exponentially with the ratio of electrical to thermal
energy.
12C.2
Bringing Bioaerosols into a Microfluidic Cell using
Electrospray. HERMES HUANG, Richard Chang, Yale
University.
Biochemical assays can be used for accurate identification
of bioaerosols. The task of putting an aerosol into solution
to perform the assay is not an easy one. We have
developed a new technique, however, using electrospray
for putting particles into solution directly within the input
well of a microfluidic cell, where the biochemical assay
can then be performed.
Electrospray is a technique where a monodisperse conical
spray is generated out of a thin nozzle due to a high
electrostatic field. This spray is highly charged, and is
often used to generate ions for ion-mobility mass
spectrometry or to charge particles.
By using a novel new design where an electrode is built
into the input well of a microfluidic cell, we are able to
generate an electrospray directly into the input well of the
microfluidic cell. When this spray intersects an aerosol
stream, the spray will charge the aerosols. The spray
droplets also impart a perpendicular momentum to the
aerosol, further directing the aerosol towards the
microfluidic cell. This combination of charging,
momentum transfer, and the electric field which is also
used to produce the electrospray all work together to
direct the aerosol into the input well of the microfluidic
cell. As an added bonus, the electrospray liquid also
replenishes the fluid within the well, so that the system
could be run continuously if desired.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12C.3
Sampling and Measurement of Mainstream Cigarette Smoke
Puffs with a Cascade Impactor. David B. Kane, Steven S.
Larson, Philip Morris USA.
Mainstream cigarette smoke, the smoke directly inhaled
by the smoker, is highly concentrated submicron aerosol
typically generated in discrete 1 - 3 s puffs. In the
literature a wide range of size distributions for
mainstream smoke have been reported. This range of
distributions is in part due to difficulties in the sampling
and measurement of such highly concentrated aerosols
and the various conditions under which the cigarette
smoke is generated.
As part of an ongoing effort to develop a fast, accurate
and reliable method to measure the particle size
distribution of mainstream cigarette smoke, a sampling
inlet for a cascade impactor capable of puffing on a
cigarette has been developed. The inlet, which generates
the mainstream smoke puffs and introduces them directly
into the sample stream of the impactor, minimizes the
effects of evaporation and coagulation by minimizing the
time between generation and collection of the impactor
substrates.
This puff sampling inlet has been tested with a MicroOrifice Uniform Deposition Impactor (MOUDI). The
particle size distribution from several industry monitor
and reference cigarettes was studied using a 35 cc, 2 s
puff with an approximately sinusoidal puff profile. The
mass median aerodynamic diameters (MMADs) of the
smoke aerosols were found to be between 0.3 and 0.4
microns. These diameters agreed well with the MMADs
calculated from particle size distributions of single puffs
collected with a scanning mobility particle sizer. The
mass concentrations of the various cigarettes determined
with the MOUDI were in reasonable agreement with the
FTC tar measurements of the various cigarettes,
suggesting that evaporation effects were minimal.
Furthermore the MMAD of the cigarettes varied with the
dilution of the cigarette as predicted by a simple
coagulation model.
12C.4
Shape selection of aerosol particles using electrostatic
classifiers. RAJAN K. CHAKRABARTY, Hans Moosm
Aerosol shape influences aerosol chemical and physical
properties and therefore is very important for the
commercial use of aerosols including applications in
medicine and nano technology. However, it is difficult to
control aerosol shape and this parameter is rarely used to
modify other aerosol properties. Here, we discuss a novel
technique of shape selection of aerosols using
electrostatic classifiers. This technique was tested on
fractal-like soot particles generated with a fuel-rich
laminar premixed ethene flame. Electrostatic classifiers
size select nanoparticles based on their electric mobility,
which is related to the particle mobility diameter Dp
divided by the particle charge q. The soot particles were
size selected first with a prevalent diameter Dp = 220 nm
and a prevalent charge q = 1. A second setup selected soot
particles with a prevalent diameter Dp = 220 nm and a
prevalent charge q = 2. Structural and fractal analysis of
soot particles transmitted for q =1, Dp = 220 nm and for
q =2 Dp = 220 nm were performed. For these identical
Dp, it was found that particles transmitted with q = 2 were
more elongated in shape than those transmitted with q =
1. The q = 2 particles had a lower aspect ratio and lower
mass-fractal dimension. This result will be explained and
discussed in detail. Our method of shape selection using
electrostatic classifiers should be applicable to all nano/
micro particles and should be useful to tailor physical and
chemical particle properties to specific applications.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12C.5
Aerodynamic Focusing of Aerosol Particles Through a
Micro-Nozzle: Modeling and Experiment. JUSTIN HOEY,
Iskander Akhatov, Orven Swenson, Doug Schulz, North Dakota
State University.
New techniques for aerosol delivery use nozzles in the
micro-scale regime. To accurately determine aerosol
trajectories in a micro-nozzle (100 micro-meter inner
diameter), a model applying the physics governing
aerosols has been devised and verified experimentally.
An examination of a micro-nozzle with silver-ink aerosol
droplets flowing at a velocity on the order of 100 m/s was
completed. Laser illumination perpendicular to the flow
at the exit of the nozzle was used to visualize particle
streaks in the aerosol beam. Experimental results show a
focal point of aerosol particles on the order of 1 mm from
the exit of the nozzle. The beamwidth of aerosol particles
at the focal point was minimized to less than 5 micrometers. An analytical approximation of the focusing of
aerosol particles using MATLAB and a computational
model using Ansys CFX are compared to the
experimental data. A physics model that approximates
the focusing behavior of the aerosol particles is discussed.
Along with the model, a method of improving the
focusing technique is presented.
12C.6
A Mobile Air Quality Monitoring Trailer for Developing
Countries, First Results. T. PETAJA, L. Laakso, H. Laakso,
P.P. Aalto, T. Pohja, E. Siivola, P. Keronen, S. Haapanala, M.
Kulmala, University of Helsinki, Finland; H. Hakola, Finnish
Meteorological Institute, Finland; N.Kgabi, M. Molefe, D.
Mabaso, J.J. Pienaar, The North-West University, Republic of
South Africa; E. Sjoberg, M. Jokinen, Department of
Agriculture, Conservation and Environment, Mafikeng,
Republic of South Africa.
Southern African savannah background environment
lacks continuous long-term combined sub-micrometer
aerosol number concentration and air quality
measurements. We built a mobile measurement trailer,
which contains measurements of aerosol number size
distributions from 10 to 840 nm in diameter, positive and
negative air ion size distributions from 0.4 to 40 nm,
aerosol mass concentration PM10, PM2.5, PM1,
inorganic aerosol composition in PM10 and PM2.5,
selected trace gases (SO2, NO, NOx, CO, O3) and basic
meteorological parameters. In savannah we periodically
we measured VOC concentrations as well. Prior
deploying the trailer to the field, we compared the mobile
setup against instruments in a state-of-the-art air quality
research station in HyytiГѓВ¤lГѓВ¤, Southern Finland. The
trailer is well-protected against thunder storms, electricity
breaks and other such problems. It is connected to the
internet via a GPRS-modem, which enables automatic
data backup to a server and remove monitoring of trailer
performance. Preliminary results from a period July, 23
to August, 15, 2006 showed that in a clean savanna
environment, new particle formation took place every
sunny day, with relatively high formation rates and very
-1
high growth rates (up to 15 nm h ). A typical new particle
formation event produced 1-3E4 cm-3 new particles in
sizes above 10 nm. The formation rate J10 was
approximately 1 cm-3 s-1. Typical night-time concentration
of submicron number concentration was approx. 1000 cm
-3
. Average mass concentrations were 12.8, 14.7 and 24.4
micro grams m -3 for PM1, PM2.5 and PM10, respectively.
The VOC concentrations were generally quite low as the
sum of monoterpene concentrations varied from 0.3 to 1.6
micro-grams m-3. The VOCs observed included e.g. alfapinene, beta-pinene, benzene, nonane and isoprene.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12D.1
Molecular Dynamics simulations of the size dependence of
deliquescence in atmospheric nano-particles : Effect of
surface tension. RANJIT BAHADUR, Lynn M. Russell,
Scripps Institution of Oceanography, UCSD.
Studies on water uptake by atmospheric sea-salt particles
in the nano-size regime show a strong dependence of both
the deliquescence relative humidity (DRH) and
hygroscopic growth factor on particle size below 50 nm.
Ionic aerosols exist as crystals covered by a liquid layer
rather than as the anhydrous phase, and contain both
solid-liquid and liquid-vapor interfaces, which strongly
influence hygroscopic properties. Surface tension
measurements for these liquid interfaces may have a high
degree of accuracy, however corresponding properties for
solid interfaces are only inferred and contain large
uncertainties.
Sodium chloride is the major constituent of atmospheric
particles. NaCl-water-air provides a simplified model
system for which upper bounds of the surface tensions are
calculated from Molecular Dynamics simulations. The
calculations are based on energy differences between each
bulk phase and the interfaces. At 1 atm and 300 K, the
calculated upper bounds for interfacial tensions are 107
mN/m for NaCl-air, 59 mN/m for NaCl-solution, 89 mN/
m for solution-air, and 73 mN/m for water-air. These
upper bounds have uncertainties between 5% and 10%,
which are slightly higher than measurements in liquid
interfaces, but reduce measurement uncertainty for solid
interfaces by as much as a factor of 15. A more
comprehensive integral-based calculation of surface free
energy for single-component phases is used to estimate
surface tension values for comparison. The upper bounds
are also simulated for nano-particles, and related to the
bulk values using Tolman lengths, which are all of the
order of 0.1 nm and positive in the model system.
Incorporation of the calculated upper bounds in a bulk
thermodynamic model significantly improves agreement
between predicted and measured hygroscopic behavior.
Sensitivity studies indicate that while DRH is strongly
dependent on the surface tensions, the curvature
dependence has a minor effect in the 5-20 nm size range.
12D.2
Homogeneous Nucleation in the Ozone - Alpha-pinene
Reaction studied by tunable vacuum UV Photoionization
Mass Spectrometry. ERIN R. MYSAK, Michael P. Tolocka,
Tomas Baer, University of North Carolina; Paul J. Ziemann,
University of California Riverside; Eric Gloaguen, Kevin R.
Wilson, Musahid Ahmed, Lawrence Berkeley National
Laboratory.
Atmospheric reactions of ozone with biogenic terpenes such as
alpha-pinene contribute significantly to global amounts of
secondary organic aerosol (SOA). SOA is a result of gas-toparticle conversion of organic compounds, and recent work by
various other laboratories has suggested that the reaction
products responsible for the nucleation are oligomeric or
polymeric in nature, although the specific species responsible
for nucleation are not yet identified. In the current study, the
particle size (2- 90nm) and composition of particles (30-90nm)
formed by this gas-to-particle conversion process are studied
simultaneously as a function of reaction time. One of the main
novelties of this experiment is that particles in this small size
range are studied in real-time, which is possible with the
combination of both delicate and continuous vaporization and
ionization sources. Due to the delicate nature of the analysis
technique, the current study is among very few in which alphapinene ozonolysis products are detected in the entire range from
43-402 m/z. Compositional analysis of the aerosol particles
indicates that over 540 different molecular products are formed.
Furthermore, the envelopes of peaks observed in the high mass
range (m/z < 300) support the conclusion that a large variety of
precursors react in various combinations to yield a wide array of
complex, multifunctional oligomeric products. In order to probe
the reaction pathways leading to the formation of the large
oligomeric species formed in alpha-pinene ozonolysis,
experiments were preformed in low versus high relative
humidity conditions (to effect the stabilized criegee intermediate
(SCI) channel), in the presence and absence of a hydroxyl
radical scavenger (to affect the hydroperoxide channel). The
presence of water, which acts as a SCI scavenger, affects the
induction time for nucleation to a small extent, but does not
affect the product formation either compositionally or in terms
of aerosol mass formation. By scavenging the hydroxyl radical,
which is a high yield product of a-pinene ozonolysis, we observe
changes in the composition of the high m/z products formed and
in the yield of aerosol mass produced. These results indicate
that SOA is formed primarily through the hydroperoxide
channel, and that water vapor can enhance nucleation (probably
through clustering), but does not impact SOA growth.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12D.3
Laboratory-Measured Nucleation Rates of Sulfuric Acid and
Water from the SO2 + OH Reaction. SHAN-HU LEE, David
R. Benson, Kent State University.
Particle nucleation is an important step in the chain
reactions that lead to cloud formation, but the nucleation
mechanisms are poorly understood. Current nucleation
theories are not rigorously tested by experiments and
large uncertainties exist in their predictions even over
many orders of magnitude. Whereas there are hundreds
of publications on nucleation models, laboratory studies
of nucleation are very limited because of the various
difficulties associated with nucleation experiments.
Binary homogeneous nucleation of H2SO4/H2O is the
most basic nucleation system, yet its nucleation
mechanisms are not well-studied. Here, we present
results of the laboratory study of this binary nucleation
system. H2SO4 was produced through the reaction of
SO2 + OH -> HSO3 in the presence of SO2, OH, O2, and
H2O in a fast flow reactor at 288 K and atmospheric
pressure. OH was produced from the photolysis of water
vapor. The power dependence of nucleation rate (J) on
sulfuric acid concentration ([H2SO4]) was 2 - 10 in the
[H2SO4] range from 3e6 - 1e9 cm-3. This power
dependence increased with decreasing RH and increasing
nucleation time. The power dependence of J on RH was
10 - 15 for the RH values from 10 - 50 %. The measured
aerosol sizes ranged from 4 - 20 nm. These aerosol sizes
were larger for higher [H2SO4], higher RH, and higher
nucleation times. The effects of RH on aerosol growth
were also more pronounced at higher [H2SO4] and with
higher nucleation times. These results will provide
important information that can be used to test and
improve the nucleation theories. We will also discuss
these results by comparing with previous studies by other
investigators.
12D.4
Measurements of Homogeneous Nucleation Rates of nalcohols in a Supersonic Nozzle by Small Angle X ray
Scattering. BARBARA WYSLOUZIL, The Ohio State
University; David Ghosh, Reinhard Strey, University of
Cologne, Germany.
In our earlier publication[1,2] we presented the results of
axial pressure measurements characterizing the
condensation behaviour of a series of n alcohols (CnH(2n
+1)OH, n = 3 to 5) in a supersonic nozzle. Although we
determined the temperature T, the condensible partial
pressure pv, and the characteristic time Delta_t,
associated with the peak nucleation rate, we were only
able estimate the nucleation rate J = N / delta_t, because
we could not directly measure the number density N of
the aerosol. Here, we present the results of our flow rate
resolved Small Angle X ray Scattering (SAXS)
experiments that characterize n alcohol droplets at a fixed
position in a supersonic nozzle with a comparable
expansion rate. By fitting the radially averaged scattering
spectrum, we obtain information on the mean radius r, the
width of the size distribution sigma, and the particle
number density N of the aerosol. The variation of these
parameters with the alcohol flow rate deepens our
understanding of the growth behaviour of nano droplets
during supersonic nozzle expansion. By combining the
data from both sets of experiments, we find that our
measured nucleation rates are by factors of 2 to 11 higher
than our previous estimates.[1,2] Finally, we use Hale's
[3,4] scaling formalism to compare our experimental
nucleation rates to those available in literature.
[1] Gharibeh, M., Kim, Y., Dieregsweiler, U., Wyslouzil,
B. E., Ghosh, D., Strey, R., J. Chem. Phys. 122, 94512
94521 (2005).
[2] Nucleation of alcohols in supersonic nozzles,
Gharibeh, M., Wyslouzil, B. E., Kim, Y., Ghosh, D.,
Strey, R., Oral presentation AAAR 2004 Annual Meeting,
Atlanta, Georgia.
[3] B. Hale, Phys. Rev. A 33, 4256 (1986).
[4] B. Hale, Metall. Trans. A 23, 1863 (1992).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
12D.5
Heterogeneous Nucleation on Single Microdroplets. ASIT K.
RAY, James L. Huckaby, University of Kentucky.
12D.6
Impurity Effect On A Nucleation Rate Of Single Vapor.
LYUBOV ANISIMOVA, Binghamton University.
We have examined heterogeneous nucleation on single
droplets that were suspended in an electrodynamic
balance mounted inside a thermal diffusion cloud
chamber. The droplets were exposed to vapors of various
immiscible compounds, and the vapor concentration
around a droplet was increased from unsaturated to
supersaturated levels in small steps by altering the
temperature difference between the top and bottom plates
of the chamber. A resonance based light scattering
technique was used to detect formation of a second phase
in the droplets. We have examined dioctyl phthalate
(DOP) droplets exposed to water vapor, Santovac droplets
(i.e., five ring polyphenyl ether) to invoil 90 (i.e., straight
chain alkane) vapor, hexadecane vapor and Fomblin (i.e.,
perfluorinated polyether) vapor. Results show that
when a droplet is exposed to an environment that is
supersaturated with an immiscible vapor, heterogeneous
nucleation occurs through three possible ways: by
dropwise condensation, nucleation inside the droplet, and
nucleation leading to the formation of a layer. High
surface tension compounds (e.g., water vapor on DOP
droplets) do not form layers. We also observed that when
a Santovac droplet is exposed to invoil 90 vapor, above a
certain supersaturation level a second phase forms by
nucleation inside the droplet phase, and nuclei grow to
form an emulsion in the droplet. Nucleation of low
surface tension compounds on a droplet results in the
formation of a layer, and transpires in the absence of any
significant supersaturation. The results show that the
formation an adsorbed layer depends on the surface and
interfacial tensions. For example, hexadecane layers form
on Santovac droplets at hexadecane saturation ratios in
the range S=1.005 to 1.02. Below S=1.005, a droplet
absorbs hexadecane vapor, while the droplet remains
homogeneous. After the formation of a layer on a droplet,
the layer initially grows slowly, but the growth rate
increases as the layer thickness increases. Above a certain
thickness, the layer behaves like a macroscopic phase
whose growth rate can be described by diffusion
equations.
Currently the requirements for nucleated vapour purity
are studied, but a impurities effect is not fully understood.
If one has a high purity vapour sample then this sample
contains not less than trillion admixture molecules per cc.
Is it too mach or no? Seemingly all vapors under
investigation are presenting multi-component systems.
Fortunately the current experimental results illustrates a
relatively low sensitivity of nucleation rate on an
admixture contains. For example, Strey et al. [1] found
experimentally that 10 percents of a second component
shift nucleation rate less than 10 orders of magnitude. I.e.
one percent of admixtures produce practically
undetectable changes in water-alcohol nucleation rate [1].
Seemingly the current state of nucleation experiment
accuracy allows the admixtures contain 0.5 percent and
less. Obviously that any activation of chemical activity
or/and a molecule (atom) radical state can rise nucleation
sensitivity enough to detect each ion practically. A
nucleation rate surface for binary ideal solution will be
presented to illustrate an impurity effect on a nucleation
rate of single vapour.
1. R. Strey, Y.Viisanen, and P.Wagner, J. Chem. Phys.
1995, 103, 4333
[email protected]
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
13D.1
Interaction of Gas-Phase Nitric Acid and Primary Organic
Aerosol in the Atmosphere of Houston, TX. Luke Ziemba,
ROBERT GRIFFIN, Casey Anderson, Jack Dibb, Sallie
Whitlow, University of New Hampshire; Barry Lefer, James
Flynn, Bernhard Rappenglueck, University of Houston.
Concentrations of aerosol and gas-phase pollutants were
measured on the roof of an 18-story building during the
Texas Air Quality Study II Radical and Aerosol
Measurement Project from August 15 through September
28, 2006. Aerosol measurements included size-resolved,
non-refractory mass concentrations of ammonium, nitrate,
sulfate, chloride, and organic aerosol in submicron
particles using an Aerodyne quadrupole aerosol mass
spectrometer (Q-AMS). Particulate water-soluble organic
carbon (PWSOC) was quantified using a mist chamber/
total organic carbon analysis system. Concentration data
for gas-phase pollutants included those for nitric (HNO3)
and nitrous acids (HONO), collected using a mist
chamber/ion chromatographic technique, oxides of
nitrogen (NO x), collected using a chemiluminescent
method, and carbon monoxide (CO), collected using an
infrared gas correlation wheel instrument.
Aerosol speciation was dominated by organics and
sulfate, constituting over 87% of the observed aerosol
mass on average. Coincident increases in nitrate and
organic aerosol concentrations occurred, generally during
morning rush hour as confirmed by the observed levels of
CO and NOx. Based on the lack of organic aerosol
processing (defined by the strength of the signal at m/z =
57 and the weakness of the signal at m/z = 44 in the QAMS spectra), the spikes in organic aerosol were likely
associated with primary organic aerosol (POA). During
these events, PWSOC also showed no strong increase,
underscoring the lack of organic aerosol processing. Gasphase HNO3 concentration decreases were simultaneous
with the increases in Q-AMS organic and nitrate aerosol
and gas-phase HONO concentrations. These data likely
indicate uptake of HNO3 and potential heterogeneous
formation of HONO involving POA. Q-AMS chloride
data indicated that strong acids other than HNO3 may also
partition to POA. This phenomenon was not observed
during the evening rush hours, though it was observed at
random non-rush hour times at other points during the
campaign.
13D.2
The Impact of Organic Coatings on the Heterogeneous
Hydrolysis of N2O5: Interaction of Atmospheric Transport
and Chemistry. NICOLE RIEMER, Stony Brook University;
Heike Vogel, Bernhard Vogel, Forschungszentrum Karlsruhe;
Tatu Anttila, Finnish Meteorological Institute; Thomas F.
Mentel, Astrid Kiendler-Scharr, Forschungszentrum Juelich.
Both tropospheric photochemistry and the formation of
particulate nitrate depend critically on the budget of
nitrogen oxides (NOx). The NOx budget in turn is tied to
nocturnal N2O5 reaction taking place in the aqueous
phase of aerosols. Through this reaction, NOx is removed
from the atmosphere and HNO3 is formed, which then
partitions between the particle and the gas phase. The
efficiency of the heterogeneous N2O5 hydrolysis is
quantified by the reaction probability gamma. In recent
years, laboratory measurements have shown that gamma
depends on the aerosol particle composition. While for
ammonium sulfate a value of gamma ~ 0.02 applies,
gamma is found to decrease by up to one order of
magnitude if aerosols contain nitrate or they are coated
with organics.
With this study we investigate to what extent a decrease
in gamma due to the presence of nitrate and/or organic
coatings can influence photochemistry and nitrate
formation in the ambient atmosphere. We present results
of regional scale model simulations for Europe with the
comprehensive model system LM-ART. To treat the
N2O5 hydrolysis, we use recent results from laboratory
studies that quantify gamma depending on nitrate content
and organic coating thickness.
The results show that the impact of the N2O5 hydrolysis
depends crucially on the characteristic development of
vertical profiles of gas phase and aerosol phase species in
the nocturnal boundary layer. At a given time, the
maxima of N2O5 and aerosol surface area profiles
typically do not coincide in space. This consequently
leads to smaller reaction rates compared to what a simple
box model treatment would give. Moreover, the
importance of an organic coating depends on the amount
of nitrate in the aerosol. In a low-NOx environment the
differences in e.g. NO3 concentrations without and with
organic coating are up to 60 %, whereas in a high-NOx
environment those differences are negligible.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
13D.3
Humidity and Nitric Acid Effects on Particle Formation for
Monoterpene Ozonolysis Using the Nanometer Aerosol Mass
Spectrometer. KATHERINE J. HEATON, Murray V.
Johnston, University of Delaware.
The goal of this study is to better understand how
water and acidity affect the formation of secondary
organic aerosol (SOA) particles from the reaction of
alpha- and beta-pinene with ozone. Alpha- and betapinene are endocyclic and exocyclic monoterpenes,
respectively, which represent a sizeable portion of the
global biogenic emissions and that produce SOA.
However, it is not completely understood how other
species, such as acidity (represented by nitric acid) and
water, affect the chemical reactions leading to aerosol
formation. In this study, the particles formed during the
early stages of growth are analyzed with the Nanometer
Aerosol Mass Spectrometer (NAMS).
A flow tube reactor was used in this study to combine
the monoterpenes with the ozone and water vapor or nitric
acid. The reaction times were varied between 9s and 23s.
On-line analysis was performed with the Scanning
Mobility Particle Sizer (SMPS) to acquire size
distributions and NAMS acquired elemental
compositional data for the particles. NAMS is a new
technique developed by our lab, in which aerosol particles
in the size range of 10nm to 25nm are trapped and ablated
with a high energy laser to produce positively charged
atomic ions. Previous work in our laboratory showed
that the C:O ratios of monoterpene SOA ranged from 1.9
to 2.7. Preliminary results for the current study suggest
that the C:O ratio is smaller in the presence of water
compared to dry air, which could indicate that either
water is being incorporated into the particles formed or
that the distribution of organic functional groups in the
particles have changed. Future work will include off-line
analysis of molecular components to understand these
changes at the molecular level.
13D.4
Modeling and Computation of Thermodynamic Equilibrium
for Mixtures of Aerosol Inorganic and Organic Species.
Neal Amundson, ALEXANDRE CABOUSSAT, Jiwen He,
Andrey V. Martynenko, University of Houston; John H.
Seinfeld, California Institute of Technology.
The computation of thermodynamic equilibrium of
atmospheric aerosols containing mixtures of inorganic
and organic compounds is a crucial issue in air quality
modeling and climate prediction.
We present here a model (UHAERO), that is flexible and
efficiently computes the thermodynamic equilibrium of
atmospheric particles containing inorganic and organic
compounds. It is applied to mixtures of inorganic
electrolytes and dicarboxylic acids, and to mixtures that
include inorganic electrolytes and liquid-liquid phase
separation between aqueous and organic phases. The
model does not rely on any a priori specification of the
phases present in certain atmospheric conditions.
The multicomponent phase equilibrium for a closed
organic aerosol system at constant temperature and
pressure and for specified feeds is the solution to the
equilibrium problem arising from the constrained
minimization of the Gibbs free energy. An accurate and
efficient method for the computation of the minimum of
energy allows to compute the equilibrium state and phase
diagrams for mixtures of inorganic and organic species.
The Gibbs free energy is modeled by a hybrid model,
namely the CSB approach with the UNIFAC model for
the organic compounds, the PSC model for the inorganic
constituents and a Pitzer model for interactions.
Numerical results show the efficiency of the model in
computing phase diagrams for mixtures of inorganic
electrolytes and organic acids. Preliminary results for
mixtures of inorganic and organic species are presented
demonstrating the influence of liquid phase separation on
the salt crystallization.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
13D.5
Secondary Organic Aerosol (SOA) Formation from Reaction
of Isoprene with NO3 Radicals. NGA LEE NG, Arthur Chan,
Puneet Chhabra, Jason Surratt, Richard Flagan and John
Seinfeld, California Institute of Technology.
Recent studies have shown that the photooxidation of
isoprene leads to the formation of secondary organic
aerosol. Isoprene is the most abundant non-methane
hydrocarbon so even a small aerosol yield may have a
large effect on both local and global SOA production. In
this study, aerosol formation from the reaction of isoprene
and NO3 radicals is investigated. Experiments are carried
3
out in the Caltech dual 28m environmental chambers,
with T=20-22В°C and RH <10 %. Aerosol growth is
monitored by differential mobility analyzers (DMAs) and
an aerosol mass spectrometer (AMS) is used to
characterize the aerosols formed. Aerosol samples are
also collected on Teflon filters for offline chemical
analysis. The SOA yields from these experiments will be
presented, and preliminary aerosol composition data will
be discussed.
13D.6
What Controls the Relative Abundance of Organic and
Sulfate Aerosol Mass in the Northeastern United States?
CHARLES BROCK, Joost de Gouw, Adam Wollny, NOAA
Earth System Research Laboratory; Rodney Weber, Rick
Peltier, Georgia Institute of Technology; Amy Sullivan,
Colorado State University.
Measurements of aerosol particle size distributions and
composition and of trace and reactive gas mixing ratios
were made on the National Oceanic and Atmospheric
Administration (NOAA) WP-3D aircraft downwind of
mixed urban and industrial sources in the northeastern U.
S. Most particulate organic matter (OM) originated from
secondary production in urban plumes, while sulfate was
formed from sulfur dioxide emitted by industrial point
sources near and within the urban areas. When submicron
particle mass concentrations exceeded 15 micrograms per
cubic meter, sulfate and associated ammonium dominated
the composition; at lower mass concentrations OM
dominated. The apportionment of aerosol mass between
OM and sulfate compounds was governed largely by the
differences between the oxidation timescales of sulfur
dioxide and those of precursor volatile organic
compounds (VOCs) relative to their transport time. Even
with higher-than-expected secondary OM formation, in
mixed urban/industrial plumes in the northeastern U.S.,
the potential inorganic particulate mass represented by
sulfur dioxide emissions significantly exceeds the
potential OM from anthropogenic VOC emissions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
14D.1
The connection between symmetry and the polarization state
of scattered light. MATTHEW J. BERG, Christopher M.
Sorensen, Amit Chakrabarti, Kansas State University.
The scattering of light from a small particle can be used to
measure the size, shape and refractive index of the
particle and hence constitutes an important unintrusive
characterization method in aerosol science and industry.
Most measurements collect the intensity of polarized light
scattered in different directions to infer characteristics of
the particle. An often overlooked ``hidden variable," is
the polarization state of the scattered light which can also
communicate information about the particle. This work
studies how the shape of a particle influences the behavior
of the polarization state in different directions. It is
shown that the symmetry of the particle shape is a major
factor controlling the evolution of the polarization state
with direction. A polarization scheme is described that
exploits this symmetry to provide the possibility for
differentiating between a variety of particle shapes.
14D.2
Relative Humidity Influence on Aerosol Light Absorption
and Scattering by Biomass Burning Aerosol. W. Patrick
Arnott, Kristin Lewis, Guadalupe Paredes-Miranda, Stephanie
Winter, University of Nevada, Reno; Derek Day, National Park
Service; Rajan K. Chakrabarty, Antony Chen, Hans
Moosmueller, Desert Research Institute.
A very interesting case of smoke aerosol with very low
single scattering albedo, yet very large hygroscopic
growth for scattering is presented. Several samples of
chamise (Adenostoma fasciculatum), a common and often
dominant species in California chaparral, were recently
burned at the USFS Fire Science Laboratory in Missoula
Montana, and aerosol optics and chemistry were
observed, along with humidity-dependent light scattering,
absorption, and particle morphology. Photoacoustic
measurements of light absorption by two instruments at
870 nm, one on the dry channel, one on the humidified
channel, showed strong reduction of aerosol light
absorption with RH above 65 percent, and yet a strong
increase in light scattering was observed both at 870 nm
and 550 nm with nephelometers. Multispectral
measurements of aerosol light absorption indicated an
Angstrom coefficient for absorption near unity for the
aerosols from chamise combustion. It is argued that the
hygroscopic growth of scattering is due to uptake of water
by aerosol with significant inorganic salt composition.
Furthermore, the reduction of aerosol light absorption is
argued to be due to the collapse of chain aggregate
aerosol as the RH increases wherein the interior of aerosol
does no longer contribute to absorption. Connections
with latent heat absorption by moist aerosol and its effects
on photoacoustic measurements will also be discussed in
relation to the reduction of aerosol light absorption.
Implications for biomass burning in general are that
humidity processing of aerosols from this source and
others like it tends to substantially increase its single
scattering albedo, probably in a non-reversible manner.
The chemical pathway to hygroscopicity will be
addressed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
14D.3
Comparison of Measured and Calculated Scattering from
Aerosols at the Surface Using Three Size Distribution
Instruments (PCASP, SMPS, UHSAS) and Nephelometers.
YONG CAI, Derek C. Montague, Wiesje Mooiweer, Terry
Deshler, University of Wyoming.
Tropospheric aerosols were sampled at the surface for two
weeks in summer and winter in Laramie, Wyoming, and
for two weeks in summer on Elk Mountain (3,320 m),
near Laramie. Aerosol size distribution, scattering, light
absorbance, composition, and hygroscopicity were
measured. Here we focus on comparing the measured
aerosol scattering with calculated scattering from size
distribution measurements. The overall goal is to clarify
the important factors which affect the retrieval of
scattering from particle size information. Aerosol size
distributions were measured with a passive cavity aerosol
spectrometer probe (PCASP, diameter: 110 - 3000 nm), a
scanning mobility particle sizer (SMPS, 16 - 800 nm), and
an ultra high sensitivity aerosol spectrometer (UHSAS, 55
- 800 nm). Aerosol scattering was measured with two
single wavelength nephelometers and one three
wavelength nephelometer. To complete the scattering
calculations two types of particle refractive indices were
used, a 24-48 hour average from filter measurements, and
a size and time dependent index from particle
composition measurements using an Aerodyne aerosol
mass spectrometer (AMS). Time-dependent aerosol
scattering, calculated from the measured size distributions
and the two types of refractive indices, are compared to
the measured time dependent scattering at four different
wavelengths. Preliminary work indicates a linear
relationship between measured and calculated scattering.
The uncertainties introduced by using 24-48 hour
averages for refractive indices versus the highly resolved
time and size dependent indices from the AMS are
discussed. Comparisons between the scatterings
calculated from the different aerosol size distribution
measurements will also be presented. Size independent
imaginary indices of refraction were also estimated using
an iterative calculation and the measured size
distributions. The indices of refraction are optimized for
the best match between measured and calculated
absorption. Preliminary calculations suggest reasonable
agreement of scattering calculations and measurements;
however, that consistency significantly depends on the
size-resolved particle refractive indices.
14D.4
Electrical Mobility of Aerosol Nanowires: Theory and
Experiment. Soo Kim, Pusan National University; GEORGE
MULHOLLAND, Michael Zachariah, University of Maryland.
This study is motivated by the need for a rapid
measurement of the length of nanowires. A theoretical
model has been developed to describe the behavior of
nanowires undergoing Brownian rotation in an electric
field in the free molecular limit. The probability of a
given orientation is proportional to a Boltzmann
expression containing the potential energy associated with
the orientation dependent torque in the electric field. The
potential energy includes the contribution from the singly
charged nanowire and from the polariability of the
nanowire. The polarizability potential energy is
approximated as the value for a prolate ellipsoid with the
same volume and aspect ratio as the nanowire. The
orientation probability is a Gaussian with a normalization
factor involving the imaginary error function.
For comparison with experiment, the nanowire system
chosen is the carbon nanotube which is easily synthesized
in the aerosol phase (S.H. Kim and M.R. Zachariah J.
Phys. Chem. B., 110, 4555, 2006). The transit time of
CNTs through the DMA are computed using the
orientation probability and the CNT drag force in the free
molecular limit. This analysis accounts for the electric
field dependence of the electric mobility of the CNT. The
theory predicts that small CNTs will be randomly
oriented, larger CNTs up to about 400 nm in lengths with
a 15 nm diameter will have a broad distribution of
orientations with the free charge the dominant energy
term, and the longest CNTs will be essentially totally
aligned in the direction of the field with the polarizability
term dominating as a result of its dependence on the
electric field squared. The predicted CNT lengths will be
compared with measured results for multi-walled CNTs
(Nanotechnology 16, 2149, 2005) and for single-walled
CNTs (D. Tsai, NIST and Moisala et al., Carbon 43,
2066, 2005).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
14D.5
On the Role of the Electric Field in the Scale-up of the
Electrospray in High-Density Microfabricated Multiplexed
Systems. WEIWEI DENG, Alessandro Gomez, Yale
University; Chris Mike Waits, Nick Jankowski, Bruce Geil,
Army Research Laboratory.
The electrospray has been traditionally plagued by low
flow-rate limitations that impeded its application to fields
other then mass-spectrometry. Multiplexing by
microfabrication was shown to be a promising approach
to remedy this drawback, as demonstrated in silicon
multiplexed electrospray atomizers operating at a packing
density of 250 sources per square centi-meter using an
extractor electrode configuration. This presentation
discusses the role and limitations that the space charge
electric field can impose on the scale-up of the
electrospray. Experimentally, we found that under the
action of the space charge field and because of their high
charge-to-mass ratio, the satellite droplets experience
flight reversal, with the droplets accumulating on the
extractor plate and eventually short-circuiting the
electrodes. Proximity with other electrosprays in
multiplexed systems exacerbates this phenomenon. To
establish a criterion for maximizing the density (#
sources/area) of a multiplexed electrospray system by
accounting for space charge limitations, first we
developed a computational approach of the space charge
field of a single electrospray using a Lagrangian model.
The model faithfully reproduced the shape of the
observed electrospray plume. Because of the prohibitively
long computational time that would have been required to
extend this model to a multiplexed system, an analytical
alternative was identified using a line-of-charge
approximation, yielding results in good agreement with
those of the full model. An extension to multiple lines of
charge was applied to model the multiplexed system and
an algebraic expression of the system behavior in the limit
of an infinitely large array of electrospray sources was
derived, identifying the current emitted per unit area and
the droplet residence time as key variables. Experimental
results corroborated these findings.
14D.6
Charge-to-mass Ratio of Progeny Droplets Produced by
Coulombic Fissions. Harry H. Hunter, ASIT K. RAY,
University of Kentucky.
During evaporation the charge density of a droplet
reaches a level where the electrostatic repulsion forces
overcome the surface tension force, and the droplet
becomes unstable. The instability causes disintegration of
the droplet, leading to the formation of small progeny
droplets. This phenomenon, referred to as a Coulombic
fission, reduces the droplet charge below the instability
limit. The charge limit at the instability is predicted by the
Rayleigh limit. No theory currently exists for the
prediction of charge and mass losses at a Coulombic
fission, or the characterization of progeny droplets. The
experimental data on charge and mass losses show wide
variability. Recent empirical analysis and experimental
data suggest droplets of low conductivity liquids
disintegrate into a few large droplets through "rough
fission" modes, while breakups of droplets of high
conductivity liquids proceed under "fine fission" modes
with the formation of large number of fine droplets with
significantly lower mass losses. In this study we have
examined charge-to-mass ratios of progeny droplets
produced by fissions of droplets of varying conductivity.
Experiments were conducted on single droplets that were
suspended in an electrodynamic balance. An optical
resonance based light scattering technique was used to
determine the size and the size change of a droplet at
fission. The charge level and the charge loss were
obtained from the dc voltages required to gravitationally
balance the droplet prior to and following the breakup. In
this study we have examined diethyl phthalate and
dimethyl phthalate droplets, whose electrical
conductivities were increased by dissolving either
tridodecylmethylammonium chloride or
tridodecylmethylammonium nitrate. The results show
that the fractional charge loss increases, while the mass
loss decreases as the electrical conductivity increases,
thus the charge-to-mass ratio of progeny droplets
increases with conductivity. An extrapolation of the data
indicates that at very high conductivities the charge-tomass ratio approaches the limit of an ion.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
16D.1
Particle Resuspension in Turbulent Flow: A New
Theoretical Model. Allison Harris, CLIFF DAVIDSON,
Carnegie Mellon University.
We describe a new model for the resuspension of particles
from exposed soil surfaces. This model, based on first
principles, is successfully able to particle motion under a
range of wind and meteorological conditions.
Our model treats aerodynamic forces in a theoretical
manner accounting for lift, drag, and interparticle
adhesion under turbulent flow conditions. By numerically
solving a system of second order differential equations,
we are able to calculate trajectories for the full range of
real-world particle sizes from particles in the submicron
range through particles that are several hundred microns
in diameter. We address the quantification of the lift
force, which persists as one of the major barriers to
accurate resuspension prediction. We also describe a new
quantification for the adhesive effect of soil moisture as a
function of rainfall rate, time since last rainfall, and
humidity.
For wind flow fields we account for near-surface coherent
structures such as turbulent bursts that can entrain surface
particles carrying them out of the quasi-laminar sublayer.
This is treated stochastically using a joint probability
distribution function for turbulent wind fluctuations in
two-dimensions.
This model tells us much about the relative importance of
the physical mechanisms governing particle resuspension.
Results indicate that the lift force is more important than
previously thought, and may be a dominant factor under
some conditions. Additionally, we have determined that a
drag-induced torque rather than lift is the factor likely
responsible for overcoming adhesion. We furthermore
improve upon the current understanding of the role of
capillary forces in particle adhesion.
16D.2
An Approach to Analytically Model Diffusional
Nanoparticle Deposition under Low Pressure Conditions.
CHRISTOF ASBACH, Heinz Fissan, Institute of Energy and
Environmental Technology (IUTA); Jing Wang, David Y.H.
Pui, University of Minnesota.
The manufacture of the next generation of semiconductor
chips requires a very clean environment and a low
pressure level around 50 mTorr (6.7 Pa) or even below.
The deposition of nanoparticles can be a major concern
for the success of new lithography tools. Photomasks are
particularly vulnerable because common pellicles can no
longer be used to protect the masks. A fully analytical
model to describe the inertial particle motion under low
pressure conditions has been established by the authors in
the past. Diffusional deposition was not included in this
model, but estimation revealed that under low pressure
conditions diffusional nanoparticle contamination can be
a major risk for the cleanliness of the photomasks. Based
on the existing inertia model and the subsequent
estimation of the diffusional particle deposition, an
analytical extension of the model has been developed that
determines the risks of nanoparticle contamination at low
pressure under the influence of gravity, electrophoresis,
thermophoresis and drag force. As a first approach the
model assumes that inertial and diffusional transport can
be decoupled and treated separately. The inertial transport
is solved first in the shape of the particle stopping
distance. It is assumed that the diffusional contribution to
the stopping distance is negligible, which is justified in
most cases, where the stopping distance is much larger
than the simultaneous diffusional displacement. The
location where the particle stops is taken as the starting
point for the diffusion model. When no external forces act
on the particle, it can move in any direction with the same
probability, i.e. it is somewhere within a spherical plume.
When external forces, such as gravity, electrophoresis
and/or thermophoresis act on the particle, the shape of the
plume is changed to an ellipsoidal shape.
The approach for the model will be presented along with
initial comparisons with experimental results.
This work was supported by the Deutsche
Forschungsgemeinschaft (DFG).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
16D.3
Effects of Inter-Particle Collisions and Two-Way Coupling
on Particle Deposition Velocity in a Turbulent Channel
Flow. HOJJAT NASR, Goodarz Ahmadi, John B. McLaughlin,
Clarkson University.
This study is concerned with the effect of particleparticle collisions and two-way coupling on particle
deposition velocity in a turbulent channel flow. The time
history of the instantaneous turbulent velocity vector was
generated by the two-way coupled direct numerical
simulation (DNS) of the Navier-Stokes equation via a
pseudospectral method. The particle equation of motion
takes the Stokes drag, the Saffman lift, and the
gravitational forces into account. The effect of particles
on the flow is included in the analysis via a feedback
force on the grid points. Several simulations for different
particle relaxation times and particle mass loading were
performed, and the effects of inter particle collisions and
two-way coupling on the particle deposition velocity,
fluid and particle fluctuating velocities, particle normal
velocity, and particle concentration profile were
discussed. It was found that when particle-particle
collisions are considered in the calculation, the particle
deposition velocity increases. When the particle feedback
force is taken into account (two-way coupling), the
particle deposition velocity slightly decreases. When both
inter-particle collisions and two way coupling are taken
into account (Four-way coupling), the particle deposition
velocity increases. The present simulation results were
compared with the available experimental data and earlier
numerical results.
16D.4
Anomalies in the Evolution of Particle Size Distributions.
JAMES W. GENTRY, University of Oklahoma.
Two general methods have been developed for examining
the effect of coagulation (or breakage) kernels on the
evolution of particle systems. One method based on
generating functions was originally developed by
Smoluchowski and used more recently by Alex
Lushnikov in examining the properties of self-preserving
distributions. The second method is based on numerical
simulations of the distribution and employs the criterion
that the logarithmic moments of the distributions evolved
from self-preserving kernels will approach constant
asymptotic values. Particularly we consider the third
logarithmic moment which is a measure of skewness ( the
second logarithmic moment is the standard deviation).
For a lognormal distribution the second moment is the
standard deviation and the third moment is zero.
Among the anomalous and significant behaviors that have
been observed from the simulations include: (1) some
kernels do not give self-preserving models with the
second and third logarithmic moments diverging, (2) In
some cases all of the volume moments converge to an
asymptotic moment, (3) To obtain the behavior shown
by atmospheric distributions it is necessary to add a
continuous source, (4) The behavior of the distributions
evolved in time depend both on the nature of the kernel
but also the initial distribution. We develop a test for
examining this behavior based on a bimodal distribution
consisting of two lognormal distributions in which each of
the initial lognormal distributions have the same volume.
(5) We believe that the third moment is a significant
indicator because it normally undergoes a maximum
before converging to an asymptotic value. Our
simulations led us to conclude that it would be necessary
to know the moments with greater precision than can be
measured experimentally in determine the degree of
homogeneity in the collision kernels.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Aerosol Chemistry and Physics
2007 AAAR Annual Conference Abstracts
16D.5
Thermophoretic Effect on Scavenging of Brownian Particles
by a Condensing Knudsen Droplet. MAREK SITARSKI,
Husson College.
The cross-effects, such as thermophoresis and
diffusiophoresis, occuring in the vicinity of colliding
aerosol particles are demonstrated to be important factors
in predicting kinetics of Brownian coagulation. The
coagulation kernel is affected directly by the concurrent
non-equilibrium phenomena: condensation/evaporation,
and the associated heat release/absorption. In some cases,
like for example marine aerosols at the deliquescence
point, additional heat sources due to intraparticle reactions
and/or phase transitions need to be considered. As a result
of these processes, the formation of non-unifom
temperature and vapor concentration profiles around the
condensing droplets affects the rate of Brownian
coagulation. The presented theoretical analysis is
formulated in terms of Brownian motion of the colliding
particles in the presence of hydrodynamic forces:
diffusiophoretic (due to the Stefan flow of the condensing
vapor) and thermophoretic (due to the heat flow). The
corresponding steady-state Fokker-Planck equation with
the net hydrodynamic force is solved by the Grad's
moment method at conditions close to the thermodynamic
equilibrium. As results from this theoretical analysis, the
thermophoresis moderates the large diffusiophoretic
enhancement of the scavenging rate of small particles by
a condensing droplet. The specific calculations are
performed for marine fog droplets at the deliquescence
point capturing nanometer size (primary) soot particles.
16D.6
Algorithm Based on Self-Organizing Map for Classification
of New Particle Formation Events. HEIKKI JUNNINEN,
Ilona Riipinen, Miikka Dal Maso, Markku Kulmala, University
of Helsinki, Finland.
Continuous measurements of aerosol size distribution are
coming more and more common, but we are missing an
objective way to classify and compare the data measured
from different places, by different instruments and
groups. In this work we are presenting an algorithm that
can be used to classify aerosol size distributions in
automatic manner from large data sets. First we
emphasized the problem of separating automatically days
with the regional new particle formation events (RNPF)
and days without particle formation. The performance of
the algorithm was verified against manually made
classification, and a very good agreement was found.
Although, the new algorithm can provide more complex
classification here we present the results of classification
into three classes, RNPF events, non-events and the days
that don't fit to previous two classes.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
2L.1
Application of Multivariate and Trajectory-Based Receptor
Models to Regional Source Apportionment in the Eastern U.
S. JOHN G. WATSON, Douglas H. Lowenthal, L.-W. Antony
Chen, Darko Koracin, David Dubois, Desert Research Institute;
Naresh Kumar, Eladio Knipping, EPRI; Neil Wheeler, Stephen
Reid, Sonoma Technology, Inc.
2L.2
Simulating IMPROVE-like Data for Use in Evaluating
Receptor Models. NEIL J. M. WHEELER, Kenneth J. Craig,
Stephen B. Reid, Erin K. Gilliland, Sonoma Technology, Inc.;
Naresh Kumar, Eladio Knipping, EPRI; Douglas H. Lowenthal,
L.-W. Antony Chen, John G. Watson, Darko Koracin, Desert
Research Institute.
The ability of receptor models to estimate regional
contributions to PM2.5 was assessed with synthetic,
speciated data sets at Brigantine National Wildlife Refuge
(BRIG), NJ and Great Smoky Mountains National Park
(GRSM), TN. IMPROVE-style concentrations were
generated for summer, 2002 using the Community
Multiscale Air Quality (CMAQ) model. The 205 source
categories accounting for 95% of the emissions of PM2.5,
SO2, NOx, VOC, CO, and NH3 emitted in the eastern U.
S. were matched to 43 speciated source profiles from the
EPA's Speciate and DRI's PM source profile databases.
CMAQ estimated the actual regional contributions to
species concentrations and individual source contributions
to primary PM2.5 at both sites. The positive matrix
factorization (PMF) receptor model was applied to
simulated and actual IMPROVE PM2.5 data sets at BRIG
and GRSM. Seven-factor solutions were found for each
site, explaining ~99% of the variability in the data sets. At
BRIG, the receptor model captured the first four major
contributing sources (including a secondary sulfate
factor), although diesel and gasoline vehicle contributions
were not separated. At GRSM, however, the resolved
factors did not correspond well to major PM sources.
Instead, minor PM sources such as aluminum processing
and industrial manufacturing we identified. There was no
clear correlation between the factors and the true regional
contributions to sulfate. The trajectory mass balance
regression model (TMBR) was used to apportion sulfate
concentrations to the 7 source regions. Ambient sulfate
concentrations were regressed on the number of 1-hour
HYSPLIT trajectory endpoints in each region. The largest
estimated sulfate contributions at both sites were from the
local regions; this agreed qualitatively with the "true"
regional apportionments. Estimated regional contributions
depended on the starting elevation of the trajectories.
Receptor models have been widely used for source
apportionment of atmospheric particulate matter (PM) to
aid in developing effective emission reduction strategies.
To assess the ability of receptor models to estimate
regional contributions to particulate matter less than 2.5
microns (PM 2.5 ) simulated IMPROVE data at Brigantine
National Wildlife Refuge (BRIG) New Jersey and Great
Smoky Mountains National Park (GRSM) Tennessee
were created for summer 2002 using the Community
Multiscale Air Quality (CMAQ) model. In preparation
for the modeling 205 source categories accounting for
95% of the emissions emitted in the eastern United States
were matched to 43 unique speciated source profiles. The
CMAQ model was modified to include 69 additional
aerosol species: 43 PM2.5 tracers, one for each unique
source profile; and 26 PM2.5 tracers representing the
species reported in IMPROVE data. A three-month base
CMAQ simulation was performed on a 12-km resolution
domain in the eastern United States and hourly speciated
PM2.5 concentrations were extracted from the model
output and evaluated. The base simulation provided
simulated IMPROVE data at BRIG and GRSM and
information about PM2.5 contributions from each of the 43
unique source profiles. To determine the regional
contributions to PM2.5 seven additional simulations were
performed; each simulation reduced emissions in one of
seven unique source regions within the modeling domain.
The results for each of these simulations were compared
to the base simulation and the resulting differences used
to determine the contribution of each region to PM2.5 at
BRIG and GRSM. The resulting data sets were provided
to the study team for use in testing the ability of
multivariate receptor models such as positive matrix
factorization (PMF) UNMIX and principal component
analysis (PCA) and trajectory-based models such as
trajectory mass balance regression (TMBR) to resolve
regional sources of PM2.5 .
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
2L.3
Identification of Source Regions of Aerosols in the Eastern
Mediterranean Atmosphere by Exploiting Receptor
Oriented Models. FATMA OZTURK, University of Maryland;
Gurdal Tuncel, Middle East Technical University.
2L.4
Impacts of Plug-in Hybrid Electric Vehicles on Regional
Haze and PM. UARPORN NOPMONGCOL, John Grant,
Alison Pollack, Greg Yarwood, ENVIRON; Eladio Knipping,
Mark Duvall, Charlie Clark, EPRI.
The primary objective of this study is to find the
contribution of the sources on the observed levels of
particulate pollution of Eastern Mediteranean atmosphere.
For this purpose, continous daily aerosol samples were
collected at a station located on the Mediterranean coast
of Turkey (30.34 E, 36.47 N) between March 1992 and
February 2001 with PM-10 Hi-Vol aerosol sampler
(Sierra Andersen Model SAUV-10H) on Whatman 41
cellulose fiber filters. Energy Dispersive X-Ray
Fluorescence (EDXRF) and Inductively Coupled Mass
Spectrometry (ICP-MS) were the analytical techniques
employed to find the trace element content of the
collected aerosol samples. 55 trace constituents from Li to
U were determined by these techniques. Major anions,
SO42-, NO3-, and Cl-, in the samples were measured by
using a VYDAC 302 Ion Chromatography (IC) anion
exchange column. A non-parametric bootstraped Potential
Source Contribution Function (PSCF) was calculated and
results were interpolated with a GIS software, namely
MapInfo, and Positive Matrix Factorization (PMF), a
new type of factor analysis method, was applied to the
generated data set to identify the source regions of the
pollutants affecting Eastern Mediterranean aerosol
composition. Preliminary results have demonstrated that
there are no significant long term variations in the
concentrations of elements, which are the markers of
crustal and marine origin, like Al, Ca, K, Na and Cl. On
the other hand these elements have showed well defined
seasonal variation. The combination of PMF and PSCF
results imply that aerosols have been affected by both
local sources, this is especially true for the crustal and
marine originated elements, and by distant sources from
which pollutants have been transported to the region.
A plug-in hybrid electric vehicle (PHEV) uses gridelectricity to power a significant fraction of driving
resulting in decreased vehicle emissions but increased
electrical generation (utility) emissions. PHEVs offer
advantages of increased energy independence/flexibility
and, potentially, reduced greenhouse gas emissions.
However, vehicle and power plants both contribute to
regional haze, particulate matter (PM) and other air
pollution impacts. Evaluating the air quality impacts of
PHEVs is complex because emissions reductions/
increases occur at different locations (e.g., urban vs.
rural), different times (e.g., rush hour vs. nighttime) and
for different pollutants. This paper uses a current
generation, three-dimensional aerosol model (CMAQ) to
project future PM and visibility conditions with and
without PHEVs. Continental-scale modeling is
performed for North America to assess the broad scale
impacts, and then finer-scale modeling is performed for
both California and Ohio to evaluate impacts in greater
detail for areas with different types of air-quality
problems and different mixtures of emission sources.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
2L.5
Regional Air Quality-Atmospheric Nucleation Interactions.
JAEGUN JUNG, Peter J. Adams, and Spyros N. Pandis,
Carnegie Mellon University (S.N. Pandis also University of
Patras, Patra, Greece).
The creation of new atmospheric particles from in-situ
nucleation influences climate through cloud-aerosol
interactions and may negatively impact human health.
Although recent observations show that nucleation is
widespread over most continents, the corresponding
pathways remain uncertain. A computationally efficient
multi-component aerosol dynamics model (DMAN) that
simulates the full aerosol size distribution and
composition starting at a diameter of 0.8 nm has been
developed. Several proposed nucleation rate expressions
for binary (H2SO4-H2O), ternary (NH3-H2SO4-H2O),
and ion-induced nucleation are evaluated using DMAN
against ambient measurements from the Pittsburgh Air
Quality Study. The ternary NH3-H2SO4-H2O nucleation
model is successful in predicting the presence or lack of
nucleation on nineteen out of nineteen days with complete
datasets in July 2001 and on twenty-five out of twentynine days in January 2002. DMAN has been added to the
three-dimensional Chemical Transport Model PMCAMxUF and is tested in the Eastern United States. Reductions
of ammonia emissions are predicted to decrease the
frequency of nucleation events during both summer and
winter, with a more dramatic effect during the summer.
The response to changes in emissions of sulfur dioxide
during the summer is counterintuitive. Reductions of
sulfur dioxide and the resulting sulfate by up to 40%
actually increase the frequency of the summer nucleation
events. Modeling predicts the opposite effect in winter,
with reductions of sulfur dioxide leading to fewer
nucleation events.
2V.1
Experiments and Modelling on the Behaviour of Ruthenium
Oxides at High Temperature. TEEMU KARKELA, Ulrika
Backman, Ari Auvinen, Yuko Enqvist, Riitta Zilliacus, Maija
Lipponen, Tommi Kekki, Unto Tapper, Jorma Jokiniemi, VTT
Technical Research Centre of Finland; Jorma Jokiniemi,
University of Kuopio; Jouko Lahtinen, Helsinki University of
Technology.
The release of highly radiotoxic fission product isotopes
of ruthenium is a particular concern in the event of a
severe nuclear accident. In order to verify, whether
ruthenium would be transported within a nuclear facility,
it is of interest to know how its volatile oxides are formed
and how they behave.
Ruthenium behaviour was studied by conducting 23
experiments, in which either RuO2 powder was oxidised
in a high temperature tube furnace or gaseous RuO4 was
fed to the furnace. Gas flow through the furnace was set
to 5 l/min (NTP). Gas composition was a mixture of air,
argon and steam.
Upon cooling of the gas flow, a large fraction of
gaseous ruthenium oxides decomposed to RuO2 particles
or reacted on the tube surface. Aerosol particles were first
filtered from the gas and gaseous RuO4 was then trapped
in 1 M NaOH-water solution. Mass of ruthenium aerosol
and gas was determined with instrumental neutron
activation analysis. The particles were analysed also with
scanning mobility particle sizer, electron microscopy,
electron diffraction and X-ray photoelectron
spectroscopy. Ruthenium transport kinetics and
deposition profile were measured by conducting some
experiments with gamma active 103Ru isotope.
Ruthenium release rate increased from 0.11 to 25.4
mg/min as temperature increased from 1100K to 1700K.
Especially at high temperature, release rate decreased
with decreasing oxygen partial pressure. The fraction of
released ruthenium transported as aerosol particles
increased with furnace temperature from 0.6% to 35%.
Formation of aerosol particles competed with
decomposition of RuO3 on the surface. As much as 70%
of the released ruthenium was transported in gaseous form
through the facility at 1300K. At higher temperature
gaseous ruthenium transport decreased close to zero.
Retention of gaseous ruthenium on stainless steel surface
was significant at temperature close to 100 degrees
Celsius.
This work is supported by Finnish Research Programme
on Nuclear Power Plant Safety (SAFIR2010), Fortum
Nuclear Services Ltd (FNS) and the Nordic Nuclear
Safety Research (NKS-R).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
7E.1
Regulatory Decision Making using Advancements in Aerosol
Science. RALPH MORRIS, Bonyoung Koo, Bo Wang, Greg
Yarwood, ENVIRON International Corporation; Gail Tonnesen,
Chao-Jung Chien, UC Riverside; Dennis McNally, Greg Stella,
Alpine Geophysics.
The Regional Haze Rule (RHR) has the ambitious goal of
natural visibility conditions at Class I areas by 2064.
Class I areas include specific national parks, wilderness
areas and wildlife refuges throughout the U.S. To achieve
this goal, a RHR State Implementation Plan (SIP) is due
in December 2007 that, among other things, demonstrates
Reasonable Progress by 2018 toward natural visibility
conditions in 2064. This paper discusses how recent
advances in aerosol science are used in the current
generation of three-dimensional aerosol models to project
future-year visibility impairment at Class I areas to
demonstrate reasonable progress toward natural
conditions. Based on recent measurement and laboratory
studies, the representation of Secondary Organic Aerosol
(SOA) formation is improved in both the CMAQ and
CAMx models. The enhanced models are then used to
simulate current conditions, which includes a model
performance evaluation and conditions in the future-year
(2018) accounting for emissions growth and control. The
models are then used to justify emissions controls on
many sources that lead to reduced PM at Class I areas and
visibility improvements. The paper concludes with
examples of the expected visibility improvements at Class
I areas from the RHR SIP modeling.
7E.2
Regional Modelling of PM2.5: Case Study for the Po Valley
(Italy). GIOVANNI LONATI, Giovanni Sghirlanzoni, Andrea
Zanoni, DIIAR - Politecnico di Milano Guido Pirovano.
The work discusses the results of a model simulation for
PM2.5 over a mesoscale domain by comparing modelled
PM2.5 bulk and speciated mass concentration to
observations.
The Po valley is a flat area about 400-km long and 100km wide extending over the great part of Northern Italy.
The area is bordered by high mountain chains (The Alps
and The Apennines) which reduce air masses circulation
and favour pollutants stagnation. In the main cities of the
area very high PM10 concentration levels are observed
during the whole year and air quality standard (AQS) are
frequently not attained. Currently PM2.5 is not
extensively monitored since not regulated by AQS, but
some data, also speciated for the main chemical
components, are available for Milan area.
CAMx (Comprehensive Air quality Model with
eXtensions) model was used for simulation with a 10 x 10
square km spatial resolution based on emission and
meteorological data for the 01/04/2003-31/03/2004
period. Emission data include criteria pollutants, NMVOC
and related chemical speciation, PM size distribution and
chemical speciation.
Model outputs are the concentration fields of PM2.5
concentration and of its main components (nitrate,
sulphate, ammonium, elemental carbon and organic
matter). Comparison between model results and daily
measurements is performed for the city of Milan, where
the highest PM2.5 concentrations are predicted, both for
PM2.5 mass and for single chemical components. Since
model results are not concurrent with the available
measurements, a qualitative comparison considers the
distributions of the modelled and observed
concentrations. Bootstrapping technique is applied in
order to extract from model results restricted warmseason and cold-season data populations suitable for
quantitative comparison with measured data.
Underestimation of PM2.5 mass is observed in both
seasons, essentially due to a large underestimation of
organic matter concentration, whereas a rather good
agreement is observed for the other species.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
7E.3
Strengths and Limitations of Multivariate Receptor Models:
Experiments with Simulated Regional-Scale PM2.5 Data.
L.-W. Antony Chen, Douglas H. Lowenthal, John G. Watson,
Darko Koracin, Desert Research Institute; Naresh Kumar,
Eladio Knipping, EPRI; Neil Wheeler, Stephen Reid, Sonoma
Technology, Inc.
Receptor models have been widely used for source
apportionment of atmospheric particulate matter (PM) to
aid in developing cost-effective emission reduction
strategies. All receptor models solve the chemical mass
balance (CMB) equation. Multivariate receptor models,
including UNMIX and PMF (positive matrix
factorization), identify factors in a chemically-speciated
dataset and determine source profiles and contributions
simultaneously. The solutions, however, are often not
unique without imposing additional constraints to limit
the inherent rotational ambiguity in these models. Though
multivariate analyses have been gaining in popularity, the
degree to which the resolved factors correspond to realworld pollution sources and strategies for improving
model performance are rarely assessed because the actual
source contributions are not known. In this study, the
Community Multi-scale Air Quality (CMAQ) modeling
system coupled with a state-of-the-art emission inventory
and dispersion scheme was used to create speciated
PM2.5 datasets at two remote IMPROVE sites in the
eastern U.S. The simulated data include "true"
contributions to chemical species from 7 geographical
regions and to primary PM2.5 from 43 different source
categories. Datasets representing averaging times of 6 to
24 hours served as input to the PMF and UNMIX models.
Following general guidance for selecting the number of
factors and rotational parameters, PMF and UNMIX
resolved unique factors, but these were not necessarily the
related to the major contributors to PM2.5 mass. Sources
with similar chemical profiles or that were collocated or
collinear in space were often merged into the same factor.
Secondary species such as ammonium sulfate were not
effectively apportioned to sources of their precursors.
Instead, the models determined exclusively secondary
factors. Separation of local and regional contributions
may be better achieved with solutions using fewer factors.
The influence of PMF model uncertainty (measurement
and conceptual) and data averaging time will be
discussed.
7E.4
Evaluation Receptor Models with Synthetic IMPROVE
Data. DOUGLAS LOWENTHAL, Antony Lung-Wen Chen,
John Watson, Darko Koracin, Dave Dubois, Desert Research
Institute; Naresh Kumar, Eladio Knipping, EPRI; Neil Wheeler,
Stephen Reid, Sonoma Technology, Inc.
The USEPA Regional Haze Rule (RHR) seeks to reduce
haze in Class 1 areas to natural background levels by
2064. Receptor models will be used by states to devise
control strategies to meet this goal. Urban-scale receptor
models attempt to identify individual sources such as
vehicles, incinerators, power plants, and industrial
activities. Because most large-scale regions contain
similar sources, identifying local and distant regional
sources is a qualitatively different problem. A study was
designed to test the ability of multivariate receptor models
such as PMF (positive matrix factorization), UNMIX, and
PCA (principal component analysis) and trajectory-based
models such as TMBR (trajectory mass balance
regression) to resolve regional sources of PM2.5 sulfate at
two eastern U.S. national parks: Brigantine National
Wildlife Refuge, NJ and Great Smoky Mountains
National Park, TN. Hourly synthetic IMPROVE aerosol
concentrations in the eastern US during summer, 2002,
were generated with the SMOKE/MM5/CMAQ modeling
system. Forty-three source profiles were chosen to
represent PM2.5 emissions from 106 source categories in
the study domain, which was divided into 7 sub-divisions
of the MANE-VU, VISTAS, MIDWEST, and CENRAP
Regional Planning Organizations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
7E.5
Variable Moment General Dynamic Equations for Global
and Regional Aerosol Modeling. BORIS GRITS, Anthony
Wexler, University of California, Davis.
Particles influence a wide range of atmospheric processes
such as cloud and fog formation, light scattering and
absorption, visibility, and health effects. Each of these
processes depends on one or more moments of the
particle size distribution and composition. Most
atmospheric particle dynamics models simulate one
moment of the distribution, typically mass or number, and
may infer other moments by making uncertain but
computationally necessary assumptions about particle
morphology, density, and composition. Substantial
uncertainties may also be introduced moving from mass
to number or vice versa since these moments are related
by the cube of the particle diameter, thereby amplifying
diameter uncertainty substantially. Finally, uncertainties
may be introduced due to sharp gradients in the particle
size distribution due to numerical artifacts introduced
during integration of these equations.
In the present study we propose a novel form of the
multicomponent general dynamic equation that simulates
arbitrary moments of the distribution. First we regard the
case of a pure condensation/evaporation particles growth.
A general equation for the evolution of r-th moment of
aerosol size distribution is derived that has the same form
for any r (both integer and non-integer), allowing us to
obtain both number and mass distributions from a single
function while minimizing moment transformation errors.
This function includes both these distributions and
optionally an intermediate smoothing component that
reduces numerical error. To test our algorithm we perform
simulations of aerosol population growth in several cases
and compare these results with those obtained using other
schemes.
Next we add coagulation, production and loss terms to the
equation and express them in the r-th moment form.
Again, the scheme proposed is tested against several
existing schemes.
11J.1
Ultrafine particles from boreal wildfires: Long range
receptor estimates of emission factors and rates. KEITH
BEIN, Yongjing Zhao, Anthony Wexler, University of
California Davis; Murray Johnston, University of Delaware.
During the Pittsburgh Supersite experiment, unprocessed
emissions from large scale wildfires in the boreal forest of
Quebec, Canada, heavily impacted the site for two
separate periods between July 6th and 8th, 2002.
Measurements of composition-resolved, number-based
particle size distribution (PSD), PM2.5 mass, EC and OC,
and NOx and CO were used to estimate size-distributed
particle number emission factors (EFs), a total particle
number EF of (1.2 +/- 0.2)E15 particles/kg biomass,
ultrafine particulate mass (PM0.1) EF of 0.18 +/- 0.04 g
PM0.1/kg biomass, 2.6 +/- 0.4 g PM2.5/kg biomass, 1.3
+/- 0.3 g PM2.5 OC/kg biomass, 0.6 +/- 0.2 g PM2.5 EC/
kg biomass, 0.2 g PM2.5 K/kg biomass and a NOx EF (as
NO) of 7 +/- 1 g NO/kg biomass. These EFs exhibit a
good degree of self-consistency in terms of what is known
about the emissions of high-severity crown fires.
Estimates of total direct carbon emissions taken from the
literature were used in conjunction with the EFs above to
estimate annual emissions from boreal forest fires in
Alaska, Canada, Russia and the entire circumpolar boreal
forest for select years. Results from these calculations
were compared to estimates of annual global emissions
for the same species from all wildland fires to elucidate
the relative contribution of boreal forest fires. During the
high fire year of 1998, boreal forest fires emitted an
estimated 8E26 particles, 0.9 Tg PM2.5 OC, 0.4 Tg
PM2.5 EC and 4.8 Tg NOx (as NO), which represent ~
14%, 6%, 22% and 30% of global wildland fire
emissions, respectively. Projections of increasing fire
activity in boreal forests under a warming climate, as well
as the indirect effects of this increase on carbon dynamics
and balance, is likely to increase the global relevance of
these ecosystems as a direct source of emissions and as a
key variable in the climate change system.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.2
Phreatomagmatic to Magmatic: The Evolution of Aerosol
Size and Composition during the 2006 Eruptions of
Augustine Volcano. CATHERINE CAHILL, University of
Alaska Fairbanks; Thomas Cahill, DELTA Group,University of
California, Davis; Jonathan Dehn, Stephen McNutt, Ken Dean,
Peter Webley, University of Alaska Fairbanks.
Augustine Volcano, an island stratovolcano 275 km
southwest of Anchorage, Alaska, erupted on January 11,
2006. Additional explosive events occurred on January
13-14, 17 and 28, 2006. The volcano then entered a
period of continuous ash and pyroclastic flow emission
that lasted from January 29 to February 4, 2006.
An 8-stage DRUM aerosol impactor sited in Homer,
Alaska, from January 13 to February 11 collected sizeresolved (35-5, 5-2.5, 2.5-1.15, 1.15-0.75, 0.75-0.56, 0.56
-0.34, 0.34-0.26 and 0.26-0.09 microns in aerodynamic
diameter) and time-resolved (3-hour resolution) aerosols
from Augustine Volcano. The aerosol samples were
analyzed for aerosol mass by beta-gauge and elemental
composition by synchrotron x-ray fluorescence.
Preliminary results show a dramatic shift in aerosol size
and composition during the transition from the 13
phreatomagmatic (water and magma) explosive events in
January to the continuous magmatic (only magma)
emission period at the end of January and the beginning
of February. For example, crustal elements such as Al,
Si, Fe, Ca, etc. are present in high concentrations in the
largest size fraction but low concentrations in a smaller
size fraction (0.75-0.56 microns) during the
phreatomagmatic events. However, during the magmatic
period, the concentrations of these elements in the large
size fraction decreased, but greatly increased in the
smaller size fraction. The composition of the aerosols
also changes during this transition. Sodium, which is
present in high concentrations in the large size fraction
during the phreatomagmatic events is present in lower
concentrations during the magmatic period. These shifts
in size distribution and composition imply a change from
rock-breaking explosive events to open-throat fresh
magmatic emissions, and a systematic shift from high to
low ascent rate.
11J.3
Variation of Perceived Visibility with Aerosol Optical
Property in the Urban Area of Seoul, Korea. KYUNG W.
KIM, Gyeongju University, Korea; Young J. Kim, Gwangju
Institute of Science and Technology, Korea; KYUNG W. KIM,
Gyeongju University; Jinsang Jung, Young J. Kim, Gwangju
Institute of Science and Technology; Taesik Kim, Gyeongju
University, Jaeyong Ryoo, Korea Institute of Environmental
Science and Technology.
Extensive aerosol, optical, and scenic monitoring were
conducted at the urban site of Seoul, Korea in order to
investigate the relationship between aerosol optical
properties and perceived visibility degradation.
Elemental, ionic, and carbonaceous species were analyzed
on the samples of PM2.5 and PM10 observed during the
extensive monitoring period. Light extinction, scattering,
and absorption coefficients were measured
simultaneously using a transmissometer, a nephelometer,
and an aethalometer, respectively. The chromatic
parameters of the color difference were calculated from
the scenic images using the HSI color difference method.
Perceived visibility varied with light attenuation due to
aerosol components, particle size distribution, and
ambient relative humidity. The hue difference, saturation
difference and the intensity difference calculated from the
scenic images showed different characteristics according
to aerosol optical properties.
This paper will discuss the changes observed in aerosol
size and composition during the transition from
phreatomagmatic events to open-throat magmatic
emission during the 2006 eruptions of Augustine
Volcano.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.4
Estimation of the source contributions from long range
transport to particulate matters in Seoul, Korea. KYESEON KIM, Jong-Bae Huh, Hyun-Sun Kim, Seung-Hee Kim,
Yong-Seok Seo, Bora Choi, Eun-Mi Choi, Seung-Muk Yi,
School of Public Health, Seoul National University.
11J.5
Characterization of Ambient Aerosol in Summer and
Winter in a Small Urban Setting and in Summer at a
Remote Mountaintop Site. DEREK C. MONTAGUE, Mariya
M. Petrenko, Wiesje Mooiweer, Yong Cai, Terry Deshler,
University of Wyoming.
Korea has been interested in the degradation of air
quality by the long range transport (LRT) since Korea is
mostly downwind from China. The objective of this study
was to investigate episodes of long range transported
pollution by looking at the enhancement ratios between
species such as ?PM2.5 /?CO and ?PM10/?CO in Korea.
Real-time monitoring using TEOM-FDMS was
performed to measure the mass concentration of PM2.5 for
every 30 minutes from September 2004 through
December 2005. The concentrations of PM10, SO2, NO2,
O3, and CO were obtained from ambient air monitoring
system at Hyojea-dong operated by Seoul City. In this
study, there were many well-defined periods of
enhancements in CO and PM. The high concentration
episodes (average 24 hours PM 2.5 more than 65
3
microgram per meter ) were observed totally 33 times
during the sampling period. The concentration trends of
PM2.5 and PM10 during those episode were not similar to
the typical patterns of CO concentration that shows high
concentrations during the traffic rush hours indicating that
those episodes were affected mostly from LRT sources
rather than local sources. The previous studies also
suggested that the CO enhancement ratios could be used
to identify the source contributions from LRT compared
to local sources. Almost a half of the episodes were
associated with high CO enhancement ratios of ?PM2.5 /?
CO and ?PM10/?CO above 0.03 microgram per meter3 per
ppbv indicating that LRT can be attributed to about 50%
of PM episodes in Korea.
Physical and chemical characteristics of dried ambient
aerosols have been measured during both summer and
winter in a small urban setting (Laramie, Wyoming:
LAR) and in summer at a remote clean mid-continental
mountaintop (Elk Mountain, 3,320 m: EMO). Size
distributions were measured with a Scanning Mobility
Particle Sizer (SMPS) (0.015 - 0.8 micrometer), a Passive
Cavity Aerosol Spectrometer Probe (PCASP) (0.11 - 3.0
micrometer), and an Aerodynamic Particle Sizer (APS)
(0.5 - 20 micrometer), augmented by an Ultra High
Sensitivity Aerosol Spectrometer (0.055 - 1.0 micrometer)
at EMO. Long period (24 - 48 hour) averages of
particulate composition were obtained using parallel
channel 47 mm twin-filter pack systems for TSP, and, by
employing a 1 micrometer cut cyclone, for PM1. Size and
time resolved (5 min. average) particle composition was
acquired by an Aerodyne Aerosol Mass Spectrometer
(AMS). Characterized aerosols had low PM1 mass
3
loadings, particularly at EMO (~3.7 microgram/m ) and in
3
winter in LAR (~2.3 microgram/m ), and were dominated
by organics (43 - 67%). Larger particles (PM<1),
composed mainly (<75%) of involatile mineral material
contributed ~70% of the TSP mass in LAR, and ~35% at
EMO. Optimal estimates of both the organic matter/
organic carbon (OM/OC) mass ratio and the collection
efficiency of the AMS were derived by combining the
filter and mass spectrometer data, assuming mass closure.
Similarly, independent AMS collection efficiency
estimates were derived from comparisons of SMPS/
PCASP/APS size distributions with those simultaneously
obtained from the AMS. Deconvolution of the mass
spectra of organic aerosols using the Zhang et al.'s (2005)
algorithm shows that whereas hydrocarbon-like organic
aerosol (HOA) is essentially absent at EMO, both HOA
and oxygenated organic aerosol (OOA) are present at
LAR in continuously varying amounts. Comparison of
HOA, OOA, and inorganic ionic species size distributions
suggest that OOA and sulfate in the accumulation mode
are mostly internally mixed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.6
Aerosol Number and Volume Concentrations During the
Rocky Mountain Nitrate and Sulfate Study (ROMANS).
EZRA LEVIN, Gavin McMeeking, Christian Carrico, Jeffrey
Collett, Jr., Sonia Kreidenweis, Colorado State University;
William Malm, National Park Service.
The Rocky Mountain Nitrate and Sulfate Study
(ROMANS) was conducted in Rocky Mountain National
Park in two phases during the spring and summer of 2006.
An aerosol sizing system was used to measure dry
number and size distributions of particles having
diameters ranging between 0.04 and 20 Вµm at 15-minute
time resolution. An alignment technique was applied to
these data to reconcile the differences between output
from an optical particle counter and a differential mobility
analyzer, and the refractive index corresponding to the
best-aligned size distribution recorded. The retrieved
refractive index during spring was slightly higher than
that retrieved during the summer. We computed dry
aerosol scattering coefficients from the size distribution
and refractive index data and compared them to scattering
coefficients measured by a nephelometer collocated at the
site, for conditions when the ambient relative humidity
was below 20%. Aerosol volume concentrations,
calculated from size distributions assuming spherical
particles, correlated with 24 hour filter measurements of
aerosol mass concentrations with an R squared of 0.75.
We have compared time series of aerosol concentrations
from the two periods to distinguish differences in aerosol
volume and number concentrations and variability, and
fine mode volume fraction. From this analysis it can be
seen that variations in the aerosol during the spring period
are dominated by episodic transport events, while the
variations in summer are dominated by diurnal patterns.
Also, higher aerosol concentrations, especially in the fine
mode, were observed in the summer period. These time
series are also statistically compared to time series of
wind direction to elucidate meteorological influences on
particle concentrations at the site.
11J.7
AMS measurements at Melpitz supersite (Germany) during
winter 2007. LAURENT POULAIN, Gerald Spindler, Thomas
Gnauk, Erika Bruggemann, Birgit Wehner, Hartmut Herrmann,
Leibniz-Institute for Tropospheric Research.
During January and February 2007, an Aerodyne Aerosol
Mass Spectrometer was deployed at the Melpitz supersite
(Germany) to measure non-refractory (NR) aerosol
composition (ammonium, nitrate, sulphate, chloride, and
total organics). The measurement place is considered as
an urban influenced rural site.
The preliminary results show that generally nitrates
represent the most important mass fraction of the aerosol.
The time profiles of the specific markers of oxygenated
organic species (m/z 44) and hydrocarbon species (m/z
57) show that the organic mass fraction of the aerosols is
mainly composed of oxygenated compounds during all
the measurement time.
In parallel to AMS measurement, aerosols were also
monitoring by SMPS-APC and daily filters sampling.
Comparison with these measurements and the impact of
the air mass origin to the aerosol composition will also be
presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.8
Characterization Of Particulate Matter Along A North.
PIERRE HERCKES, Jenny Cox, Kandis Knight, Nabin
Upadhyay, Panjai Prapaipong; Arizona State University; Rainer
Lohmann, University of Rhode Island; Luca Nizzetto,
University of Insubria.
In recent years an increasing number of studies have
investigated aerosols in urban and remote locations.
However, little data on particulate matter especially
carbonaceous aerosols in the remote marine environment
exists.
We will present results on trace metal and organic aerosol
characterization performed on samples collected along a
North/South Transect from Bremerhaven (Germany) to
Capetown (South Africa) aboard the FS Polarstern
research vessel. Total Suspended Particulate matter (TSP)
filter samples were analyzed for a variety of trace metal
species including Pb, Fe, Zn as well as for total carbon
and carbon isotope ratios, using high resolution
inductively coupled plasma mass spectrometry (HR-ICPMS) and isotope ratio mass spectrometry. Individual
organic species (including n-alkanes, n-alkanoic acid,
polyaromatic hydrocarbons, hopanes, steranes,
dicarboxylic acids, and levoglucosan) were quantified in
discrete (12h) and pooled (24-72h) samples, following
solvent extraction and analysis by gas chromatography
coupled to mass spectrometry (GC/MS).
Our results indicate low concentrations of carbonaceous
material during most of the research cruise (~1micrograms/m3) with higher concentrations in less pristine
areas, mostly along Europe. Carbon isotope ratios were
variable (-20 < delta-13C < -27) and consistent with a
change from C3 to C4 vegetation, although effects from
algae cannot be excluded. Hopanes and cholesterol were
nearly universally detected at very low concentrations. A
sample contamination appears highly unlikely and the
results seem to indicate a ubiquitous background
concentration in the case of hopanes while algae might
contribute to the ubiquity of cholesterol. Biomass burning
markers retene and levoglucosan were occasionally
detected. As for trace metals, high iron concentrations
were observed consistent with a dust event. A detailed
composition analysis as a function of air mass history will
be presented.
11J.9
Three Years Measurement of sulfate at Okinawa, Japan in
Spring Period. AKINORI TAKAMI, Xiaoxiu Lun, NIES;
Takao Miyoshi, RIHN; Akio Shimono, SPS; Shiro Hatakeyama,
TUAT.
East Asia is one of the most developing regions and thus
emissions of aerosol and its precursor are increasing.
Since Japan is situated at the east end of Asian continent
and the westerly wind is prevailing in winter-spring
period, aerosols are expected to be transported from Asian
continent to Japan. In order to understand both
concentration levels and chemical compositions of aerosol
in East Asia, we have continued the aerosol
measurements at Cape Hedo in Okinawa Island (128.5E,
26.5N) since October 2003. An instrument we deployed
was an Aerosol Mass Spectrometer (AMS) produced by
Aerodyne Research Inc.
Sulfate is dominant species. We also observed ammonium
and organics. However, nitrate and chloride were very
little in fine particle. Sulfate measured using Q-AMS was
compared with a low pressure impactor (LPI) data and a
filter pack sampling data. Average sulfate in spring period
were 4.28 (+/- 3.60) micro-g m-3, 6.37 (+/- 4.27) micro-g
m-3 and 6.07 (+/- 4.83) micro-g m-3 for 2004, 2005 and
2006, respectively. This is about 1.5-2 times higher than
the sulfate value (about 3 micru-g m-3) measured in 1992
- 1994 using a filter sampling method. We observed an
increasing tendency of sulfate at Cape Hedo, Okinawa.
One of the factors for high sulfate is air mass history.
When air mass is transported from China, high sulfate is
observed, while sulfate is low when air mass is transport
from Pacific Ocean. High sulfate is due to high sulfur
emission. Another factor is SO2 conversion to sulfate.
SO4/SOy (SOy = SO2 + SO4) at Cape Hedo is close to
unity, indicating that efficient conversion of SO2 to
sulfate occurs during transport.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.10
Particulate Matter Characteristics During Transport
Between Two Ground Sites in the 2006 MILAGRO
Campaign. XIAO-YING YU, Nels S. Laulainen, M. Liz
Alexander, J. Christopher Doran, Jerome D. Fast, Carl M.
Berkowitz, Pacific Northwest National Laboratory; Timothy B.
Onasch, Douglas R. Worsnop, Aerodyne Research Inc.; Eben S.
Cross, Boston College; W. Pat Arnott, Desert Research Institute.
In order to understand particle formation, transformation,
and transport in Mexico City, we deployed the T1-T2
study in the Megacity Initiative: Local and Global
Research Observations (MILAGRO) field campaign in
March 2006. The evolution of aerosols and their
associated chemical, physical, and optical properties was
investigated by deploying two sets of instruments at two
surface sites, T1 and T2, including a Sunset organic and
elemental carbon analyzer (OCEC), particle soot
absorption photometer (PSAPs), and nephelometer, as
well as a suite of in-situ and remote sensing meteorology
measurements. In addition, an Aerodyne Aerosol Mass
Spectrometer (C-ToF-AMS) was deployed at the T1 site.
During the field campaign meteorological conditions
favored transport between the T1 and T2 sites. Secondary
organic aerosols were predominant in fine particulate
mass loading. More aged particles were found at T2 than
at T1. Particle optical properties were studied based on
our observations.
11J.11
Continuous measurements of inorganic Reactive Gases and
aerosols across Europe during the EMEP Intensive
Measurement Campaigns 2006/07. EIKO NEMITZ, Rick
Thomas, Gavin Phillips, Centre for Ecology and Hydrology,
Edinburgh, UK; Chiara di Marco, Edinburgh University, UK;
Rami Alfarra, Andre Prevot, Paul Scherrer Institute, CH; Rene
Otjes, Jan Willem Erisman, Energy Research Centre of the
Netherlands (ECN), NL; Ari Laaksonen, Jukka Rautiainen,
University of Kuopio, FI; Laurent Poulain, Institute for
Tropospheric Research, D.
The new EMEP monitoring strategy foresees more detailed
measurements at typically one or two \supersites\ in each
member states to the UNECE Convention on Long-Range
Transboundary Air Pollution (CLRTAP). This includes daily
concentrations of inorganic gases and aerosols. Until the
strategy is fully implemented, EMEP has started to arrange
focused intensive measurement campaigns. The first two took
place in June 2006 and Jan 2007 to measure inorganic gas and
aerosol concentrations across Europe and assess the gas/aerosol
partitioning. Several sites operated two types of (semi-)
continuous monitors: the Aerodyne Aerosol Mass Spectrometer
+
2(Q-AMS) for non-refractory NH4 , NO3 , SO4 , Cl and total
organics in PM1 , and a wet chemistry monitor based on a
denuder / steam jet aerosol collector with online IC analysis for
+
gas phase NH3 , HNO3 , HCl, SO 2 and HONO as well as NH4 ,
-
2-
-
NO3 , SO4 and Cl in PM 2.5 and PM 10.
Measurements were made at Mace Head (Ireland), Auchencorth
Moss (UK), Harwell (UK), Cabauw (NL), Melpitz (D) and
Payerne (CH) and provide a detailed picture of the behaviour of
gases and aerosols at the same time, at different parts of Europe
under contrasting climatic conditions. They provide a database
for the assessment of European chemical transport models in
terms of emissions, meso-scale and local scale transport and
thermodynamic module. In particular: a) Secondary aerosol
concentrations increase from the North West (Ireland, Scotland)
towards the South. b) Sulphate concentrations agree between
PM1 , PM 2.5 and PM 10 measurements, indicating that all sulphate
is contained in sub-micron particles. c) Concentrations of nitrate
are governed by long-range transport at the North European sites
and by local meteorology at the central European sites. d) At the
UK sites a significant fraction of the nitrate is found in the
coarse fraction, while at the Dutch site all nitrate is found in the
fine fraction. e) Ammonium tends to be fully neutralized by the
sum of sulphate and NR nitrate. f) The English site and the
Dutch site differ greatly in their SO2 /NH3 ratio, reflecting the
high NH3 concentrations in the Netherlands.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.12
Characterization of chemical constituents in PM2.5 during
yellow sand events in Seoul, Korea. HYUN-SUN KIM, JongBae Huh, Bo-Ra Choi, Kye-Seon Kim, Seung-Muk Yi, School
of Public Health, Seoul National University; Jang-pyo Cheong,
KyungSung University.
Ninety nine chemically speciated samples were collected
on the roof of the School of Public Health building of
Seoul National University from March through May in
2003, 2004, 2005, and 2006 in the Seoul Metropolitan
area. The objectives of this study were to investigate the
characteristics of the major components in PM2.5 and to
characterize the chemical variations between yellow sand
and non-yellow events. The average PM2.5 concentration
3
was 49 microgram per m , that is almost three times
higher than the US NAAQS annual PM2.5 standard of 15
3
microgram per m . During this sampling period, yellow
sand and smog events were observed on 26 days and 13
days, respectively. The PM2.5 concentrations during
yellow sand events and smog events (PM2.5 more than 65
3
microgram per m ) were 1.5 and 2 times higher than those
during the spring non-events, respectively. Especially,
smog-yellow sand events were observed 6 days during the
yellow sand events. The concentrations of PM2.5 and its
constituents during smog-yellow sand events were 1.5 to
4 times higher than those during the spring non-events.
The concentrations of sulfate, nitrate, and ammonium
during smog-yellow sand events were higher than those
during the spring non-events and smog events.
Additionally, the concentrations of trace elements during
smog-yellow sand events were 2 and 4 times higher than
those during smog events and the spring non-events,
respectively. Five-day backward air trajectory analysis
showed that the air parcels during yellow sand events
passed through the desert areas in China and Mongolia,
while the air parcels during smog-yellow sand events
passed through not only the desert areas in China and
Mongolia but also major industrial areas in China. These
results suggested that the control strategy for yellow sand
events in Korea should be considered the desertification
as well as long range transport of air pollutants from
China.
11J.13
A Mass Spectral Fingerprint of Ship Emission Particles by
Aerosol Time-of-Flight Mass Spectrometry and Applications
for Source Apportionment. ANDREW P. AULT, Gerardo
Dominguez, Hiroshi Furutani, Mark Thiemens, Kimberly
Prather, University of California San Diego; Kimberly Prather,
Scripps Institution of Oceanography.
Mass spectral fingerprints for ship emission particles
were observed by aerosol time-of-flight mass
spectrometry (ATOFMS) and compared with hysplit back
trajectories and measurements of ship traffic (location,
speed, direction, and size) off the Southern California
coast to identify the potential impact of ship emissions on
the Southern California regional area. A unique class of
particles containing a strong vanadium signal as well as
iron and nickel was observed and is suggested to be
linked with ship traffic. This class of particles exhibits a
similar temporal trend with elemental carbon detected by
ATOFMS. These particles are likely the result of lower
grade fuel combustion from ship exhaust.
To assess the validity of the mass spectral fingerprint
comparisons were made with fresh ship exhaust data in
the open ocean during a clean marine period from the
CALCOFI 2004 cruise. The same vanadium-rich particles
were observed. This mass spectral signature has also been
detected during additional field campaigns where
ATOFMS data were collected in Southern California.
Using this mass spectral fingerprint, the ship particles
were estimated to be a major contributor to the submicron
size range during distinct periods of measurement at the
Scripps Pier. These data suggest that ship traffic could
play a major role in air quality for the Southern California
region and that the contribution of ship emission particles
needs to be accounted for in source apportionment.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.14
Aviation-Related Meteorological Changes Of Fog In
Southern Nigeria. Onifade, Yemi Sikiru.
This work was carried out by making use of Fog
measured Data for 5 years 4 months (64) months (1995
-2000). The monthly mean values of Fog days at the three
stations were determined. A Fast Fourier Transform
Program was used to obtain their Fourier analyzed Data.
The monthly frequency and Variational trend of Fog were
determined for the three stations. Graphical
representations were made. These show a significant time
variations which is an important conclusion for flight
plans and forecast of turbulent and other phenomenon.
11J.15
Methodology Using Surrogate Surface for the Estimation of
Atmospheric Dry Deposition Applicable in the Korean
Peninsula. JANGPYO CHEONG, Seung-Hoon Lee, Kungsung
University; Seung-Muk Yi, Seoul National University.
In this study, the fluxes of gaseous (HNO3 and SO2) and
particulate (NO3- and SO42-) species were measured
using two surrogate surfaces, the water surface sampler
(WSS) and the knife-edge surrogate surface with greased
Mylar strips (KSS). Sampling was conducted at the
agricultural area near Janghowon in Kyunggido from
March 2001 to June 2001. Ambient particle size
distributions were measured using a cascade impactor and
a coarse particle rotary impactor (CPRI).
Average fluxes of total mass, nitrate, and sulfate
measured with the KSS were 105.0±45.19, 3.20 +/1.70, and 2.41 +/- 1.11 mg m-2 day-1, respectively.
Average fluxes of nitrate and sulfate measured with the
WSS were 4.31 +/- 2.57 and 6.44 +/- 1.82 mg m-2 day-1,
respectively. These fluxes compare well with fluxes
obtained at rural sites in Korea in other studies. Average
gaseous HNO3 and SO2 fluxes calculated by subtracting
the particulate phase flux measured with KSS from the
total flux (gaseous phase + particulate phase) measured
with the WSS were 1.78 +/- 1.75 mg m-2 day-1 and 2.96
+/- 0.81 mg m-2 day-1, respectively.
The particle size distributions measured with a cascade
impactor and a CPRI were well fitted to the Weibull
probability distribution function. The distributions
showed a typical trimodal pattern peaked at around 5
micrometer and 15 micrometer in fine particle range and
42.5 micrometer in coarse particle range. Atmospheric
dry deposition fluxes of total mass and ionic species
estimated by various techniques using Weibull
distribution function (one-step, multi-step, equalconcentration, subdivision for only the coarse particle
range, etc.) were compared to the measured fluxes using
KSS. The results suggested that the deposition fluxes
estimated using the each particle size range determined
from the particle size distributions and dry deposition
velocity using Weibull probability distribution function
were the most applicable method.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.16
Effect of Atmospheric Deposition to Juam Reservoir in
Korea. JangPyo Cheong, YOUNG-HOAN JANG, Kungsung
University; Il-kyu Kim, Pukyong National University; Namik
Jang, Yeongsan River Environment Research Center.
11J.17
Atmospheric Aerosol Composition during the Convective
and Orographically-induced Precipitation Study (COPS).
WILLIAM MORGAN, Hugh Coe, Jonathan Crosier, James
Allan, Paul Williams, University of Manchester, UK.
The objective of this study was to investigate the impact
of atmospheric deposition on the water quality using the
several estimation methodologies including the mass
balance approach, statistical analysis of corelation and
regression, etc. Target water body of this research was
Juam reservoir which is one of the major water supply
resources in Chollanamdo, Korea.
The dry deposition fluxes and ambient concentration of
several ionic and gas-phase pollutants were measured
with DDP(dry deposition plate), WSS(water surface
sampler), CPRI(coarse particle rotary impactor) and 3stage filter pack from August in 1999 to November in
2000.
The direct atmospheric deposition(wet+dry) fluxes for
+
both NO3 and NH4 to Juam reservoir was estimated to be
183.13 ton per month and the flux of NH4+ was about 1.5
times higher than that of NO 3 . While the flux of NO 3 by
wet deposition was similar to that of dry deposition
In order to figure out the relationship between the dry
deposition flux of NO3 and the concentrations of NO3 -N
in Juam reservoir, the linear regression analysis was
conducted. The linear regression equation and coefficient
of determination, r2 are obtained was follows ; Y = 0.02X
+ 0.39, r2 = 0.77(The dry deposition flux of NO3- is taken
as the independent parameter of X and the concentration
of NO3 -N in Juam reservoir is taken as the dependent
parameter of Y)
Based on the mass balance analysis for nitrogen
containing compounds using hydrological data,
atmospheric deposition data and water quality data, the
estimated atmospheric loading of nitrogen containing
+
compounds(NO3 , NH4 ) was about 36 percent (27 percent
of dry deposition and 9 percent of wet deposition) of the
annual nitrogen load to Juam reservoir. This estimated
value in Juam reservoir was compatible with the values
obtained by previous studies in US.
The Convective Orographically induced Precipitation
Study (COPS) is an international project designed to
address the formation and development of convection
over hilly terrain and to improve predictions of heavy
convective precipitation. A key part of the study that links
the orographically forced flow and the microphysical
development of the clouds anchored my the mountains is
the transported particulate. Both the physical and
chemical properties of the aerosol have an effect on the
subsequently formed cloud. To study these effects the
COPS field campaign was held in the summer of 2007 in
the Black Forest and Swabian Jura of SW Germany and
Vosges Mts of E. France.
During COPS, airborne measurements were made using
the UK Facility for Airborne Atmospheric Measurements
(FAAM), a BAe146 aircraft and at a number of ground
based locations, in particular at the top of a mountain in
the Black Forest, the Horingsgrinde. The number size
distribution, aerosol composition and physical properties
were measured from both platforms. The aircraft
conducted low level passes of the valley region to
characterise the upwind aerosol and subsequently
characterised the cloud as it developed over the terrain
using a suite of cloud microphysics instrumentation.
Here we present an overview of the experiment from an
aerosol perspective, showing the suite of instruments at
Horningsgrinde and on board the aircraft and providing a
description of the main flight patterns conducted. Data
from both the ground and aircraft sites will be presented
demonstrating a characterisation of the aerosol in the
region during COPS and identifying diurnal variability
and export of aerosol from the Rhine valley region during
convective events.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.18
Measurement and Derivation of Emissions Factors for
Cotton Field Preparation. APRIL L. HISCOX, David R.
Miller, The University of Connecticut; Junming Wang, New
Mexico State University; Britt A. Holmen, The University of
Vermont; Wenli Yang, Crocker Nuclear Laboratory.
With a growing interest in emission of particulate
matter from farming operations, the development of
reliable emissions factors for these practices is important.
This presentation will give the specifics of a
comprehensive data set which will allow the computation
of emission factors from field preparation activities.
Available measurements in this area are limited, and the
previously reported emission factors and methods for
estimating emissions are based soil on surface soil silt
content. Emission factors derived directly from in-field
aerosol samplers and remotely from lidar measurements
will be presented. Additional micrometeorology
measurements onsite also allow of an analysis of emission
dependence on short term winds.
11J.19
Does Phytoplankton DMS Affect Iron Bioavailability in
Marine Atmospheric Aerosols? ANNE M. JOHANSEN,
Lindsey M. Shank, Central Washington University.
Iron availability limits open-ocean phytoplankton growth,
and because phytoplankton account for half of the Earth's
photosynthesis they are key players in modulating global
climate. The present study pertains to elucidating
mechanisms that control one of the prevalent avenues by
which this iron is supplied to remote oceans: the
deposition of atmospherically processed dust particles.
Chemical reactions on dust particles are believed
responsible for transforming aerosol iron into soluble
forms that are more available for phytoplankton
metabolism than the iron trapped inside sparingly soluble
mineral lattices. Previous laboratory observations
established a chemical link between iron reductive
dissolution and methanesulfinic acid (MSIA), an
oxidation product of dimethyl sulfide (DMS) emitted by
iron-starved phytoplankton. To investigate this
relationship in-situ, aerosols were collected over the
Equatorial Pacific Ocean between Hawaii and Papua New
Guinea during a 2-month research cruise (R/V Kilo
Moana) in Aug-Oct 2006. Results from these analyses
will be presented in the context of the proposed ironMSIA reaction.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
11J.20
Characterization of Saharan Dust Physical/Optical
Properties as Derived from the NASA NAMMA Airborne
Observations. GAO CHEN, Bruce Anderson, Lee Thornhill,
Eddie Winstead, Kuan-man Xu, and Yali Luo
11J.21
Implications of atmospheric SO2 and aerosol SO42variability and transport on particle acidity in Toronto,
Canada. KRYSTAL J. GODRI, Greg J. Evans, University of
Toronto.
The Saharan Air Layer (SAL) is thought to have a major
impact on suppressing the tropical cyclone activity over
the North Atlantic. The NASA African Monsoon
Multidisciplinary Analysis (NAMMA) airborne campaign
that was conducted during summer 2006 from Cape
Verde examined the characteristics of the SAL and sought
to elucidate the mechanisms of its influence over
convective storm dynamics. On ten different flights, SAL
properties were characterized using the extensive suite of
particle instruments and chemical sensors that were flown
aboard NASA DC-8 research aircraft. The SALs were
visually-evident, horizontally-extensive layers that
typically resided between 2 and 4 km in altitude. The
layers exhibited relatively low particle number
concentrations (300 to 600 cm-3), but scattering
coefficients that at times exceeded 300 Mm-1.
Extinctions integrated over the depths of the layers
typically yielded aerosol optical depths (AOD) of 0.4 to 1.
A majority of the particles present in the SALs were
nonvolatile at temperatures below 300 degreesC and
occupied two distinct two distinct size modes: an
accumulation mode with a geometric mean diameter
between 0.1 and 0.2 micro-m that accounted for 90% of
the number concentrations and a coarse mode with a
volume mean diameter of about 2 micro-m that was
responsible for >75% of the optical extinction and
essentially all of the mass loading. The prevalence of
large, non-absorbing dust particles in the layers resulted
in scattering coefficients that were relatively constant
with wavelength (e.g., near zero angstrom exponents) and
single scattering albedos that ranged from 0.98 to 0.99.
To assess the influence of the dust on tropical cyclones,
the DC-8 in situ measurements were used to constrain a
cloud model simulation. Along with a summary of the
airborne observations, we will discuss preliminary
modeling results and the predicted Saharan dust influence
on cloud formation and precipitation rates.
The Ontario Power Generation Nanticoke coal fired
power station is of concern to Toronto's air quality;
located ~150 km southwest of Toronto, emitted pollutants
are regularly transported into the city and observed at the
sampling site situated in the downtown core. A fourmonth sampling campaign conducted during the summer
of 2006 semi-continuously measured SO2 and sulphate
PM2.5 concentrations using a Dionex Gas Particle Ion
Chromatograph with 15 minute intervals. Local and
regional sources contributed to the total measured SO2.
The geographic origins of regional measured gaseous and
particulate species were elucidated using NOAA
HYSPLIT Trajectory and Dispersion Model analysis.
Local SO2 sources contributed only weakly to total SO2
concentrations: the diurnal fraction of local SO2
concentrations contributed a maximum of 25% to the total
from 03:00 to 07:00. Three categories of sulphur
episodes were identified: SO2 episodes, haze episodes
and simultaneously elevated particle and gas
concentrations. The first was characterized by sharp
narrow peaks of elevated SO 2 levels and low sulphur
particulate fractions. Haze episodes were distinguished by
low visibility and elevated sulphate particulate
concentrations. During the summer 2006 sampling
campaign, a total of seven sulphur particulate episodes
originating from Nanticoke were identified. However,
occasions also existed when air masses traveled over
Nanticoke and proceeded into Toronto but no episodes
were monitored. Temporal variations in measured
sulphate concentrations were examined using
meteorological parameters and particle acidity
information. The latter parameter was influenced by
ammonium and sulphate concentration which dominated
+
the total PM2.5 mass during summer months. Low NH 4 /
2SO4 ratios indicated that the extent of particle
neutralization was low. Application of the Aerosol
Inorganic Model (AIM-II) to measured data provided in+
situ aerosol acidic properties including free H
concentrations and aerosol pH levels, the latter of which
also accounts for the degree of bisulphate ion
dissociation. Sulphate particle size was also related to
shifts in inorganic aerosol chemical composition and
relative humidity. Thermodynamic simulation results
were used to understand the formation pathway and size
of these sulphate aerosols.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
16C.1
Coupled measurements of the size, chemical mixing state,
and optical properties of individual atmospheric particles.
KIMBERLY PRATHER, Ryan Moffet, University of California
at San Diego.
Single particle measurements are now becoming a routine
component of atmospheric measurements in an effort to
better understand the chemical mixing state of aerosols,
which plays an important tole in determining the forcing
of the atmosphere. Recently, we have added the ability to
make coupled optical measurements in our on-line single
particle mass spectrometer which measure the
aerodynamic size and chemical composition of single
particles. Measurements in Mexico City and California
will be described. Specifically, a discussion will be
presented on the single scattering albedo as a function of
size of different particle types including soot-sulfate and
organic carbon particles in the atmosphere.
16C.2
Transboundary Pollutant Impacts of Emissions in the
Imperial Valley-Calexico Region and from Southern
California. SANTOSH CHANDRU, Yongtao Hu, Armistead
G. Russell, Georgia Institute of Technology; Ana yael Vanoye,
Arturo Moran Romero, Alberto Mendoza, Instituto Tecnologico
y de Estudios Superiores de Monterrey.
Air pollution continues to be an increasing problem with
the socio-economic and industrial growth in both sides of
the US-Mexico border. Earlier Studies in the region have
primarily focused on analysis of primary PM and to
certain extent on secondary PM. In the current study, the
dynamics of Ozone and particulate matter is studied with
a modeling perspective using the Models-3 framework
(USEPA, 1999) over the US-Mexico border of the
Mexicali- Imperial valley region and Southern California,
including Los Angeles and San Diego areas, for pollution
events in August 2001, where peak Ozone concentrations
reached 190 ppbv in the LA basin on August 26, 2001.
The study also addresses the air quality impacts of two
power plants located three miles south of the US-MX
border on the Mexicali-Imperial region. Cross-boundary
transport of pollutants and its resulting impacts is also
studied.
The Mesoscale Meteorological model (MM5) is utilized
to simulate the meteorological fields. The BRAVO
Emissions Inventory (USEPA, 2001) along with the NEI
files (USEPA, 2001) and Mexico NEI 1999: Six Northern
States is used for emissions modeling using SMOKE. The
Chemical Transport Model of the Community Multiscale
Air Quality (CMAQ v4.4) model simulated pollutant
concentrations, including PM components. Several
sensitivity analyses using the Direct Decoupled Method in
CMAQ (Cohan et al., 2005) were conducted, including
the impact of area and mobile sources of NOx and VOC
on particulate matter and Ozone, and how SO2 impacts
particulate matter levels. Sensitivities were also studied
for perturbations in VOC, and NOx concentrations in the
Los Angeles basin and Tijuana-San Diego regions.
Mobile NOx emissions from Tijuana-San Diego region
had impacts of upto 5 ppbv of O3 in the Mexicali area.
The impacts of SO2 emissions from the two power plants
on PM concentrations are also discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
16C.3
Lead in single atmospheric particles. DANIEL MURPHY,
Karl Froyd, Troy Thornberry, David Thomson, NOAA Earth
System Research Laboratory; Paula Hudson, University of Iowa;
Daniel Cziczo, Stephane Gallavardin, ETH Zurich; Murray
Johnston, Melissa Reinard, University of Delaware; Anthony
Wexler, UC Davis.
Single particle measurements using PALMS have shown
that a significant fraction of accumulation mode particles
contain trace amounts of Pb. This talk will explore the
reasons for the frequency of Pb in fine particles now that
most gasoline is unleaded. Trace amounts of Pb were
found in 5 to 25% of 250 to 3000 nm diameter particles
sampled by both aircraft and surface instruments in the
eastern and western United States. Lead was found on all
types of particles, including Pb present on biomass
burning particles from remote fires. Less common
particles with high Pb contents contributed a majority of
the total amount of Pb. Single particles with high Pb
content often also contained alkali metals, Zn, Cu, Sn, As,
and Sb. The association of Pb with Zn and other metals is
also found in IMPROVE network filter data from surface
sites. Sources of airborne Pb in the United States are
reviewed for consistency with these data. The frequent
appearance of trace Pb is consistent with widespread
emissions of fine Pb particles from combustion sources
followed by coagulation with larger particles during longrange transport. Industrial sources that directly emit Pbrich particles also contribute to the observations. Clean
regions of the western United States show some transport
of Pb from Asia but most Pb over the United States comes
from North American sources. Resuspension of Pb from
soil contaminated by the years of leaded gasoline was not
directly apparent.
16C.4
Long-Term Measurements of Size-Resolved Particle
Chemistry and its Dependence on Air Mass Origin in the
German Lowlands. GERALD SPINDLER, Erika
Brueggemann, Thomas Gnauk, Achim Gruener, Konrad
Mueller, Birgit Wehner, Alfred Wiedensohler, Hartmut
Herrmann, Leibniz-Institute for Tropospheric Research, Leipzig,
Germany; Thomas M. Tuch, UFZ Centre for Environmental
Research, Leipzig, Germany; Markus Wallasch,
Umweltbundesamt, Dessau, Germany.
A joint investigation (supported by the
Umweltbundesamt, project 351 01 022) for a sizesegregated physical-chemical characterization of
tropospheric aerosol has started in spring 2004 at the
research station of the Leibniz-Institute for Tropospheric
Research (IfT) in Melpitz in the river Elbe valley (12
degrees 56' E, 51 degrees 32' N, 86 m asl.). This spot is
integrated in EMEP activities and a supersite in the
EUSAAR network. 24 hour samples for PM10, PM2.5
(every day) and PM1 (at least every six days) were
collected using high volume samplers.
Particle mass concentration was determined
gravimetrically and water-soluble ions were detected by
ion chromatography. Organic and elemental carbon were
quantified by a thermographic method. The particle
number size distribution was measured between 3 and 800
nm. During selected days with a distinct air mass origin
particles with diameters between 0.05 Вµm and 10 Вµm
were size-fractionated using a five stage BERNER-type
low pressure impactor and analyzed for mass, water
soluble ions, carbon and selected organic species.
The mean source regions for wintertime
anthropogenically influenced air masses are inside and
outside of the European Union. The mean concentrations
in the Melpitz area for PM10, PM2.5 and PM1 were 19.9,
15.2 and 12.7 micro-g/m3 in 2004 and 22.4, 17.6 and 12.9
micro-g/m3 in 2005, respectively. Therefore, particles
were physically and chemically characterized for two
years after size-segregated sampling. The main results of
the project are the differences in the mean particle mass
concentration, the chemical distribution and the physical
properties of particles distinguished for air masses
transported from West or East in summer and winter by
classification of the daily results. These differences show
the possible influence of long-range transports from the
east, mostly in wintertime, to the region of Saxony near
the Polish border in Germany.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
16C.5
Hygroscopic Properties of Sub-Micrometer Atmospheric
Aerosol Particles Measured with H-TDMA Instruments in
Various Environments- A Review. KAARLE H
The hygroscopic properties of atmospheric submicrometer aerosol particles are vital for a proper
description of how the particles interact with water vapour
at sub- and super-saturated conditions, and are thus of
major importance in describing the life cycle of the
aerosol and the direct and indirect effects of aerosols on
climate. The hygroscopic properties can be measured in
great detail using H-TDMA instruments (Hygroscopic
Tandem Differential Mobility Analyzers), providing online and in-situ information regarding the extent of
external versus internal mixing of the atmospheric
aerosol, since the H-TDMA determines the hygroscopic
growth of individual aerosol particles. The primary
parameters measured with an H-TDMA as a function of
dry particle diameter are i) the ratio between humidified
and dry particle diameter at a well-defined relative
humidity RH - often denoted hygroscopic growth factor;
ii) the number fraction of particles belonging to each of
the observed and separable groups of hygroscopic growth;
and often also iii) the spread of diameter growth factors
around the arithmetic mean value. While the bulk of HTDMA data is available for a nominal and high RH (often
between 80-90%), some field studies have performed
scans in RH to explore the aerosol deliquescence and
efflorescence behaviour as well as water uptake at low
RH. This work reviews and summarizes the existing HTDMA data sets, with an emphasis on those published so
far in peer-reviewed journals. The aim is to present the
data in a way that will make it useful in evaluating models
on various spatial and temporal scales incorporating a
more detailed aerosol description than simply aerosol
mass. To facilitate comparison between sites, growth
factors are recalculated to an RH of 90% whenever
possible, and classified according to the air mass
properties and geographical location.
16C.6
Water-Insoluble Particles in Spring Snow at Mt. Tateyama,
Japan: Characteristics of the Shape Factors in Relation with
Their Origin, Transportation and Preferential Settling.
JING-MIN LI, Kazuo Osada, Nagoya University, Japan
The shape factors of water-insoluble particles (WIP) in
four dirty snow layers in the spring of 2001 at Mt.
Tateyama, Japan, were analyzed using scanning electron
microscopy and optical microscopy together with
imaginary analysis software. Results show that the
median aspect ratio (ratio of the longest dimension a to
the orthogonal width b; a/b) of the WIP varied within 1.22
-1.31. Only a few particles with an aspect ratio of more
than 2.5 were observed. The median circularity factor (4 x
Pi x S / l^2 ); S is projection area and l is periphery
length) varied from 0.83-0.97. Bias of dust particles'
centers of gravity was observed: L1 (the longest distance
from the center of gravity to the boundary of particles) is
5% (of L1, on average) longer than L2 (1/2 of the longest
axis of particles).
Combined with backward air trajectories, visibility
reducing surface weather reports, additional results of rain
and dry deposition samples containing Saharan dust and
Asian dust, the sources of those WIP were identified. By
comparing the shape factor distributions of dust particles
observed in Japan and their source area, an interesting
result was discovered: the proportions of elongated
particles were less in the sample that had been transported
a longer distance from the dust source areas.
To explain observational results above, Ginoux's
mathematic model was studied. By applying the
preferential orientation of particles settling heavy side
down based on the observed bias of particles' center of
gravity, the settling velocity for ellipsoidal particles with
Reynolds numbers lower than 2 was estimated. Results
show that, for particles less than 11 micro-meter, the
settling velocity of spheroids is lower than that of
ellipsoids having equal surface area, which would imply a
lower proportion of elongated particles after long range
transport, as observed previously.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
17C.1
Wintertime Measurements of Fine Aerosol Chemical
Composition and Gas Phase Precursors Near the Flatirons
in Boulder, Colorado. R. BAHREINI, B.M. Matthew, H.D.
Osthoff, J.A. Neuman, T. Fortin, A.G. Wollny, E.J. Williams, B.
Lerner, and F.C. Fehsenfeld, University of Colorado, CIRES
and NOAA Earth System Research Laboratory, CSD; A.M.
Middlebrook, S.S. Brown, C.A. Brock, and T.B. Ryerson,
NOAA Earth System Research Laboratory, CSD; A. Swanson
and F. Flocke, National Center for Atmospheric Research; P.K.
Quinn and K. Schulz , NOAA Pacific Marine Environmental
Laboratory.
Smog episodes have been observed in the Denver
metropolitan area, more so with wintertime temperature
inversions, and have been the subject of several field
studies dating back to late 70s. Here we present results
from measurements of aerosol chemical composition and
mass size distributions by an Aerodyne Quadrupole
Aerosol Mass Spectrometer (Q-AMS) as well as gas
phase precursors made during the Winter Nitrogen Oxides
and Aerosol experiment (Winter NOaA 2005- Jan. 27Feb. 9) in the foothills in Boulder, CO, 25 miles
northwest of Denver. On average, sulfate and organics
contributed similar amounts to aerosol mass, each about
3
1.1 micro-g/m . Non-refractory aerosol composition was
dominated by ammonium nitrate with an average
3
concentration of 3.4 micro-g/m . Bimodal mass
distributions were occasionally observed with nitrate,
ammonium, and/or organics dominating the smaller
mode. Small mode nitrate aerosols were present
predominantly at low temperatures and high relative
humidity, with easterly winds. Very low gas phase nitric
acid concentrations at these times point to nitric acid
limited conditions for aerosol nitrate formation. N2O5
concentration at night showed an inverse relation with
aerosol nitrate concentration when relative humidity was
high. Nighttime uptake of N2O5 on available aerosol
surface area will be examined to determine the extent of
nitrate formation by N2O5 hydrolysis. Principal
component analysis on the organics mass spectra showed
that the organics were predominantly fresh and
hydrocarbon-like (HOA) as opposed to aged and
oxygenated like (OOA), consistent with previous
observations in urban areas during winter. Correlations
between HOA vs primary and OOA vs secondary markers
will be presented.
17C.2
The Role of Climate and Emission Changes on PM2.5 over
North America and Uncertainty Assessment of Global
Climate Change Impacts. EFTHIMIOS TAGARIS, Kuo-Jen
Liao, Kasemsan Manomaiphiboon, Armistead G. Russell,
Georgia Institute of Technology,; Jung-Hun Woo, Shan He,
Praveen Amar, Northeast States for Coordinated Air Use
Management (NESCAUM); Lai-Yung (Ruby) Leung, Pacific
Northwest National Laboratory; Chien Wang, Massachusetts
Institute of Technology.
The objective of this study is to assess the impacts of
global climate and emissions changes on regional air
quality over North America with particular focus on
global climate change uncertainties. PM2.5 concentrations
for historic episodes (i.e. 2000s) are compared to future
episodes (i.e. 2050s). Meteorological fields are derived
from the GISS GCM and have been regional downscaled
using the Penn State/NCAR Mesoscale Model (MM5).
CMAQ with SAPRC-99 chemical mechanism is used for
the regional air quality modeling. MIT's Integrated Global
System Model (IGSM) is used to suggest uncertainties in
future climate (i.e. temperature and absolute humidity).
Low-extreme and high-extreme scenarios of air
temperature and absolute humidity which derived from
IGSM are used to perturb the base scenario which derived
from GISS model. Results show that uncertainties from
impacts of climate changes on PM2.5 concentrations are
larger in the higher extreme case and may have significant
regional variations. These results imply that uncertainties
of climate change should also be included when
investigating the influences of climate change on regional
air quality.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
17C.3
Nucleation and particle growth over/in a forest. S.C.
PRYOR, Indiana University - Bloomington and Risoe National
Laboratory, Roskilde, Denmark; R.J. Barthelmie, University of
Edinburgh, UK and Indiana University - Bloomington; F.
Rahman and V. Cordova, Indiana University - Bloomington.
We are conducting continuous measurements of particle
size distributions (6 to 400 nm) at three-levels above and
within a deciduous forest in the Ohio River Valley, and
analyzing the data to quantify the frequency and
characteristics of nucleation events and high ultra-fine
particle concentrations, including the chemical
composition of the ultra-fine particles, the principal
mechanisms of nucleation, limitations on nucleation and
growth and the ultimate fate of the resulting ultra-fine
particles. Here we focus on winter-time (leaf-off) periods
with high number concentrations of ultra-fine particles
and show that nucleation events appear to occur above the
canopy. At the start of these events net radiation (NR)
levels are frequently below the threshold of 300 W/m2
that has been proposed as characterizing nucleation
events, but during the period of highest observed particle
concentrations NR < 300 W/m2. Initiation of increased
particle concentrations appears to be coincident with a
transition from highly stable conditions over night
towards unstable conditions. This transition is consistent
with erosion of a nocturnal inversion, and vertical
transport of precursor gases and/or nucleated particles.
The ultra-fine particles have relatively low growth rates of
approximately 2 nm/hr, almost all of which can be
attributed to coagulation. We are currently examining
whether periods associated with high ultra-fine particle
concentrations have unique signatures in aerosol products
derived from the MODIS Terra satellite in order to
provide a regional context for our measurements and
potentially a prognostic tool.
17C.4
Holme Moss 2006: Overview. James Allan, The University Of
Manchester, UK; Betsy Andrews, NOAA; Karl Beswick, Keith
Bower, Rachel Burgess, Hugh Coe, BENJAMIN CORRIS, Ian
Crawford, James Dorsey, Michael Flynn, Martin Gallagher,
Nicholas Good, Martin Irwin, Dantong Liu, Gordon McFiggans,
William Morgan, The University Of Manchester, UK; John
Ogren, NOAA; Paul Williams, The University Of Manchester,
UK.
The field of aerosol science has direct relevance to the
understanding of a wide range of phenomena including
their effects on cloud formation and perturbations to
radiative forcing. Due to the particles short residence time
in the atmosphere their distributions and chemical
character vary with both time and geographical location
making quantifying their effects exceedingly difficult.
Compounding these problems is a lack of understanding
of the detailed microphysical behaviour of these particles
in the atmosphere and the obscure nature of the organic
fraction. (IPCC 2001).
The Holme Moss experiment ran from the beginning of
November 2006 to the beginning of December 2006.
Holme Moss is a moor in the south Pennines of England,
on the border between Derbyshire and the West Yorkshire
district of Kirklees (Co-ordinates 53 degrees 31'58'' N, 1
degree 51'21'' W). The sampling site was down-wind from
Manchester and frequently subjected to hill cap cloud.
An extensive array of instrumentation was deployed to
measure both the aerosol and gas phase constituents of the
atmosphere in addition to meteorological measurements.
These instruments included an Aerodyne Time of Flight
Aerosol Mass Spectrometer (ToF-AMS), Differential
Mobility Particle Sizer (DMPS), Hygroscopic Tandem
Differential Mobility Analyser (HTDMA),
Nephelometers, Particle Soot Absorption Photometer
(PSAP), Multi-Angle Absorption Photometer (MAAP)
Cloud Condensation Nuclei Particle Counter (CCN).
During in-cloud events part of the instrumentation was
switched from an interstitial inlet to an inlet equipped
with a counterflow virtual impactor (CVI) to allow the
sampling of the relatively large cloud drops.
This work will provide an overview of aerosol data taken
at Holme Moss, England in 2006 along with comparisons
between the physical, chemical and optical properties of
the aerosol. Preliminary results indicate the aerosol was
externally mixed and contained a significant amount of
absorbing material.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Remote and Regional
2007 AAAR Annual Conference Abstracts
17C.5
Investigating apparent particle emission fluxes over forests.
R.J. BARTHELMIE, University of Edinburgh and Indiana
University - Bloomington; S.C. Pryor, Indiana University Bloomington and Risoe National Laboratory, Roskilde,
Denmark.
Particle number fluxes over forests are typically
downwards (negative), but multiple previous studies have
indicated the presence of a substantial number of positive
(upward) fluxes. Here we present analyses of data from a
beech forest in Denmark that indicate these apparent
emission fluxes are frequently statistically different from
zero flux, and are observed in fluxes computed from a
variety of micrometeorological methods. In this data set
over 1/3 of all half-hour periods exhibit upward fluxes.
These upward fluxes are not solely observed during
periods when other micro-meteorological fluxes are illdefined, which in conjunction with other evidence implies
that they derive from a/multiple physical cause/s. Upward
fluxes are slightly more frequent at night, but are
observed in all hours of the day. They do not appear to be
dependent on wind direction or speed. The rate of upward
fluxes (emission velocity) scales with prevailing
geometric mean diameter (GMD), with higher emission
velocities being associated with smaller GMD, but the
vertical exchange velocity during emission periods
appears not to scale with changes in GMD.
17C.6
Airborne measurements of the export of gaseous and
particulate species from the UK. JONATHAN CROSIER,
Hugh Coe, James Allan, Keith Bower, Paul Williams, Gerard
Capes, University of Manchester, UK; Debbie Polson, David
Fowler, Centre for Ecology and Hydrology, Edinburgh, UK;
Dave Stewart, University of East Anglia, Norwich, UK.
The export of most primary anthropogenic pollutants and
greenhouse gases is relatively well known and is
documented in emissions inventories for many countries.
Quantifying these anthropogenic emissions helps us to
understand the possible impact on important atmospheric
processes such as global warming and acid deposition, as
well as air quality. The AMPEP field campaign (Aircraft
Measurement of chemical Processing and Export fluxes
of Pollutants over the UK) was conducted in the UK
between April and September 2005 to directly measure
the export of major pollutants and their derivatives.
Fifteen research flights onboard the UK BAE-146 Facility
for Airborne Atmospheric Measurement (hereafter
referred to as FAAM) were conducted between 21st April
and 29th September 2005.
The export of pollutants from the UK are calculated using
a simplified model, using measured concentration
enhancements of various species compared to upwind
values, along with average wind speed, wind direction
and boundary layer height determined by the National
Atmospheric Modelling Environment (NAME), driven by
the Met Office Unified Model (UM). Aerosol chemical
composition was measured using an Aerodyne Research
Inc. Quadrupole Aerosol Mass Spectrometer (Q-AMS).
Under strong westerly flow (usually a result of low
pressure to either the north or west of the UK) measured
annual emissions of NOx and its oxidation products
compare well to NAEI (National Atmospheric Emissions
Inventory) annual emissions for 2004. Nearly all of the
nitrogen in this meteorological situation is found in the
gas phase, mainly in the form of NO2, with little
Secondary Organic Aerosol (SOA) produced. In high
pressure, stagnant and European outflow condition the
amount of nitrogen found in the form of nitrate in the
aerosol phase is significantly increased, as is the amount
of SOA formed. UK exports of SO2, particulate non-seasalt sulphate and particle number are also shown.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.1
Applications of the Advanced EPA PMF and PMF2 Model
for PM2.5 Source Apportionment. INJO HWANG, Philip K.
Hopke, Clarkson University; Pentti Paatero, University of
Helsinki.
In order to estimate mass contributions and chemical
composition of PM2.5 sources using the PMF2 and EPA
PMF V2.0, PM2.5 samples collected at Washington, DC
IMPROVE site from August 1988 to December 1997
were analyzed. A total of 718 samples were used with 35
species determined by PIXE (particle-induced X- ray
emission), XRF (X-ray fluorescence), PESA (proton
elastic scattering analysis), IC (ion chromatography), and
IMPROVE/thermal optical reflectance (TOR) method.
These data were selected for PMF2 and EPA PMF
analysis. Both PMF models identified ten sources:
secondary sulfate I, gasoline vehicle, secondary sulfate II,
secondary nitrate, secondary sulfate III, incinerator,
airborne soil, aged sea salt, oil combustion, and diesel
emission, respectively. In a comparison of source profiles
resolved by both models, the source profiles of each
source showed good agreement with the differences
limited to a few species. The calculated average
concentrations of PM2.5 were consistent with between
PMF2 and EPA PMF analysis (17.93 +/- 0.30 microgram/m3 and 17.94 +/- 0.30 micro-gram/m3). Also, each
estimated source contribution showed good agreement
between PMF2 and EPA PMF. Thus, the next version of
EPA PMF (V2.0) that will include rotational capabilities,
will provide reasonable solutions and because of its ease
of use, it can be more widely applied to solving air quality
management problems.
2J.2
Source Apportionment for Semi-Continuous Data at St.
Louis Supersite. INJO HWANG, Philip K. Hopke, Clarkson
University.
Semi-continuous PM2.5 species were collected at the St.
Louis-Midwest supersite. Time-resolved samples were
collected one week in each of June 2001 (22 June to 28
June), November 2001 (7 November to 13 November),
and March 2002 (19 March to 25 March). The sampling
procedure consisted of 1-hour sampling intervals.
Elements were determined using the Semi-continuous
Elements in Aerosol System (SEAS). The SEAS samples
were analyzed for eleven species (such as Al, As, Cd, Cr,
Cu, Fe, Mn, Ni, Pb, Se, and Zn) by graphite furnace
atomic absorption spectrometry (AAS). Elemental carbon
(EC) and organic carbon (OC) were measured with a field
Sunset OC/EC analyzer using the ACE-ASIA protocol
(NIOSH/TOT method), and semi-continuous PM2.5 sulfate
and nitrate measurements by particle-into-liquid sampler
(PILS).
The objective of this study is to estimate the mass
contributions and chemical composition of sources of
PM2.5 at St. Louis supersite. PMF was applied to identify
the sources and apportion the PM2.5 mass to each source
for highly time resolved data. In addition, the conditional
probability function (CPF) and nonparametric regression
(NPR) was applied to identify the predominant directions
of local sources relative to wind direction. Also, this
study will be comparing the source directions between
CPF and NPR analysis. Finally, resolved source profiles
were comparing with source profiles obtained by previous
St. Louis studies.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.3
Source Identification of PM2.5 Measured at Tae-In Dong,
Gwangyang in Korea near Large Steelworks Using Positive
Matrix Factorization (PMF) Model. JONG-BAE HUH,
Yong-Seok Seo, Hyun-Sun Kim, Seung-Hee Kim, Seung-Muk
Yi, Seoul National University.
Taeindong, Gwangyang in Korea is an industrial region
where a large steel mill and related industrial complex are
located. During the last several years, there has been
growing concerns about the adverse health and
environmental effects of air pollutants in the residential
area near the steel mill in this area. The objectives of this
study were to investigate the characteristics of major
components in PM2.5 and to identify the source types and
contributions of PM2.5 using PMF model. Samples were
collected on the roof of the Taeindong Development
Association building (127.76E, 34.94N, and 17 m) from
August 2003 through January 2005. The site is
surrounded by commercial and residential buildings
located on the north of the steel mill and related industrial
complex. PM2.5 and chemical speciated samples were
collected using Partisol speciation sampler (Rupprecht &
Patashnick, USA).
The PMF analysis resolved eight sources with their
average contributions: B-C fuel and coal chemistry
(29%), motor vehicle (19.4%), secondary nitrate (14%),
nitrogen-compound producing process (10.1%), fresh sea
salt and chlorine producing process (8.7%), soil and road
dust (7.9%), steel producing process (6.8%), and biomass
(2.9%). The overall average source contribution from
steelworks and its related industries accounted for about
60% of PM2.5 in this study. The contribution from steel
producing processes increased from 6.8% to about 50%
when wind was from south indicating that steelmaker
could be an important contributor to PM2.5 in this area
particularly during the summer with prevailing southern
wind.
2J.4
Roadside, Near-Road and Regional Detailed Chemical
Composition and Source Apportionment of PM2.5 at
Atlanta, GA in Two Seasons. BO YAN, Mei Zheng, Amy
Sullivan, Rodney Weber, Sangil Lee, Charles Evan Cobb,
Santosh Chandru, Hyeon Kook Kim, Armistead G. Russell,
Georgia Institute of Technology; Eric S. Edgerton, Atmospheric
Research & Analysis, Inc.
To investigate chemical composition and source
contributions of fine particulate matter (PM2.5) in
Atlanta, GA, a three-channel particle composition monitor
(PCM) was used to collect fine particle ambient samples
during two intensive episodes (summer 2005 and winter
2006). Three sampling sites were utilized: a roadside
highway site (directly beside the I-75/85 connector in the
midtown Atlanta); a more typical urban site in the
Georgia Tech campus (approximately 450 meters away
from the highway) and a rural site in Yorkville, GA
(impacted primarily by biogenic emissions and regional
transport).
In this study, particle phase chemical composition of
PM2.5 was investigated including organic carbon (OC),
elemental carbon (EC), 40 trace metals, ions, and organic
compound speciation (e.g., molecular markers).
Temporal and spatial variations of PM2.5 mass and
composition were compared and investigated. A
chemical mass balance (CMB) model was applied to
estimate major source contributions to PM2.5 at each site.
Results indicated that on-road vehicle emission dominated
ambient EC concentrations at the highway site. However,
this impact dramatically decreased with increasing of the
distance away from the highway. Within a distance of
450 meters from the I-85/75 highway site to the Georgia
Tech campus site, ambient EC and OC dropped by 74%
and 35%, respectively.
In summer time, the average ambient OC and EC
-3
-3
concentrations were 8.2 and 4.0ug.m , 5.2 and 1.0ug.m ,
-3
3.8 and 0.2ug.m at the highway roadside site, Georgia
Tech campus site and the Yorkville site, respectively. In
winter time, on average ambient OC and EC data were
-3
-3
reported as 5.1 and 2.7ug.m , 3.6 and 0.9ug.m , 2.1 and
-3
0.3ug.m at the three sampling sites above, respectively.
Higher OC concentrations in summer time imply strong
photochemical activities leading to an elevated level of
secondary organic aerosol (SOA) in PM2.5. Comparison
of the summer and winter roadside samples, in particular
comparing metal tracers with specific organic tracers and
total OC, suggested higher OC emissions from on-road
motor vehicles in the winter. Detailed discussion of
source apportionment results will be presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.5
Source apportionment of suspended particulate matter in a
clean area of Delhi using chemical mass balance receptor
model. ARUN SRIVASTAVA, V. K. Jain, Jawaharlal Nehru
University, New Delhi.
2J.6
Simultaneous Factor Analysis of Organic Particle and Gas
Measurements in Downtown Toronto. JAY SLOWIK,
Alexander Vlasenko, Maygan McGuire, Greg Evans, Jonathan
Abbatt, University of Toronto.
Source Apportionment of Suspended Particulate Matters
was carried out at Jawaharlal Nehru University (JNU), an
extremely clean location of Delhi using Chemical Mass
Balance Receptor Model (CMB8). 24 hour samples were
obtained continuously for 32 days during February 25 to
March 27, 2000. Results were obtained by CMB for each
day. The average results of 32 days reveal that diesel
vehicles are the major contributor (over 50%) among all
the sources, followed by the industrial source
(approximately 24%). The other significant contributions
were observed from paved road dust (10.2%), Gasoline
Vehicles (6.2%) and solid waste (5.8%). Soil and crustal
dust's contribution was lowest (approximately 1.7%). By
and large the contributions of most of the sources are
variable, except diesel vehicles which is relatively stable.
Possibility of some unknown sources of few metals (Ni,
Mn, Fe and Cu) among the considered species could not
be ruled out.
During the winter component of the SPORT (Seasonal
Particle Observations in the Region of Toronto),
particulate non-refractory chemical composition and
concentration of selected VOCs were measured by an
Aerodyne time-of-flight aerosol mass spectrometer and a
proton transfer mass spectrometer, respectively.
Sampling was performed in downtown Toronto, ~5 m
above ground level and ~15 m from a busy roadway.
Positive matrix factorization was used to deconvolve the
organic spectra collected by the two instruments into
factors related to chemical composition and emissions
sources. Six major factors have been identified: (1)
oxygenated organic aerosol and long-lived VOCs; (2)
less-oxygenated organics and shorter-lived VOCs; (3)
roadway traffic emissions; (4) local traffic emissions; (5)
charbroiling emissions; and (6) cooking emissions.
Factor 1 is characterized by a high ratio of m/z 44 to total
organics (~0.17) in particles and elevated concentrations
of VOCs such as acetic acid, which is attributed primarily
to secondary anthropogenic sources and has a lifetime of
~15 days. Factor 2 shows less-oxygenated particulates
and elevated concentrations of VOCs such as
acetaldehyde, a marker for secondary oxidation with a
lifetime of ~0.8 days. The particulate mass spectra of
factors 3 and 4 are similar, but the concentrations of
aromatic gases such as benzene and toluene are higher in
factor 4. Additionally, the temporal variation of factor 3
is characterized by large spikes in the particulate
concentration with a typical duration of less than a
minute, while factor 4 exhibits more gradual diurnal
variation coinciding with the morning and evening rush
hour. Between the hours of approximately 11 AM and 5
PM, factor 5 is also present in the spikes, and is attributed
to charbroiling emissions from a nearby open-air cooking
stand. Factor 6 is typically of maximum intensity during
the evening and has a mass spectrum qualitatively
consistent with fatty acids.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.7
Aerosol Impacts from Secondary Roadways. THOMAS A.
CAHILL, David E. Barnes, Steve Cliff, DELTA Group,
University of California, Davis; Thomas M. Cahill, Arizona
State University.
Reductions in emissions from diesel engines in the past 2
decades have resulted in emissions per kilometer that are
only about 10 times that of the average automobile. Thus,
since on secondary roads cars outnumber diesels by
typically 50 to 1, roadway emissions are generally
dominated by spark emission vehicles. We have studied
aerosols at both freeways and heavily traveled secondary
streets in Northern California, and compare the emitted
aerosols as a function of size, time, and composition,
including organic matter (PAHs, others) versus particle
size. The measurements were then compared to two
independent models: mass emission rates from the
Tuscarora Tunnel data of Gertler et al (2002) and diesel
particulate mass and elemental profiles from Zielenska,
Cahill et al (2000). Good agreement was achieved
between the two models and the data, showing high levels
(up to 7 micrograms/m3) of less than 0.25 micro-meter
automobile aerosols from the secondary roadways at
downwind receptor sites. Burned lubricating oil was a
major contributor to this mass. At Lake Tahoe, we were
also able to establish the sources and influence of soil and
phosphorus containing winter aerosols from sanding and
salting on Lake Tahoe, with important implications on
water quality.
2J.8
Sources of Ambient Fine Particulate Matter at Two
Community Sites in Detroit, Michigan. DAVYDA
HAMMOND, Timothy Dvonch, Gerald Keeler, James Barres,
Ali Kamal, Edith Parker, Wilma Brakefield-Caldwell,
University of Michigan; Fuyuen Yip, National Center for
Environmental Health, CDC.
Detroit, Michigan is a non-attainment zone of the annual
PM2.5 National Ambient Air Quality Standard (NAAQS),
and contains a host of local pollution contributors
including several automotive factories, multiple
manufacturing plants, and high diesel traffic from a
nearby international border crossing. When an area
experiences high particulate concentrations, particularly
when the concentrations are in violation of the NAAQS,
identification of the contributing emission sources aids
researchers in accessing source-specific health impacts
and assists policymakers in developing effective control
strategies. A source apportionment analysis was
conducted using PM2.5 data collected from 1999 to 2002
by the Community Action Against Asthma (CAAA)
project in Detroit, Michigan. CAAA uses a communitybased participatory research approach to identify and
address the environmental triggers for asthma among
children residing in southwest and east Detroit. The
partnership, established in 1998, is comprised of
representatives from community-based organizations,
health agencies and academia, and is affiliated with the
Detroit Community-Academic Urban Research Center.
The data used for the study included 24-hour
measurements of PM2.5 mass, elemental and organic
carbon, and a suite of trace element species. Positive
matrix factorization (PMF) was used to quantitatively
apportion the sources of ambient PM2.5 at each of two
Detroit community sites. Results show that southwest
Detroit PM2.5 levels can be described by seven source
categories: coal combustion, gasoline vehicles, diesel
vehicles, refinery/oil combustion, iron-steel
manufacturing/waste incineration, automotive
electroplating, and crustal/sewage sludge incineration.
The PMF model apportioned the east Detroit PM2.5 data
into five source categories: coal combustion, motor
vehicles/combustion, refinery/oil combustion, iron-steel
manufacturing/waste incineration, and automotive
electroplating. For both locations, approximately 60% of
the PM2.5 mass was attributed to coal combustion sources,
30% to vehicular sources, and 5-7% to local industrial
sources. The unexplained variance in the data accounted
for 3-5% of the PM2.5 mass.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.9
Source Apportionment of PM10 at Santiago, Chile.
HECTOR JORQUERA and Luis Cifuentes, Universidad
Catolica de Chile.
Santiago, Chile is one of the most polluted cities in Latin
America. Although PM10 and PM2.5 levels have
decreased since 1990, in the last 5-6 years that downward
trend has slowed down, and average PM2.5 and PM10
annual ambient levels currently exceed 30 and 65 (ug/
m3), respectively. To understand what sources are
currently responsible for the ambient PM impacts, we
have conducted a source apportionment by using Positive
Matrix Factorization. Ambient samples of fine and coarse
fractions of PM10 were analized by XRF for elemental
composition. We chose two sites for analysis: one located
within the downtown area, surrounded by traffic and
industrial sources, and another one in a suburban area.
Results show that the major contributors to fine particles
are motor vehicles, sulfates and residential combustion,
with 50, 20 and 10%, respectively, at the downtown site.
At the suburban site the same sources dominate with 55,
10 and 15%, respectively. Street dust and marine aerosol
contribute each around 5%. Coarse particles are
dominated by dust street, followed by construction
activities and fugitive dust from regional sources.
Comparisons are made with a similar source
apportionment campaign conducted in 1999, to assess
trends in source contributions. There is a clear need for
additional regulations to curb down ambient PM10 and
PM2.5 levels in the city.
2J.10
Identifying the Impact of Local and Regional Sources of
Fine Particles and Hazardous Air Pollutants in the Midwest:
An Observation-Based Approach. Soner Yorgun, BIRNUR
BUZCU-GUVEN, Michigan State University.
Previous studies identified regional PM sources as a
group, especially the coal combustion operations located
along Ohio River Valley on the air quality of the
Midwestern US. However, this region encompasses the
locations of many large electricity generating power
plants and industrial sources and the relative contributions
of individual sources have not been identified and
emissions from specific facilities have not been
pinpointed to date. The overall objective of this study is to
identify and quantify the largest contributors to local and
regional air pollution specifically individual coal
combustion operations concentrated along the Mid-Ohio
River Valley. The combination of multivariate receptor
modeling techniques with meteorological data allowed a
unique resolution of distinct sources of aforementioned
pollutants and will help locate the local hot spots caused
by local large sources, mostly coal-fired power plants. We
modified two receptor models, Positive Matrix
Factorization (PMF) and Multilinear Engine (ME), to use
both particle and gaseous species to track the mass
contribution of emissions source categories and source
regions. We extended the PMF analysis to include
temperature-resolved organic and elemental carbon
fractions, air toxics and gaseous pollutants to enhance the
source separation. We explored several new support tools,
including nonparametric regression to find the locations
of the largest sources of these pollutants in the region
using the results of PMF analysis and meteorological
parameters.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2J.11
PM 2.5 Source Apportionment for the Chemical Speciation
Trends Network (STN) Site at Birmingham, Alabama.
KARSTEN BAUMANN, Atmospheric Research & Analysis,
Inc.; James B. Flanagan, R.K.M. Jayanty, RTI International.
A systematic approach employing the Positive Matrix
Factorization (PMF) receptor model and data from the
PM2.5 Chemical STN was used to estimate source
contributions to ambient PM2.5 in a highly industrialized
urban setting in the southeastern US. Model results
consistently resolved 10 factors representing 2 secondary,
5 industrial, 1 motor vehicle, 1 road dust, and 1 biomass
burning sources. Estimation of primary organic carbon
(POC) from seasonal application of the elemental carbon
(EC) tracer method significantly improved the model's
performance. Uniform increase of input data uncertainty
and exclusion of a few specially identified outlier samples
further improved the model results, explaining 97% of the
measured total PM 2.5 mass at a minimal intercept of 0.25
-3
2
micro-g m and an r of 0.96.
Particles formed by secondary atmospheric processes,
such as sulfate and secondary organic carbon (SOC,
estimated as difference between measured organic carbon
OC and POC) combined, contribute the majority of
ambient PM2.5 with strong seasonality. Motor vehicle
emissions constitute the biggest primary PM2.5 mass
contribution with almost 25 +/- 2% long-term average and
winter maximum of 29 +/- 11 %. PM2.5 contributions from
the five identified industrial sources vary little with
season and average 14 +/- 1.3%.
The approach described here can provide guidance for
state and local agencies on effectively using STN data for
identifying the most effective emissions reduction efforts.
In the case of Birmingham, AL, such efforts should focus
on traffic and certain local industrial sources. This work
demonstrates that STN data are of sufficient quality to
support modeling studies of this kind, despite issues such
as lack of blank correction for the OC fraction and low
sampler flow rates. It also shows how uncertainty values
provided by the STN dataset can be increased to yield
optimum modeling results.
2J.12
Source Apportionment of PM2.5 Using Chemical Mass
Balance and Positive Matrix Factorization at an
Industrialized City in Northern British Columbia. Juli I.
Rubin, STEVEN G. BROWN, Hilary R. Hafner, Paul T.
Roberts, Sonoma Technology, Inc.; Mark Graham, BC Ministry
of Water, Land, & Air Protection.
Two techniques commonly used for source apportionment
of speciated PM2.5 data are chemical mass balance (CMB)
modeling and positive matrix factorization (PMF). While
both analyses have been successfully used for source
apportionment, each has limitations. With CMB, sources
that impact the monitor along with their emission profiles,
must be known prior to conducting the analysis;
otherwise, a source contribution cannot be quantified.
Additionally, species are assumed to add linearly (no
reactive loss) and source emissions are assumed to be
constant over time in the model. While PMF does not
require source profiles, the factors produced are not
always representative of a single source and can be
difficult to interpret. Limitations in each technique can be
overcome by conducting both analyses on a data set, with
overlapping results providing additional confidence.
CMB and PMF were conducted on speciated PM2.5 data at
a site in Prince George, British Columbia. Prince George
was an ideal location to compare the two methods
because local point sources are dominated by a single
industry (paper/pulp mills), site specific profiles are
available, the monitor is close to industry and highways,
and unique tracers were measured. CMB and PMF were
applied to a 138 sample data set measured with a 1-in-3
day frequency from December 26, 2004, to March 27,
2006, at a downtown Prince George site. CMB was
conducted with 11 source profiles, including pulp mill,
burning, light-duty gasoline vehicle, heavy-duty diesel
vehicle, hog-fuel boiler, asphalt, soil, ammonium sulfate,
ammonium nitrate, and road salt profiles. PMF was
conducted on the same data set with seven to nine factors.
Source contributions were compared on an average and
day-to-day basis. Source apportionment results were
consistent across methods, showing pulp mills, burning,
and light-duty gasoline vehicles as the major contributors
to PM2.5 levels in the area.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.1
Characterization of carbonaceous particle emissions by
waste water treatment plants. PIERRE HERCKES, Zhuo
Chen, Paul Westerhoff, Arizona State University.
2K.2
Seasonal and diurnal variations in water soluble inorganic
fine particulate matter and associated gas precursors.
KRYSTAL J. GODRI, Greg J. Evans, University of Toronto.
As part of waste water treatment, biomass is kept in
suspension in so-called aeration basins through bubbling
of air through the waste water. In these activated sludge
reactors, the bubbling process might lead to the
aerosolization of the liquid, in a similar fashion than sea
salt formation in the marine environment. However, in a
waste water treatment plant this would lead to the
formation of particulate material containing a variety of
benign and malicious micro-organisms as well as organic
species like pharmaceuticals, personal care products or
human fecal markers.
From June 2006 to May 2007, water soluble inorganic
2+
aerosols, including Cl , SO4 , NO3 , NO2 and NH4 , and
their associated precursor gas (HCl, SO2, HNO 3, HNO 2,
NH3) concentrations were measured adjacent to a hightraffic street in downtown Toronto, Canada. Semicontinuous measurements averaged over 15 minute
intervals were performed with a Dionex Gas Particle Ion
Chromatograph. The dataset was analyzed for seasonal
and diurnal variation for each pair of inorganic aerosol
and its associated precursor gas. Particulate nitrate and
HNO3 exhibited a seasonal trend. Local gaseous vehicle
emissions contributed to HNO3 and particulate nitrate
production in the winter. Low temperatures and high
relative humidity induced gaseous HNO3 and NH3
condensation yielding NH4NO3 aerosol. Consequently,
low fractions of total nitrate (TNO3=HNO3+pNO3-) were
measured in the gas phase during winter months.
Ammonia and particulate ammonium also demonstrate
seasonal diurnal differences. During the summer, NH3
exhibited a morning rush hour maxima on weekdays. The
progression from summer to winter shifted the morning
maxima to an afternoon/evening diurnal peak and the
overall magnitude of NH3 concentrations also decreased.
Ammonium only demonstrated diurnal variation in the
winter when particulate nitrate concentrations are at a
maximum. Particulate sulphate and SO2 showed no
diurnal variation regardless of season suggesting
dominate transport from regional sources throughout the
2year. The frequency of SO2, and in particular SO4
episodes declined in the winter as did the magnitude of
the concentrations measured for each species; less
2efficient oxidation of SO2 yielded higher winter SO2/SO 4
ratios. Toronto's geographic location allows for low
ambient chloride concentrations. However, roadway
salting in the winter caused both gas and particulate phase
chloride episodes. Early morning chloride particulate
peaks were seen daily in the summer and were attributed
to the lawn sprinkler system situated close to the sampling
inlet. Highly acidic aerosols were associated with the
summer months while a shift towards neutralization
occurred as the temperature declined.
To test for emissions of activated sludge reactors, we
sampled fine and coarse aerosol particles above an
aeration basin. Aerosol as well as waste water and
biomass samples were analyzed by liquid chromatography
coupled to tandem mass spectrometry (LC/MS/MS) and
gas chromatography coupled to mass spectrometry (GC/
MS) for individual organic species.
Our results show that the particulate matter above the
aeration basins contains a substantial amount of species
common in waste water and not usually found in the
atmosphere. As an example, beta-estradiol, a female
hormone, was detected in both coarse and fine fraction at
concentrations up to 0.5ng/m3. Other species present in
biomass and being emitted into the atmosphere included
caffeine, sterols like cholesterol, musk and fragnance
components like galaxolide and a series of
pharmaceuticals.
The knowledge of aeration rates allows for the estimation
of emission rates and these results will be discussed as
well as the potential to affect molecular marker source
apportionment studies.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.3
Size Distribution of Particulate Metals in Central California.
WALTER A HAM, Michael J Kleeman, University of
California, Davis.
Epidemiological studies have linked the inhalation of
particulate matter (PM) to increased morbidity and
mortality. While the toxicological mode of action of PM
remains uncertain, some hypotheses suggest that the size
and composition of PM may be the best indicator for the
potential for toxicity. Typically, healthy human bodies
can readily remove coarse PM (2.5 microns < Dp < 10
microns). However, it has been observed that fine PM
(Dp < 2.5 microns) is not easily removed thus allowing
more time for the chemical components of the PM to react
with biological tissue. The metal content of airborne PM
is of particular interest due to the fact that certain metals
have been associated with cancer, protein synthesis
inhibition, and a number of other health effects.
Therefore, the size distribution and metal content of
airborne PM must be measured to fully evaluate the
potential for human toxicity of these particles.
In this study, the size and composition of urban
particulate matter are reported from two field exposure
studies. Field measurements and particulate collection
were performed during summer and winter campaigns in
Fresno, CA in 2006 and 2007. Micro-Orifice Uniform
Deposit Impactors were used to collect urban PM in six
size fractions between 0.056-1.8 Вµm particle diameters on
Teflon substrate. Scanning Mobility Particle Size (SMPS)
and Aerodynamic Particle Size (APS) instruments were
simultaneously used to obtain real-time size distribution
data during these sampling events. A number of metal
ions associated with respiratory toxicity including but not
limited to 34S, 55Mn, 69Ga, 75Ar, 111Cd, 118Sn, and
137Ba will be measured and quantified in these sizeresolved samples using Inductively Coupled Plasma Mass
Spectrometry (ICP-MS). Size-resolved chemical
composition data and real-time size distribution
measurements will be presented and possible
toxicological implications will be discussed.
2K.4
Diagnosis of an Aged Prescribed Fire Plume Hitting an
Urban Area. SANGIL LEE, Hyeon Kook Kim, Evan Cobb,
Sara Nichols, Nick Culpepper, Michael Chamber, Eric S.
Edgerton, John J. Jansen, Armistead G. Russell, Georgia
Institute of Technology
On February 28th, 2007, an unplanned for shift in
winds led to the plume of a 3000 acre prescribed burn to
impact Atlanta, GA, starting about 17:00. Observed 1-hr
PM2.5 concentrations at several monitors in the city
increased from less than 10 to greater than 145 microgram
per m3 in the matter of a couple of hours (U.S. National
Ambient Air Quality Standard (NAAQS) for 24-hr PM2.5
is 35 microgram per m3). Ozone concentrations also
jumped by up to 20 ppb, in spite of the late hour and
being during the winter. Such increases in pollutant
levels are expected to lead to health impacts, both from
increased acute exposure to PM and ozone.
While the event is unfortunate in terms of potential
health impacts, it was well captured by the variety of PM
and gas-phase species air quality monitors (e.g., from
STN, SEARCH and ASACA networks) around the city,
allowing for improved understanding of impacts from
biomass burning in general, and prescribed burning in
particular. Of special interest is the opportunity to capture
the composition of an aged plume, as the burn site was
approximately 70 km from the city, allowing the plume to
age before reaching Atlanta.
PM2.5 chemical speciation data from 24-hr filter
samples suggest that the elevated PM2.5 concentrations
are driven mainly by distinctive increases in organic
carbon (OC) concentration. 70 % of the increase in OC is
attributed to increased water soluble organic carbon
(WSOC). Nitrate concentrations also increased, due to
NOx and NH3 emissions from prescribed burning. K and
Cl concentrations, two major elements of PM2.5 from
prescribed burning, are significantly higher during the
smoke event across three monitoring sites compared to
both the previous and following days, while other
elements do not show enhanced concentrations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.5
Predicting near real-time PM2.5 FRM Concentrations from
Continuous Mass and Species Measurements in New York
City. DIRK H. FELTON, Oliver V. Rattigan, New York State
Department of Environmental Conservation; James J. Schwab,
Kenneth L. Demerjian, University at Albany, SUNY.
Near real-time PM2.5 mass data is used routinely by air
monitoring agencies to provide air quality information
and health warnings related to pollutant concentrations to
the public. This data is presented in concentrations,
which are as similar as possible to FRM filter data. Since
the FRM data is not available for several months
following sampling agencies generally use statistical
correlations to produce \FRM-like\ data from continuous
mass instruments. Measurements from collocated semicontinuous sulfate, nitrate, elemental carbon (EC) and
organic carbon (OC) instruments can be used in
conjunction with TEOM mass and meteorological data to
produce PM2.5 data that is more reflective of actual FRM
measurements.
For cities in the Northeastern United States, the data from
TEOM instruments that operate at 50 deg C is not
adequate to produce data for public reporting purposes.
The heated TEOM sampling inlet causes a negative bias
in relation to the FRM, which varies seasonally. This
work demonstrates one method to calculate hourly PM2.5
mass that includes estimates of the fractions of individual
PM2.5 species retained on the FRM filter. This allows
the technique to account for the seasonal bias between the
FRM and TEOM continuous mass measurements. The
continuous species data used in this study are correlated
to 24-hr integrated 1 in 3 day filter measurements from
the Speciation Trends Network.
2K.6
Fine, Ultrafine And Nanoparticle Trace Organic
Compositions Near A Major Freeway With A High Heavy
Duty Diesel Fraction. ZHI NING, Michael D. Geller and
Constantinos Sioutas, University of Southern California.
Individual organic compounds such as hopanes and
steranes (originating in lube oil) and selected polycyclic
aromatic compounds (PAHs) (generated via combustion)
found in particulate emissions from vehicles have proven
useful in source apportionment of ambient particulate
matter. Currently, little ambient data exists for a majority
of these species. Trace organic species in the sizesegregated ultrafine (<0.18 micro-meter) and
accumulation (0.18-2.5 micro-meter) particulate matter
(PM) modes were measured during the winter season next
to a busy Southern California freeway with significant
(~20%) diesel traffic. The ultrafine mode was further
segregated into 4 size ranges (18-32 nm, 32-56 nm, 56
-100 nm, and 100-180nm) with a NanoMOUDI lowpressure cascade impactor sampler. Both ambient and
concentrated size-segregated impactor samples were
taken in order to collect enough mass for chemical
analysis. Chemical composition of accumulation and
ultrafine mode particles, including four size ranges, were
analyzed. Particle acidity was also investigated by the
ratio of measured and required ammonium for
neutralization with nitrate and sulfate. All the measured
organic species exhibited decreasing concentrations with
size in the ultrafine mode. The most abundant PAHs in
the ultrafine and accumulation modes were pyrene and
benzo(ghi)perylene, and norhopane dominated the
hopanes and steranes. This study is the first to present
size-segregated organics species in an ambient
environment.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.7
Compositions of the Major Chemical Constituents of PM2.5
in Korea. YOUNG-JI HAN, Jin-Hee Jung, Sun-Young, Kan,
Kangwon National University; Jong-Bae Huh, Seung-Muk Yi,
Seoul National University.
The PM2.5 samples were collected from November 2005
through October 2006 at the two sites in Korea, in order
to investigate the characteristics of the major components
in PM2.5. One monitoring site was located in Chuncheon,
which is a relatively small town, and the other sampler
was deployed in Seoul, the biggest metropolitan area in
Korea. The average PM2.5 concentration in Chuncheon
3
3
and Seoul were 36Вµg/m and 40Вµg/m , respectively,
which is almost three times higher than the US NAAQS
3
annual PM2.5 standard of 15Вµg/m . The average PM2.5
concentration was highest in winter followed by spring,
fall, and summer. This higher concentration in winter was
most likely due to the combination of increased emissions
from combustion sources and the lower mixing heights.
In addition, large portion of fine particles could be
removed through wet deposition in summer, resulting low
PM2.5 concentrations in summer. Ionic constituents were
analyzed using ion chromatography, and the biggest anion
and cation constituents were sulfate and ammonium,
respectively, at both sites. Carbonaceous constituents
including elemental and organic carbons were also
analyzed using IMPROVE method. The mean
concentration of OC tended to be higher during the winter
than the summer at both sites. Contribution of
carbonaceous compounds to PM2.5 was much higher in
Chuncheon than in Seoul while PM2.5 in Seoul was
significantly occupied by ionic constituents. The reasons
of higher concentrations of carbonaceous compounds in
Chuncheon than in Seoul are being investigated currently.
The ratio of secondary organic carbon (SOC) to the total
OC appeared higher in Seoul than in Chuncheon.
During the sampling period, six yellow sand events were
observed, and the PM2.5 concentrations were about two
to three times higher than those observed during nonyellow sand events. These dramatically increased PM2.5
concentrations during yellow-sand events were never
observed in previous years before 2006.
2K.8
Integrated and Semi-Continuous Mass and Chemical
Species Measurements for both Fine and Coarse Particles in
Lindon, UT. BRETT D. GROVER, Russell W. Long, Robert
W. Vanderpool U.S. Environmental Protection Agency,
National Exposure and Research Laboratory; Robert W.
Murdoch, RTI International; Delbert J. Eatough, Brigham
Young University.
An extensive field sampling campaign was conducted in
Lindon, UT during January - February, 2007. During this
time period, Utah County experienced a severe inversion
with particulate matter concentrations well above
National Ambient Air Quality Standards (NAAQS). Both
integrated and semi-continuous samplers were employed
during the sampling campaign for the measurement of
particulate mass and chemical species for both fine and
coarse particles. Integrated mass measurements were
made with a Dichotomous Sampler (PM10-2.5 and
PM2.5) and two FRM samplers (PM10 and PM2.5).
Semi-continuous mass measurements were made with an
R&P FDMS (PM2.5) and a GRIMM monitor (PM10,
PM10-2.5, and PM2.5). Comparisons between the mass
measurements of the different size fractions of integrated
samplers have traditionally shown good agreement, as
was observed in this study. Comparisons between
integrated mass measurements and semi-continuous mass
measurements will be presented. Also included in the
sampling campaign were semi-continuous instruments to
measure atmospheric inorganic species for both PM10
and PM2.5, based on parallel-plate wet wall denuder
(PPWD) - ion chromatography (IC) technology. These
data will be used to elucidate specific atmospherically
interesting episodes related to the study period and to
explain any differences observed between mass
measurement techniques.
DISCLAIMER
Although this work was reviewed by EPA and approved
for publication, it may not necessarily reflect official
Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.9
Characterization of the chemical compositions in PM2.5 in
Seoul - relationship between indoor and outdoor. BO-RA
CHOI, Jong-Bae Huh, Hyun-Sun Kim, Kye-Seon Kim, SeungMuk Yi, Seoul National University.
Each Fifty four chemically speciated samples were
collected on the roof of the School of Public Health
building of Seoul National University and underground
shopping center in Seoul, Korea from December 2005
through December 2006. The objectives of this study
were to investigate the characteristics of the major
components in indoor and outdoor PM2.5 and to identify
the source types and contributions for indoor and outdoor
PM2.5.
The average concentrations of PM2.5 and its components
measured at indoor site were higher than those at outdoor
site. The average concentrations of outdoor and indoor
3
PM2.5 were 49.24 microgram per m and 65.9 microgram
3
per m , respectively.
2
+
The seasonal concentrations of SO4 -, NO3 and NH4
were significantly different. During the summer the
average sulfate concentration was higher than that during
the other seasons due to the increased photochemical
reaction. The nitrate concentration was the highest in
winter due to the lower temperature and higher humidity.
The same seasonal variations of PM2.5 and its
components were observed at indoor site. The
ammonium concentrations measured at outdoor were the
highest during the winter, while those at indoor were
highest during the summer.
The PMF results for indoor resolved 91% of PM2.5 mass
concentration and indicated that six sources independently
contributed to the PM2.5: road dust (18.6%), secondary
sulfate (23.2%), secondary nitrate (17.4%), motor vehicle
(11.3%), internal source 1 (13.5%), and internal source 2
(7.7%).
2K.10
Organic Aerosol Analysis with the Aerodyne High
Resolution Time-of Flight Aerosol Mass Spectrometer (HRToF-AMS) at T0 in Mexico City during MILAGRO /
MCMA-2006. ALLISON C. AIKEN, Michael Cubison, J. Alex
Huffman, Peter F. DeCarlo, Ingrid Ulbrich, Ken Docherty,
Donna Sueper, Jose L. Jimenez, University of Colorado at
Boulder; Dara Salcedo, Universidad Aut
Non-refractory sub-micron (approx. PM1) ambient aerosol
was analyzed from March 10 - March 30, 2006 in Mexico
City at the T0 (IMP) site with a new version of the
Aerodyne Aerosol Mass Spectrometer, the HighResolution Time-of-Flight Aerosol Mass Spectrometer
(HR-ToF-AMS, DeCarlo et al., 2006). Time series of
mass concentrations of inorganic species (Ammonium,
Chloride, Nitrate, Sulfate) and of the Organic fraction
along with averaged AM and PM size distributions are
presented and compared with results from MCMA-2003
with a Quadrupole AMS (Q-AMS, Salcedo et al., 2006).
The HR-ToF-AMS has the ability to resolve the elemental
composition of most mass fragments, especially for the
low m/z (below 100) where the majority of the signal in
the AMS occurs when using electron impact ionization
(EI). Organic mass spectra below m/z 100 have been
+
+
separated into four fragment types (CxHy , CxHyOz ,
+
+
CxHyNz , CwHxNyOz ), while still retaining quantitative
mass concentrations, and O/C ratios have been computed
for total organics. Additionally, Positive Matrix
Factorization (PMF) has been used to analyze the
components of the organic mass fraction. Primary
emissions and SOA formation are important for this
dataset, while the impact of large biomass burning plumes
appears to be more episodic at this ground site. Organic
amines are observed in the aerosol during some mornings.
References
P.F. DeCarlo, J.R. Kimmel, A. Trimborn, M.J. Northway,
J.T. Jayne, A.C. Aiken, M. Gonin, K. Fuhrer, T. Horvath,
K. Docherty, D.R. Worsnop, and J.L. Jimenez. FieldDeployable, High-Resolution, Time-of-Flight Aerosol
Mass Spectrometer. Analytical Chemistry, 78: 8281-8289,
2006.
D. Salcedo et al. Characterization of ambient aerosols in
Mexico City during the MCMA-2003 campaign with
Aerosol Mass Spectrometry: results from the CENICA
Supersite. Atmospheric Chemistry and Physics, 6, 925
-946, 2006.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.11
Lead Isotope Abundance Ratios for Ambient Particulate
Matter in St. Louis. JAY TURNER, Washington University in
St. Louis; Judith Chow, John Watson, Desert Research Institute.
Lead isotope analysis was performed on ambient
particulate matter samples collected under the St. Louis Midwest Supersite program. Daily 24-hour integrated
samples for PM2.5 and PM10 were collected from April 13,
2001 through September 25, 2003 at East St. Louis, IL,
and August 17, 2001 through November 20, 2001 at Park
Hills, MO. 116 of these filter samples were analyzed for
four lead isotopes (206Pb, 206Pb, 207Pb and 208Pb) using high
resolution inductively coupled plasma - mass
spectrometry (ICP-MS). The analysis set included 92
ambient particulate matter samples (33 PM2.5 and 33 PM10
at East St. Louis, 13 PM 2.5 and 13 PM10 at Park Hills), 16
collocated samples (8 runs x 2 samples) to determine
measurement precision, and 8 field blank samples to
determine practical detection limits. All of these samples
were analyzed by X-ray fluorescence (XRF) as a quality
control measure.
Lead isotope abundance ratios exhibited significant
sample-to sample variation. Ambient particulate matter
207
206
208
206
Pb/ Pb and Pb/ Pb isotope abundance ratios are
not adequately described by a single two-member mixing
model. There appear to be at least three distinct sources
of airborne lead, with numerous samples falling along one
of two mixing lines between the end-members
representing these source categories. Most of the
remaining samples fall within the domain between these
two mixing lines. Lead isotope analysis was performed
for paired PM2.5 and PM10 samples to investigate the
fractionation between fine and coarse (by difference)
particle size modes. In almost all cases, the 207Pb/206Pb
and 208Pb/206Pb ratios were greater for PM 2.5 than PM10.
Coarse particle lead isotope abundance ratios at these sites
are closer to Viburnum Ores than the fine particle lead
isotope abundance ratios. This trend suggests
contamination of soil in the St. Louis area from locally
mined, smelted and refined lead.
2K.12
Eddy Covariance Flux Measurements of Urban Aerosols
and Related Urban Gaseous Pollutants During the
MILAGRO Mexico City Field Campaign. RASA
GRIVICKE, Shelley Pressley, Gene Allwine, Tom Jobson, Hal
Westberg, and Brian Lamb, Washington State University; JoseLuis Jimenez, University of Colorado; Eiko Nemitz, Centre for
Ecology and Hydrology Edinburgh; Liz Alexander,
Environmental Molecular Sciences Laboratory PNNL; Erik
Velasco and Luisa Molina, Molina Center for Energy and the
Environment; Rafael Ramos, SIMAT.
During the Mexico City MILAGRO field campaign, an
urban flux tower was operated from a rooftop near central
Mexico City to measure fluxes of urban gas and aerosol
species. The measurement height was 42 m in an area
with a relatively homogeneous urban landscape. An
Aerodyne quadrupole aerosol mass spectrometer (QAMS) was operated to measure aerosol concentrations (1
min. averages) and aerosol fluxes (10 Hz, selected ion
monitoring) on an alternating 30 minute schedule. The
aerosol fluxes were derived from eddy covariance
calculations. These data are supported by additional gas
phase flux measurements for CO2 and a number of VOC
gaseous species using a combination of techniques,
including Proton Transfer Reaction Mass Spectrometry
using the disjunct eddy covariance technique and GC-FID
analysis of samples from a disjunct eddy accumulation
sampler. Surface energy fluxes were also measured over
the urban landscape. Preliminary results from the AMS
concentration and flux measurements will be presented in
the context of the other pollutant and energy flux data.
These data indicate that the urban landscape is a
significant source for organic aerosols.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
2K.13
Temporal Characterization of Individual Ambient Particles
by using an Aerosol Time-of-Flight Mass Spectrometer
(ATOFMS) in Toronto, Canada. CHEOL-HEON JEONG,
Greg J. Evans, Krystal Godri, Andrew Knox, University of
Toronto.
An Aerosol Time-of-Flight Mass Spectrometer
(ATOFMS, TSI 3801) was deployed in downtown
Toronto during June 2006 and January 2007 as a part of
the Seasonal Particulate Observation in Regional Toronto
(SPORT) campaign. The ATOFMS measures both the
aerodynamic size (0.3 to 3 micro-meter) and chemical
information on individual particles in real time basis using
laser desorption ionization dual polarity time-of-flight
mass spectrometry. During the SPORT campaigns, the
ATOFMS was deployed with other co-located chemical
speciation instruments. In the study, size resolved
continuous mass concentrations of nitrate, sulfate,
ammonium, organic carbon, elemental carbon, and
chloride were obtained by estimating peak intensities of
marker species measured by the ATOFMS. A
comparison will be made between the quantitative mass
concentrations from the ATOFMS and several high-time
resolution species concentrations measured by a GasParticle Ion Chromatograph (GP-IC), a Sunset lab OC/EC
analyzer, an R&P Nitrate monitor. The implications of
these findings will be discussed. Furthermore, in order to
determine common patterns of particle compositions, the
mass spectra obtained by the ATOFMS were classified by
using a clustering tool, Environmental Chemistry through
Intelligent Atmospheric Data Analysis (Enchilada),
developed by a group at Carleton College, MN. The
Enchilada uses several clustering algorithms, such as
ART-2 and K-MEANS to assign individual particles into
specific particle types based on their mass spectral
similarities. The diurnal and seasonal trends of particle
types characterized by the clustering methods will be also
described and discussed to provide information on likely
sources of ambient particles in the urban area.
3D.1
Chemically resolved aerosol emission fluxes above six urban
areas. EIKO NEMITZ, Rick Thomas, Gavin Phillips, Daniela
Famulari, David Fowler, Centre for Ecology and Hydrology,
Edinburgh; Jose Jimenez, Alex Huffmann, University of
Colorado / CIRES; Hugh Coe, Keith Bower, James Allan, Paul
Williams, Manchester University; Shelley Pressley, Brian Lamb,
Washington State University; Erik Velasco, Molina Center for
Energy and Environment; Mikaela Alexander, Pacific Northwest
National Laboratory; Doug Worsnop, Aerodyne Research Inc.
Micrometeorological flux measurements of chemically
resolved sub-micron aerosol fluxes were made with an
eddy-covariance flux system based on an Aerodyne
Aerosol Mass Spectrometer (Q-AMS), over six urban
areas (Boulder, Colorado; Gothenburg, Sweden;
Edinburgh, Manchester and London, all UK, and Mexico
City). This instrument was able to derive fluxes of nitrate,
sulphate and organic aerosol, which were separated into
hydrocarbon-like organic aerosol (HOA) and oxygenated
organic aerosol (OOA). Simultaneous measurements
included fluxes of carbon dioxide (by infrared gas
analyser), and during some studies fluxes of carbon
monoxide (by resonance fluorescence) and particle
numbers (by CPC and optical spectrometer).
The measurements show that sulphate fluxes are very low
in most environments, indicating that city centres are no
major sources of this compound. By contrast, nitrate was
emitted from most cities, but only on certain days,
emphasising the role of meteorology in urban nitrate
formation. HOA was emitted throughout the day,
following the diurnal pattern of traffic activity. Through
comparison with the carbon dioxide fluxes, emission
factors can be estimated. There is clear evidence for OOA
emissions above the cities, indicating that significant
chemical processing occurred within the city, before the
air masses reached the measurement height of 40 to 200
m.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
3D.2
Vertical Profile of PM Size Distribution in Milan (Italy).
Vorne Gianelle, ARPA Lombardia
Particle size distributions have been measured at the
bottom and the top of a 84-m (250 ft) high tower located
in the city centre of Milan. Simultaneous monitoring of
particle number concentration (from 300 nano-meter up to
20 micro-meter of diameter) has been performed by
running parallel two low-volume (1.2 litres per minute)
optical particle counters, which use laser light-scattering
technology for single-particle counts in 15 size range
bins. Concentration data for each size bin have been
collected at 1-minute time resolution during two
monitoring campaigns held in the cold season (13 days in
October 2005) and in the warm season (20 days in MayJune 2006). After processing for outlier detection and
elimination, 1-hour averaged particle number
concentrations have been calculated based on the
resulting data.
Overall particle number concentrations, as well as mass
concentration for PM10, PM2.5 and PM1 derived from
particles' counts, are reported. Size distributions for
number, surface area and volume observed at the towertop (TT) and tower-bottom (TB) monitoring sites are
compared. Cluster analysis is applied in order to group
together size bins characterised by similar patterns.
Regardless of the measurement height, 3 main clusters
can be identified for cold-season data: a cluster for fine
particles (0.3-2 micro-meter size range), one for fine to
coarse particles (2-15 micro-meter size range) and one for
large particles (<15 micro-meter size range); for the warm
season the fine particle cluster is divided into three
separate clusters, so that 5 main clusters are identified. TT
and TB daily patterns of particle number concentrations,
separately evaluated for each cluster, are compared: on
both seasons higher concentrations are observed at the TB
level, with the largest differences in daily patterns for
coarse and large particles, whereas the vertical profile of
the finest particles is rather uniform during the day.
3D.3
Number-based Emission Factors and New Particle
Formation/Growth Events from Mexico City SMPS Data
(MILAGRO). ALICIA PETTIBONE, Charles Stanier,
University of Iowa.
We will discuss the results from the MILAGRO field
campaign in Mexico City during the spring of 2006,
where size distributions from 10-500 nm were measured
using the SMPS at the T0 (Mexican Petroleum Institute)
Site. The Mexico City basin has been shown to have
strong thermally and topographically driven circulations.
Average particle concentrations in the 10-500nm size
range at this site were 10,000-30,000/cm3, with strong
diurnal patterns. In the morning, CO2 and particle count
data are highly correlated. In the afternoon, the CO2
drops during ventilation of the daily polluted layer, and
the coupling between CO2 and particle number breaks
down, with particle number sometimes increasing as CO2
decreases. The connection of this unique meteorology to
new particle formation is potentially important because it
may give clues to (1) the mechanisms at work for
formation and growth of new atmospheric particles, and
(2) help explain the horizontal and vertical extent of new
particle formation in central Mexico. From the ground
site, nucleation and growth events (defined by the
emergence of a clear and growing mode at 10-15nm)
were observed on 4 of the 15 days sampled (27%). New
particle formation events were observed both
simultaneous to and separately from the afternoon
ventilation of primary pollutants, which was quite
dramatic on some days. The commonality among all the
new particle formation and growth events was cleaner
than average number concentrations just prior to the
event. The events are analyzed using a nucleation box
model and compared to gas and aerosol data aloft
(LIDAR, aircraft, and 3D model values).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
3D.4
Highlights of PM2.5 Continuous Speciation Measurements
in New York. OLIVER V RATTIGAN, Dirk H. Felton, New
York State Department of Environmnetal Conservation; James J.
Schwab, Kenneth L. Demerjian, University at Albany, SUNY.
3D.5
Daily Measurements of Speciated PM2.5 in Denver, CO with
Seasonal and Weekly Patterns. STEVEN J DUTTON,
Michael P Hannigan, Shelly L Miller, University of Colorado;
Sverre Vedal, University of Washington.
Continuous multi-pollutant measurements provide
important temporal information for air quality monitoring,
for understanding emission sources and for public health
studies. Aerosol carbon, sulfate and nitrate constitute the
major fractions of PM 2.5 mass in the Eastern United
States. At a site in the South Bronx, NY semi-continuous
measurements of PM2.5 mass, sulfate, nitrate, elemental
carbon (EC), organic carbon (OC) and black carbon (BC)
and trace gas species including SO2, O3 and NOx have
been made for several years. In addition 24-hr integrated
1 in 3 day filter measurements using the Speciation
Trends Network and IMPROVE protocols are also
collected at the site.
The long-term semi-continuous measurements are used to
examine diurnal, day of week and seasonal patterns. The
measurements also provide information on short-term
stagnation plume events. EC, BC and nitrate
concentrations (and OC during winter months) track
throughout the day with peak concentrations in the
morning hours coinciding with the commute period. The
afternoon/evening peak is less pronounced due to a
change in boundary layer height, which leads to a dilution
and dispersion of pollutants. Concentrations of these
species are generally higher on weekdays compared to
weekends with some noticeable seasonal differences. This
pattern is also reflected in the primary pollutant NOX
indicating that local mobile emissions make a strong
contribution to these species in NYC. Although EC and
BC are highly correlated significant biases are observed
particularly during summer months. Throughout the year
OC concentrations are highly correlated with PM2.5 mass,
and sulfate during summer months, indicating that there is
a strong regional contribution to OC measured in NYC. In
this paper highlights of the long-term semi-continuous
measurements including temporal patterns and
comparisons with collocated 24-hr integrated filters will
be presented.
Daily 24-hour composite PM2.5 filter samples have been
collected at a site in Denver, CO for the past 5 years to
support the Denver Aerosol Sources and Health (DASH)
study. Simultaneous collection on both Teflon and quartz
filters has allowed for analysis of PM2.5 mass and
speciation for the following: sulfate and nitrate using ion
chromatography (IC), bulk elemental and organic carbon
using thermal optical transmission (TOT) and detailed
organics using gas chromatography/mass spectrometry
(GC/MS). The first 3.5 years of mass, IC and TOT as
well as the first 6 months of GC/MS data will be
presented along with interesting seasonal and day of the
week patterns for several of the species.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
3D.6
Daily Variation in Chemical Characteristics of Urban
Ultrafine Aerosols and Inference of Their Sources. ZHI
NING, Michael D. Geller, Katharine F. Moore, Constantinos
Sioutas, University of Southern California; Rebecca Sheesley,
James J. Schauer, University of Wisconsin, Madison.
A summer air quality monitoring campaign focusing on
daily variation of ultrafine (< 180 nm in diameter) particle
chemical characteristics was conducted in a typical urban
site in Los Angeles during June "“ July 2006. Ultrafine
particles (UFP) were collected weekly for two 3-hr
periods each day -- one to capture the morning commute
(06:00-09:00 PDT) (Pacific Daylight Time) and one to
investigate photochemically-altered particles (13:00
-16:00 PDT). Samples were analyzed for ionic
compounds, metals, trace elements, elemental carbon, and
organic carbon. In addition, measurements of individual
organic species and their variation with time of day at the
urban site were conducted. The relative abundances of
alkanes, PAH, and hopanes in the morning denote a
strong influence of commute traffic emissions on ultrafine
particle concentrations. By contrast, afternoon
concentrations of oxygenated organic acids and sulfate
rose, while other species were diluted by increased
mixing height or lost due to increasing temperature. These
are clear indicators that secondary photochemical
reactions are a major formation mechanism of ultrafine
aerosols in the afternoon. The concentrations of organic
species originating from vehicular emissions measured in
this study compare favorably to those from freewayadjacent measurements by using CO2 concentrations to
adjust for dilution, demonstrating the effectiveness of this
tool for relating sites affected by vehicular emissions."
9A.1
Characterization, Seasonality and Source Apportionment of
Fine Particulate Organic Matter at Urban and Rural Sites
During TexAQS II. Matthew Fraser, SHAGUN BHAT, Rice
University.
Ambient samples of fine particulate matter with diameter,
dp < 2.5 micro-meter, were collected in Dallas, Texas and
San Augustine, Texas as part of the Texas Air Quality
Study (TexAQS II) from January 2006 - August 2006.
Solvent extractable non-polar and polar compounds were
extracted from quartz filters. Concentrations of n-alkanes,
hopanes, n-alkanoic acids, n-alkenoic acid, levoglucosan
and pinic and pinonic acid were measured from all
samples and exhibit great seasonal variation at both sites.
A total of 19 molecular markers were measured every
third day over a 7-month period. Average measured
concentrations of particulate n-alkanes (C25 to C35)
ranged from 2-10 ng m-3 for San Augustine, which is a
rural site and 4.5 - 17 ng m-3 for Dallas. The Carbon
Preference Index (CPI) of ~ 1.0 for Dallas suggests that
motor vehicle exhaust is the major source of n-alkanes in
ambient fine particulate organic matter. For polar
compounds, weekday concentrations exceeded weekend
concentrations for both sites. Alpha-pinene
photooxidation products, namely pinic acid and pinonic
acid were also quantified from ambient samples. The
average concentrations for pinic acid were 53 ng m-3 and
18 ng m-3 in San Augustine and Dallas respectively. The
presence of these oxidation products in samples points
towards secondary aerosol formation from alpha-pinene
in the atmosphere. Chemical Mass Balance (CMB-8)
model was used to apportion fine particulate organic
matter using polar and non-polar species concentrations
derived from GC-MS analysis of PM2.5 samples.
Source profiles were generated using molecular marker
concentrations in the multivariate receptor model,
Positive Matrix Factorization (PMF). Major contributors
to organic carbon were gasoline-powered vehicles, diesel
exhaust, wood combustion and meat cooking.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
9A.2
Receptor Modelling of Chemically Speciated Aerosols
Sampled with High Time Resolution by an Aerosol Mass
Spectrometer and a Semi-Continuous Elements in Aerosol
System. MAYGAN MCGUIRE, Greg. J. Evans, Cheol-Heon
Jong, University of Toronto; Jeffrey Brook, Gang Lu,
Environment Canada; John Ondov, University of Maryland.
Windsor, Ontario, situated adjacent to Detroit, Michigan
along the Canada/US border, frequently experiences
episodes of poor air quality. A three week intensive air
sampling campaign performed using Environment
Canada's CRUISER mobile air pollution laboratory
during January and February 2005 sought out to
determine the local and regional contributions leading to
elevated aerosol levels in winter months. Expected local
particulate sources included continuous trans-border and
local traffic, as well as industrial complexes in both
Windsor and Detroit; expected regional sources included
coal-fired power generating stations and petroleum
refineries as well as the regional traffic contribution.
CRUISER was equipped with a suite of high time
resolution instruments: an Aerosol Mass Spectrometer
(AMS) sampling every 15 minutes, an Aethelometer, a
GRIMM dust monitor as well as a Condensation Particle
Counter. Gas monitors measured NOx, SO2, O3 and CO
concentrations. Trace refractory elemental contributions
were collected by a Semi-Continuous Elements in
Aerosol System (SEAS) every half-hour and stored for
off-site Inductively Coupled Plasma Mass Spectroscopy.
Receptor modelling using PMF 2 was first performed on
the mass spectra of the non-refractory aerosol components
generated by the AMS. Trace refractory elemental
concentrations collected using the SEAS were then
incorporated into the factor analysis; some of these trace
refractory elements served as marker species in the factor
analysis which were combined with the non-refractory
components to further factor resolution. Factor
identification was performed by comparing the resolved
source profiles from both approaches with reference
spectra. The factor contributions were also compared to
gas and particle concentrations, meteorological trends and
air mass back-trajectories generated using the NOAA
HYSPLIT model. The two factor analysis approaches
will be contrasted and the results of the receptor
modelling on the Windsor airshed will be presented.
9A.3
Source Apportionment of the Particulate Organic Mass
During Winter and Summer in Zurich, Switzerland.
ANDRE S.H. PREVOT, M. Rami Alfarra, Jisca Sandradewi,
Silke Weimer, Nolwenn Perron, Urs Baltensperger, Paul
Scherrer Institute, Switzerland; Valentin Lanz, Christoph
Hueglin, Swiss Federal Laboratories for Materials Testing and
Research, Empa, Switzerland; Soenke Szidat, University of
Bern, Switzerland.
Field campaigns of several weeks including aerosol mass
14
spectrometer (AMS), C, EC/OC, and other
measurements were performed in summer and winter in
an urban background station in Zurich, Switzerland. The
organic aerosol mass spectra were used to perform
positive matrix factorization (PMF) and hybrid statistical
methods between chemical mass balance and PMF. In
summer 6 different sources could be distinguished while
in winter 3 sources (primary traffic, wood burning, and
secondary organic aerosols) could be distinguished. Both
in summer and winter, primary traffic emissions
contribute less than 15% to the organic mass (OM). Wood
burning contributes in summer around 10% and in winter
more than 30%. Oxidized organic aerosol contributes
(probably mostly secondary organic aerosol) most to OM.
14
Combining the statistical analyses with the C analysis
reveal that most of the secondary organic aerosol is nonfossil both in summer and winter. In addition to the
studies in Zurich, a comparison to measurements at a
highway site will be shown.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
9A.4
Source Apportionment of Ultrafine Airborne Particulate
Matter During a Winter Pollution Episode. MICHAEL J.
KLEEMAN, Sarah G. Riddle, Michael A. Robert, Chris A.
Jakober, University of California, Davis; James J. Schauer,
University of Wisconsin, Madison; Michael P. Hannigan,
University of Colorado, Boulder.
Size-resolved samples of airborne particulate matter
collected at 3 sites in the San Joaquin Valley of California
were extracted with organic solvents and analyzed for
detailed organic compounds using GC-MS. The smallest
size fraction analyzed was 0.056 < Dp < 0.1 micro-meter
particle diameter which accounts for the majority of the
ultrafine particle mass (PM0.1). Source profiles for
ultrafine particles developed during pervious studies were
applied to the measurements at each sampling site to
calculate source contributions to organic and elemental
carbon concentrations. Ultrafine elemental carbon
concentrations ranged from 0.03 micro-gram m-3 during
the daytime to 0.18 micro-gram m-3 during the nighttime.
Gasoline fuel and motor oil accounted for the majority of
the ultrafine elemental carbon concentrations, with
relatively minor contributions from biomass combustion
and meat cooking. Ultrafine organic carbon
concentrations ranged from 0.2 micro-gram m-3 during
the daytime to 0.8 micro-gram m-3 during the nighttime.
Wood combustion and meat cooking were found to be the
two largest sources of ultrafine organic carbon, with
smaller contributions from diesel fuel, gasoline fuel, and
motor oil. Future inhalation exposure studies may wish to
target these sources as potential causes of adverse health
effects.
9A.5
Bayesian Approaches for Pollution Source Location
Identification and Apportionment. WILLIAM F.
CHRISTENSEN, Basil Williams, C. Shane Reese, Brigham
Young University.
We consider the integration of a deterministic dispersion
model and a statistical model for the purposes of
identifying point source locations. The approach utilizes
meteorological data, species abundance measurements,
and Toxic Release Inventory data. Posterior distributions
for major point source directions in the St. Louis,
Missouri area are presented and their value for pollution
source apportionment (PSA) is discussed. We also
present a Bayesian hierarchical model for PSA that allows
for the utilization of a priori information related to
hypothesized source profiles, seasonal variation, and
meteorology. The Bayesian PSA approach is compared
with standard approaches via simulation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
9A.6
Near-Road PM2.5 Mass Concentrations of Manganese, Iron,
Chromium and Lead: Mixed Model Analyses of
Contributing Factors. Timothy M. Barzyk, Alan Vette, Carvin
Stevens, BJ George, Carry Croghan, U.S. EPA; Jonathan
Thornburg, Charles Rodes, RTI International; Ronald Williams,
U.S. EPA.
11G.1
Fine, Ultrafine and Nanoparticle Trace Element
Compositions Near a Major Freeway with a High HeavyDuty Diesel Fraction. Leonidas Ntziachristos, Zhi Ning,
MICHAEL D. GELLER, Constantinos Sioutas*, University of
Southern California; Rebecca J. Sheesley, James J. Schauer,
University of Wisconsin, Madison.
The Detroit Exposure and Aerosol Research Study
(DEARS) was a three year air pollution study designed to
assess the influence of point and mobile sources on air
quality and human exposures. The DEARS included
measurements of PM2.5 mass concentrations of manganese
(Mn), iron (Fe), chromium (Cr) and lead (Pb) at outdoor
residential sites throughout Detroit, Michigan, and at an
ambient monitoring site (Allen Park). The first year of the
study provided 362 sample days at 48 households over
two 7-week sampling seasons (summer and winter).
Factors affecting outdoor residential levels of these metals
were assessed including concentrations measured at the
ambient site, proximity to major roadways (interstates and
arterials), and meteorology. Linear mixed effects models
were used to assess the statistical significance for each of
these contributing factors, separately and jointly, on the
outdoor residential concentrations of Mn, Fe, Cr and Pb.
Results of the mixed models indicate that ambient
concentrations of Mn, Fe and Pb were significantly
related to outdoor residential concentrations (p < 0.005)
and distance to roadway was significant for outdoor
residential concentrations of Mn and Fe. Results for wind
speed and direction were inconclusive in the Year 1
models, possibly because the residential sites were
downwind of a major roadway only during 11% of the
sample days. Measurements from the households
monitored during Year 2 of the DEARS will be added to
increase the statistical power of the models. The effect of
nearby industrial sources and other potential covariates
including mobile source related volatile organic
compounds will also be assessed with the models. The
additional data and revised models will clarify the impact
of mobile source emissions on particulate matter
components measured outdoors at residences.
Trace elements and metals in the ultrafine (<0.18 micrometers) and accumulation (0.18 - 2.5 micro-meters)
particulate matter (PM) modes were measured during the
winter season next to a busy Southern California freeway
with significant (~20%) diesel traffic. Both ambient and
concentrated size-segregated impactor samples were
taken in order to collect enough mass for chemical
analysis. Data at this location were compared to a site
located 1 mile downwind of the freeway, which was
reflective of urban background. The most abundant trace
elements in the accumulation mode detected by
inductively coupled plasma mass spectroscopy (ICPMS)
3
3
3
were S (138 ng/m ), Na (129 ng/m ) and Fe (89 ng/m )
3
3
while S (35 ng/m ) and Fe (35 ng/m ) were the most
abundant in the ultrafine mode. The concentrations of
several trace elements, including Mg, Al, and Zn, and in
particular Ca, Cu, and Pb, did not uniformly increase with
size within fine PM, an indication that various roadway
sources exist for these elements. Calculation of crustal
enrichment factors for the two sites indicates that the
freeway traffic contributed to enriched levels of ultrafine
Cu, Ba, P and Fe and possibly Ca. The results of this
study show that trace elements constitute a small fraction
of PM mass in the nanoparticle size range, but these can
and should be characterized due to their likely importance
to human health.
Although this work was reviewed by EPA and approved
for publication, it may not necessarily reflect official
Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11G.2
Particle Concentration And Characteristics Near A Major
Freeway With Heavy-Duty Diesel Traffic. Leonidas
Ntziachristos, Zhi Ning, MICHAEL D. GELLER and
Constantinos Sioutas*, University of Southern California.
This study presents the number, surface and volume
concentrations and size distribution of particles next to the
I-710 freeway during February through April 2006. I-710
has the highest ratio (up to 25%) of heavy-duty diesel
vehicles in the Los Angeles highway network. Particle
concentration measurements were accompanied by
measurements of black carbon, elemental and organic
carbon and gaseous species (CO, CO2). Using the
incremental increase of CO 2 over the background to
calculate the dilution ratio, this study makes it possible to
compare particle concentrations measured next to the
freeway to concentrations measured in roadway tunnels
and in vehicle exhaust. In addition to the effect of the
dilution ratio on the measured particle concentrations,
multivariate linear regressions showed that light and
heavy organic carbon concentrations are positively
correlated with the particle volume in the nucleation and
accumulation modes respectively. Solar radiation was
also positively correlated with the particle surface
concentration and the particle volume in the accumulation
(40-638 nm) mode, presumably as a result of secondary
particle formation. The methods developed in this study
may be used to decouple the effect of sampling position,
meteorology and fleet operation on particle concentrations
in the proximity of freeways, roadway tunnels and in
street canyons.
11G.3
Real-Time Measurement of Ambient Particle
Concentrations in Pune, India. MANISHA SINGH, TSI Inc.;
Rakesh Kumar, Vikram Shenvi, National Environmental
Engineering, Research Institute, P. Satyanarayana, Tesscorn
Systems India.
Pune is a rapidly growing city in Western India and is
located 100 miles south of Mumbai. Air quality in the city
like other urban areas is severely impacted by vehicular
emissions. Heavy-duty diesel vehicles as well as 2 and 3wheeler vehicles (which are mostly 2-stroke engines) are
major contributors to particulate matter (PM) pollution.
There are concerns regarding the health consequences of
ambient air pollution in Pune, but routine air monitoring
is limited and does not characterize exposure conditions
that people experience on a day- to- day basis, especially
while commuting. Most of the public transport buses
(with the exception of the new fleet) ply with open
windows and doors, resulting in high infiltration of
particulates from outside. Passengers of these buses as
well as semi-enclosed rickshaws and 2-wheelers have
high exposure potential. In this paper, data from real-time
measurements of particulate concentrations inside
passenger compartment of a plying bus as well as outside
on road concentrations in the near-vicinity of the bus will
be presented. Measurements were conducted using three
real-time aerosol monitors viz. an Electrical Aerosol
Detector (TSI Model 3070A, EAD), an Engine Exhaust
TM
Particle Sizer (TSI Model 3090, EEPS) and a
TM
DUSTTRAK Aerosol Monitor (TSI Model 8520). The
EAD measured particles in the size range of 10 nm to
1000 nm. The EEPS measured the number based size
distribution of particles in the size range of 5.6 to 560 nm
(total 32 size channels) while the DUSTTRAK measured
PM10 (mass concentration of particles < 10 micrometer in
aerodynamic diameter) in real-time.This data will provide
insights into the impact of vehicular emissions on ambient
air quality and on-road exposure levels of PM. PM
measurements data from another site in the city located on
the hillside away from the traffic sources as background
will be presented for comparison.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11G.4
Roadside measurements of size-segregated particulate
organic compounds near gasoline and diesel-dominated
freeways in Los Angeles, CA. HARISH C. PHULERIA, Philip
M. Fine, Constantinos Sioutas, University of Southern
California; Rebecca Sheesley, James J. Schauer, University of
Wisconsin-Madison.
Individual organic compounds such as hopanes and
steranes (originating in lube oil) and selected PAH
(generated via combustion) found in particulate emissions
from vehicles have proven useful in source apportionment
of ambient particulate matter. Detailed information on the
size-segregated (ultrafine and accumulation mode)
chemical characteristics of organic particulate matter
during the winter season originating from a pure gasoline
traffic freeway (CA-110), and a mixed fleet freeway with
the highest fraction of heavy-duty diesel vehicles in the
state of California (I-710) is reported in this study.
Hopanes and steranes as well as high molecular weight
PAHs such as benzo(ghi)perylene (BgP) and coronene
levels are found comparable near these freeways, while
EC and lighter molecular weight PAH are found much
elevated near I-710 compared to CA-110.
The roadway organic speciation data presented here is
compared with the emission factors measured in the
Caldecott tunnel, Berkeley CA (Phuleria et al., 2006) for
light duty vehicles (LDVs) and heavy-duty vehicles
(HDVs). Very good agreement is observed between CA
-110 measurements and LDV emission factors (EFs) as
well as I-710 measurements and corresponding
reconstructed EFs from Caldecott tunnel for hopanes and
steranes as well as heavier PAHs such as BgP and
coronene. Our results therefore suggest that the emission
factors for hopanes and steranes obtained in tunnel
environments, where emissions are averaged over a large
vehicle-fleet, enable reliable source apportionment of
ambient PM, given the overall agreement between the
roadway vs tunnel concentrations of these species.
11G.5
Evaluate PM emission impacts air quality concentrations
and population exposure to traffic-generated pollutants in
the near road environment. FU-LIN CHEN, Ronald
Williams, Fred Dimmick, Richard Baldauf, U.S. Environmental
Protection Agency.
A study was conducted in Raleigh, North Carolina to
assess the impacts of traffic emissions on air quality near
a heavily-traveled highway. PM10 and PM2.5 MiniVol
samplers and sequential air samplers (FRM) were used to
measure particle mass concentrations in various locations
near the I-440 highway in Raleigh, North Carolina.
MiniVol samplers (one PM10 and one PM2.5 ) were
positioned in a protective cage 10, 50, 100, and 275
meters away from the highway edge. Another set of
MiniVol samplers were placed behind a 6 meter tall noise
barrier and 10 m away from the highway. The FRM
monitors were positioned by a cage 10 and 275 meters
from the highway. Measurements were collected on a
near daily basis from July 27 to August 9, 2006. PM10-2.5
values were estimated using the differential between
independent PM10 and PM2.5 collocated MiniVol
measurements. Results indicated that PM10 and PM10-2.5
mass concentration decrease substantially with increasing
distance. PM2.5 mass concentrations were not affected by
the distance being evaluated. The noise barrier reduced
4% and 12% of the PM10 and PM10-2.5 mass concentration,
respectively. The noise barrier did not reduce PM2.5
concentrations which increased by about 2%. The
comparison of the MiniVol and FRM resultd were in
good agreement. The regression slope for 10 and 275
meters location were 0.9 and 0.92, respectively. The
intercept for 10 and 275 meters location were 3.32 and
3.61, respectively. Average correlation coefficients of
r=0.98 at 10 meter and r=0.99 at 275 meter were obtained.
Data in this study resulted in a range PM2.5 /PM10 values
ranging from 75 - 80 % by location. This study showed
additional PM size fraction field studies need to be
performed to more fully understand the exposure
assessment of adverse health effects for populations living
near major roads.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11G.6
Study of particulate mater at Mitrovica roadside in rural
and urban area. AFRIM SYLA, Agron Veliu, Kadri Berisha,
Syle Tahirsylaj, Leonora Nuli Universitet of Prishtina Research Aerosol Institute Prishtina, Kosova.
11G.7
Study Of Particulate Mater At Mtrovica Roadside In Rural
And Urban Area Of Northern Kosova. AFRIM SYLA1,2,
Emin Karakashi1, Agron Veliu1, Kadri Berisha 1, Leonora
Nuli2, Mexhit Musa 2 1Universitet of Prishtina, Mitrovic
This research present aerosol measurements using the
field study data from April 2004 to December2006 to
evaluate the particulate air pollution in selected area in
Mitrovica. The study employed microenvironment
monitoring technique to acess the exposure of suspended
particulate and airborne lead at heavly trafficked urban
roadside. A total of 25 roadside side in 8 districts were
selected which covered the most urbanized and densely
populated area. It was found that pedestrians exposed to
24 hour average and airborne lead ranged from 29,35 to
425,32 mikrogram/cubic meter and 0,0954 to 0,856
microgram/cubic meter. Most field study data was
significantly higher than the neargy fixed station data.
This research present aerosol measurements using the
field study data from April2004 to December 2006 to
evaluate the particulate air pollution in selected area in
Mitrovica . This part of northern Kosova constitutes one
of most important mining field in Europe. Consequently
the industrial activities were mostly connected to the
exploitation of the base metal ores of the Trepca district
as well as the smelter of Zvecan and the Trepca battery
factory. The town of Zvecan and Mitrovica have been
exposed for many years to the poisons emissions of the
Zvecan stacks. Indeed, according to a crude estimate,
about a ton of lead vapours were daily related to the
atmosphere through the main stack Because of the
imposing emissions, the KFOR Headquarters decided to
stop in August 2000 the roasting plant activity at Zvecan.
However, atmospheric transport of the emissions from the
Zvecan plants provoked a serious heavy metal pollution
of a vast area embarcing the Iber and Sitnica valleys:
soils, air within a radius of at least 10 kilometers from the
stacks are heavily polluted mostly for highly toxic
elements such as lead and cadmium.
The study employed microenvironment monitoring
technique to access the exposure of suspended particulate
and airborne lead at heavily trafficked urban roadside. A
total of 25 roadside side in 8 districts were selected which
covered the most urbanized and densely populated area. It
was found that pedestrians exposed to 24 hour average
and airborne lead ranged from 29.35 to 425.32
microgram/cubic meter and 0.0954 to 0.856 microgram/
cubic meter. Most field study data was significantly
higher than the nearby fixed station data.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11G.8
Experimental and modeling study of particle deposition near
roads. JOHN VERANTH, Scott Speckart, Eric Pardyjak,
University of Utah.
11G.9
Effects of a Sound Barriers and Vegetation on the
Dispersion of Ultrafine Aerosol from Highways. ANDREY
KHLYSTOV, Duke University.
Near source redeposition has been proposed as an
explaination of why emission inventory-based methods
overpredict the contribution of vehicle-generated dust to
ambient particulate. Two published field studies that
tested this hypothesis by measurement of dust flux down
wind of a test road reached contrasting conclusions
regarding the fraction of dust that is redeposited within
100 m of a road. Subsequent measurements of particle
deposition on simulated vegetation and computational
modeling of dust transport provide new data that
reconciles the field study results. Surface roughness and
atmospheric stability are important variables in near
source dust transport. A computationally efficient model
for near source dust transport that includes these effects
has been developed and simulation results show good
agreement with the experimental dust cloud concentration
and time data. Direct measurements of particle deposition
on flat and artificial vegetation surfaces placed near roads
provides new estimates on the particle removal in the near
source impact zone where the dust cloud is moving
approximately horizontally and is comparable in height to
the surface roughness elements such as buildings and
vegetation.
Understanding the dispersion of pollutants from traffic
sources is important both for urban planning and for air
quality assessments. This study examines the effects of a
roadside sound barrier on the dispersion of ultrafine
particles from a high-traffic highway in an urban area of
Raleigh, NC. The effect of the sound barrier was
investigated using a mobile unit equipped with a Global
Positioning System (GPS) and two Differential Mobility
Analyzer / Condensation Particle Counter (DMA/CPC)
combinations. The DMA/CPCs provided measurements
of \full\ size distribution (size range 10 - 270 nm) of
ultrafine aerosol and / or measurements of aerosol
concentration at two fixed sizes: 20 nm and 75 nm. The
\full\ size distributions measurements were made with a
frequency of 20 seconds, while single size measurements
were made with the frequency of 10 Hz. The
measurements were performed continuously as the van
drove over a fixed pattern, scanning the area behind the
barrier and in an open field adjacent to it. Highly resolved
gradients of aerosol concentration as a function of the
distance from the highway collected in this study are used
to assess the effects of the noise barrier on ultrafine
aerosol dispersion.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11G.10
Characterization of Seasonal Changes in Aerosol
Characteristics in Toronto, Canada through the SPORT
campaign. GREG J. EVANS, Jonathan P.D. Abbatt, CheolHeon Jeong, Xiaohong Yao, Krystal Godri, Ryan D
Seasonal variations in the characteristics of aerosols in
downtown, Toronto, Canada, were investigated through
the Seasonal Particle Observation in the Region of
Toronto (SPORT) intensive air sampling campaign
conducted during the summer of 2006 and winter of 2007.
A suite of high time resolution instrumentation, with 1 sec
to 15 min sampling times, was used to investigate
fluctuations in the concentration, composition and size
distribution of the fine and ultrafine particulate (UPF)
matter. This instrumentation included a TSI Aerosol Time
of Flight Mass Spectrometer, Aerodyne Aerosol Mass
Spectrometer, Dionex Gas-Particle Ion Chromatograph,
and TSI Fast Mobility Particle Sizer. Fluctuations in the
origin of the incoming air masses allowed pollutants from
regional and local sources to be distinguished.
Measurement of fine PM composition revealed distinct
day-to-day and seasonal trends in the fluctuation of the
particulate component and precursor gas concentrations.
NOAA HYSPLIT back trajectories indicated different
geographic origins were associated with these changes in
the composition of the regional fine PM. Measurement of
the UFP concentration indicated multiple spikes of less
than on minute in duration caused by local vehicles, on
top of a regional baseline concentration. The
presentation will contrast the aerosol characteristics
during the winter and summertime, and highlight
advantages offered through high time resolution
measurement methods.
11H.1
Linked Dependencies of PM2.5 and Ozone Responses to
Emissions Controls, Now and in the Future. KUO-JEN
LIAO, Efthimios Tagaris, Kasemsan Manomaiphiboon,
Armistead G. Russell, Georgia Institute of Technology, JungHun Woo, Praveen Amar, Shan He, Northeast States for
Coordinated Air Use Management.
Influence of precursors (e.g., NOx, SO2, NH3 and VOCs)
on regional PM2.5 and ozone formation and the
effectiveness of currently planned control strategies are
investigated over the continental U.S. both historically
(2001) and in the future under the impact of climate.
MM5, SMOKE and CMAQ with DDM-3D are used to
calculate ozone and PM2.5 sensitivities to precursor
emissions. Responses to controls are found to be only
slightly sensitive to climate changes alone. In many cases,
absolute sensitivities (e.g., ppbV/ton) to NOx and SO2
controls are predicted to be greater in the future due to
both the lower emissions due emission controls as well as
climate, suggesting that current control strategies based
on reducing such emissions will continue to be effective
for decreasing ground-level ozone and PM2.5
concentrations. SO2 emissions are predicted to be most
beneficial for decreasing summertime PM2.5 levels while
controls of NOx emission are more effective in the winter.
Effectiveness of controls of SO2, VOC and NOx
emissions on ozone and PM2.5 are investigated for five
cities (ATL, CHI, HUS, NY and LA) in the continental U.
S. Controls of anthropogenic SO2 and VOC emissions are
found to be beneficial to decrease both PM2.5 and ozone
levels in the urban areas. Controls on anthropogenic NOx
emissions are simulated to decrease PM2.5 levels but lead
to higher ozone levels in large cities.
Of particular interest is how controls interact between
species. Here, sensitivities of PM2.5 and its secondary
components (e.g., sulfate, nitrate, ammonium, and organic
components), as well as ozone, are correlated over one
year. As expected, nitrate sensitivities to SO2 emissions
are negative, though much less so than sulfate, suggesting
that SO2 controls will reduce PM2.5 on an annual basis.
There is a small, negative correlation between sulfate
formation and NOx emissions, but, again, nitrate levels go
down more than sulfate goes up. Surprisingly, there is a
small, negative sensitivity of sulfate formation to VOC
emissions in VOC-rich areas due to radical scavenging.
Using one year simulations identifies the seasonal
variations in such sensitivities, e.g., the enhanced nitrate
sensitivities in the winter and SOA and sulfate
sensitivities in the summer.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11H.2
Integrated PM10 Emission Assessment and Modeling in
Mediterranean Regions. Cristina Faricelli, Maria Chiara
Metallo, ATTILIO A POLI, Francesca Raffaele, Alessandra
Scifo, Environmental System Analysis S.r.l.
11H.3
A DSS Application to Perform Operational PM10 Forecast.
MARIA CHIARA METALLO, Cristina Faricelli, Attilio A.
Poli, Pierluca Di Giovandomenico, Francesca Raffele,
Alessandra Scifo, Environmental System Analysis S.r.l.
The assessment of the anthropogenic contribution to
PM10 plays a key role in the frame of Air Quality
evaluation. In order to accurately discern this contribution
it is necessary to assess the quantity of particulate matter
coming from natural sources as well. In this work an
integrated system for the assessment of PM10 emission is
presented. It utilizes different approaches according to
different aerosol sources, the contributions from the
various sources are then added together.
Controlling PM10 concentrations in highly inhabited
areas constitutes a critical issue both for citizens and local
authorities. Air Quality Forecast can be an important tool
in terms of knowledge and communication. At this
purpose the predictions have to be accurate, with high
spatial and temporal resolution and provided on a daily
basis.
To achieve this a DSS has been implemented in different
Spanish regions, integrating in a GIS platform emission
and dispersion models and running in operational mode to
provide 24 and 48 hour forecast. The system is running
using three nested domains with increasingly grid
resolution (54 km, 6 km, 2 km), the first one including
Northern Africa to take into account also Saharan dust
intrusions. The emission inventory is performed
according to CORINAIR methodology, the atmospheric
circulation is predicted by the Fifth-Generation NCAR/
Penn State Mesoscale Model (MM5) and finally the
transport and deposition is driven by the California
Photochemical Grid Model (CALGRID).
The DSS daily provides hourly concentrations for 24/48
hours, an Air Quality Index to easily inform the
population, the comparison of the last forecast with the
concentration measured by the local monitoring network
and the forecast performance index calculated accordingly
to the EU Directive for model evaluation.
Notwithstanding the many uncertainties related with
meteorology, emissions and atmospheric chemistry,
results of forecast show good confidence with measured
data and EU evaluation index is fully satisfied on short
and long term basis.
An emission inventory database, performed according to
CORINAIR (the official European methodology),
contains the contribution of the anthropogenic sources
(road transport, non road transport, industry, residential
heating, etc.) present in the area of interest. Information
about spatial and temporal distribution of each kind of
source is also carefully evaluated. A huge quantity of data
related to parameters of disparate nature (socioeconomical, geographical, chemical, technological,
demographic) are elaborated and organised to develop the
emission inventory.
The contribution of mineral aerosol coming from bare
soils and agricultural field, in the target area, is taken into
account associating different emission coefficients to
selected classes of soils using the Corine Land Cover.
Given its relevance in the Mediterranean basin also the
Saharan dust flux is calculated; this is done employing a
dedicated algorithm, meteorology-dependent, driven by
the Penn State Mesoscale Model (MM5). The release of
aerosol is considered when the friction velocity exceeds a
threshold velocity function of the particle diameter, the
soil moisture content and the soil roughness length. The
quantity of dust released in the atmosphere depends on
the soil texture, in particular on the fraction of clay.
The above system allows to draw a comprehensive picture
of the different contributions. The emissions calculated
with the above scheme provide the input for an
Operational air quality forecasting system. The system is
currently operative in several Spanish regions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11H.4
Improvements in Modeling Urban PM Concentrations using
the St. Louis Super Site Data. RALPH MORRIS, Bonyoung
Koo, Jeremiah Johnson, Greg Yarwood, ENVIRON
International Corporation; Jay Turner, Jennifer Garlock,
Washington University in St. Louis; Calvin Ku, Wendy Vit,
Adel Alsharafi, Missouri Department of Natural Resources.
Currently several States are developing emission control
plans that demonstrate compliance with the PM2.5
standard to be submitted to EPA by April 2008. The
demonstration of attainment is performed using threedimensional photochemical grid models (PGMs). The
performance of such models varies from quite good for
sulfate (SO4) and elemental carbon (EC) to poor for Soil
to highly variable for Organic Matter Carbon (OMC) and
nitrate (NO3). Because of the variability in emissions and
meteorology and the volatility of some PM species, they
exhibit large diurnal variability. However, most PM
measurements are obtained as 24-hour averages that limit
the ability to diagnose why the models deviate from the
measured values and when they do agree whether the
modeled agreement is for the right reason. This paper
presents the results of comparison of two models (CMAQ
and CAMx) with the high time resolution PM species
measurements from the St. Louis Super Site (SS) and how
such comparisons have improved model representation
and model performance for many PM species. The
comparisons of predicted and observed hourly sulfate
(SO4) and Organic Mass Carbon (OMC) are presented
and how they are used to improve the model is discussed.
The use of data analysis techniques to decompose the
OMC measurements into primary and secondary
components and components due to anthropogenic and
biogenic emissions and how these data are used in the
evaluation of the CMAQ and CAMx models are
discussed. The detailed evaluation of the two models
using the St. Louis SS data are presented along with areas
where more research is needed to improve model
performance.
11H.5
Numerical CFD Modelling of the Formation of an Aerosol
Distribution close to a Car Traffic Linked Source. BASTIEN
ALBRIET, Karine Sartelet, CEREA.
Nowadays, special attention is paid to the aerosol number
distribution as it might be more related to health impact
than the mass distribution. In particular, car traffic leads
to emission of large amount of nanoparticules. Those
nanoparticles are often observed to be bimodal: primary
emission of soot particles and secondary nucleated
nanoparticles. The modal aerosol model MAM has been
coupled to the CFD code Mercure Saturne, adapted to
atmospheric conditions, in order to simulate the formation
of the traffic influenced aerosol distributions. In the
simulations, two dilution phases are distinguished.
The first phase occurs in the first 10 meters in the plume
of a tailpipe. It lasts only 1 or 2 s. Emissions are mainly
composed of soot particles, sulphuric acid and semivolatil organics. Those organics are speciated following
measurements in Schauer et al. [1999]. During the
cooling of exhaust gases in the atmosphere, fast
nucleation and a rapid growth of nanoparticles by
condensation are observed. Brownian coagulation is too
slow to have a real impact here.
The second phase corresponds to a domain from the road
border to a few hundred meters downwind. Results
obtained at 10 meters in the tailpipe plume are taken as
input conditions. During this phase, nucleation is almost
inactive. Nanoparticules continue to grow through
condensation process. Brownian coagulation can have a
more important impact on the aerosol distributions.
The model seems to be able to reproduce qualitatively
well what is observed for the formation of aerosol
distributions close to car traffic linked sources for both
dilution phases. Large uncertainties subsist for emission
data. The sensitivity analysis performed allow to identify
the sensitive parameters and the important processes of
the modelling for each dilution phase.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11H.6
Understanding Source Impacts on Particulate Matter
Concentrations in the Eastern United States. KRISTINA
WAGSTROM, Spyros Pandis, Carnegie Mellon University.
An important aspect of understanding regional particulate
matter concentrations is gaining a better understanding of
the contributions of different sources to these particulate
matter concentrations. We have used PMCAMx, a
regional three-dimensional chemical transport model, to
study impacts of pollutant sources and transport on
particulate matter concentrations in the Eastern United
States using PSAT (Particulate Matter Source
Apportionment Technology). PSAT is a computationally
efficient particulate matter apportionment algorithm that
is able to track source contributions to both primary and
secondary particulate matter in regional Eulerian models
such as PMCAMx. Comparisons of apportionments
predicted by PSAT with those predicted by more
computationally expensive and benchmark methods,
showed agreement within a few percent for secondary
organic aerosol species (Wagstrom et al., 2007).
The contributions of different source types (fuel
combustion for electricity generation, fuel combustion for
industrial processes, non-combustion related industrial
emissions, transportation and biogenic sources) to
primary and secondary particulate matter concentrations
are quantified. The seasonal dependence of these
contributions is investigated. In this implementation in
PMCAMx, PSAT also allows the study of size-resolved
source contributions. These contributions to different
parts of the aerosol size distribution (e.g. ultrafine
particulate matter) are discussed.
Wagstrom, K. M., Pandis, S. N., Yarwood, G., Wilson, G.
M., Morris, R.E . 2007. \Development and Application
of a Computationally Efficient Particulate Matter
Apportionment Algorithm in a Three-Dimensional
Chemical Transport Model\. Manuscript in Preparation.
11H.7
Simulating Present-Day and Future Regional Qir Quality As
Climate Changes: Model Evaluation. JOHN DAWSON,
Pavan Racherla, Barry Lynn, Peter Adams, Spyros Pandis,
Carnegie Mellon University.
The Global-Regional Climate-Air Pollution modeling
System (GRE-CAPS) has been developed, linking a
general circulation model/chemical transport model
(GCM/CTM), a regional meteorological model, and a
regional chemical transport model (CTM). This modeling
system has enabled the examination of the effects of
changes in climate, intercontinental transport, and global
and regional emissions on regional and urban air quality.
The GRE-CAPS system consists of the GISS II' GCM/
CTM, the MM5 regional meteorological model, and the
PMCAMx regional CTM. Global-scale meteorology and
pollutant concentration fields are generated by the GCM/
CTM. Meteorology is downscaled to the regional level
using MM5. Intercontinental transport is simulated by
using the GCM/CTM-predicted concentrations around the
edge of the regional CTM domain as chemical boundary
conditions in the regional CTM.
The modeling system is evaluated for the present day,
with model predictions compared to measured ozone and
speciated PM2.5 measurements. Model predictions for five
present-day Januaries and Julys are compared to
measurements from the STN and IMPROVE databases
from 2001-2005. Concentrations at 22 sites spread
throughout the Eastern US modeling domain were used
for comparisons. GRE-CAPS performed rather well in
capturing present-day pollutant concentrations. Model
biases and errors were similar to those for traditional
model evaluation of historical air pollution episodes.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11H.8
Concentration and Composition of Fine Particulate Matter
and Resulting Human Health Effects from Using Installed
Backup Generators for Meeting Peak Electricity Demand.
ELISABETH A. GILMORE, Peter J. Adams, Lester B. Lave,
Carnegie Mellon University.
Existing generators installed for backup power during
blackouts could be operated at their marginal costs during
periods of peak electricity demand, increasing grid
reliability and supporting electricity delivery. Many
generators, however, are diesel internal combustion
engines (ICE) which have non-negligible air emissions
that could damage air quality and human health. Of
specific concern are the fine particulate matter (PM)
emissions. In this work, first, we model the concentrations
and composition of fine PM from the operation of diesel
ICEs, natural gas ICEs, and natural gas microturbines,
using both a 3-D chemical transport model (PMCAMx)
and dispersion plumes. The generators are run for a
bounding scenario producing 36,000 MWh of electricity
over 3 summer days (July 23 to 25, 2001). Second, we
transform the concentrations into their equivalent health
endpoints using concentration-response (CR) functions
and generate a social cost by multiplying the resulting
morbidity and mortality by the willingness to pay to avert
ill-health. For several urban centers in the Eastern US,
small but noticeable enhancement (up to 5 micrograms
per cubic meter) in fine PM were observed for
uncontrolled diesel engines. While secondary fine PM is
formed, the PM mass is dominated by the elemental and
organic carbon from direct emissions. Since we are
interested in controlling direct emissions, a diesel
particulate filter (DPF) is a suitable control technology.
We find that an controlled diesel ICE has a full (private
and social) cost that exceeds all other options (over $2/
kWh), including the new peaking plant (approx. $0.60/
kWh). Retrofitting the diesel ICE with a DPF reduces the
social cost from PM mortality to less than $0.10/kWh; the
full cost of this option is comparable to the natural gas
generators. On a full cost basis, controlled diesel ICEs
backup generators are a cost-effective method of meeting
peak demand.
11H.9
Regional Process Analysis of Wintertime Particulate Matter
Formation in Central California. QI YING, California Air
Resources Board.
California's central Valley experiences some of the worst
wintertime particulate air quality pollution in the nation.
Although conceptual models and air quality simulations
have been used in the past to explain the formation of
elevated air borne particulate matter (PM) in the Valley,
the significance of each major process that leads to the
high PM concentrations is unknown. The CIT/UCD air
quality model was modified to include a process analysis
scheme for gaseous and PM pollutants to reveal the
relative importance of chemical production, horizontal
and vertical advection and diffusion, direct emission and
deposition on the predicted pollutant concentrations in the
entire model domain.
In this study, the process analysis tool is applied to study
the concentrations of EC, OC, and ammonium nitrate and
its precursor gases (O3, HNO 3 and N2O5) during the
wintertime California Regional Particulate Air Quality
Study (CRPAQS). This is the first time a PM formation
process analysis is applied to study the wintertime PM
pollution problem in California. The diurnal variation of
O3 is mainly driven by the downward mixing of
background O3 from the upper atmosphere due to vertical
diffusion, and the net chemical destruction of O3 by NO
due to low photochemical reaction rates in the winter.
The formation of HNO3 and N2O5 are found to be less
important in the source regions of NOx than in the remote
receptor regions due to higher O3 concentrations.
Regional process rate analysis is used to determine the
significance of inter-region transport of nitrate precursors.
The significance of N2O5 heterogeneous reactions is
determined for different regions in the model. Local
emissions and vertical diffusion are the major processes
that determine the surface elemental and organic carbon
concentrations in the source regions. Dry deposition of
particulate matter in the PM2.5 size range is found to be
less significant.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11H.10
Predicting Future Air Quality in California's San Joaquin
Valley. MARK HIXSON, Michael J. Kleeman, University of
California-Davis.
The San Joaquin Valley in California presently exceeds
both national and state air quality standards for ozone and
particulate matter of less than 10 micro-meters and 2.5
micro-meters (PM 10 and PM 2.5). There is work
currently underway to correct this problem, but the impact
of future changes to land use and population on the ability
to achieve acceptable air quality in the region is currently
unknown. The San Joaquin Valley will have to cope with
exploding population growth over the next 25 years; it is
currently the fastest growing region in California.
Changing demographics, industry, transportation and
agriculture could have profound impact on the land use in
the region. The influence of climate change on air quality
via direct and indirect effects must also be considered.
In the present study, air quality emission estimates for the
San Joaquin Valley are generated for the year 2030.
Limiting cases are examined ranging from conservative
growth to rapid expansion in order to span the full range
of possible future outcomes. These estimates attempt to
account for the changes in population, transportation,
industry, electrical power generation, and agriculture in
different future scenarios. The impacts of these emissions
on future air quality are evaluated with current
meteorology data from the year 2000 using a regional air
quality model. Future air quality predictions are
compared to present-day results to identify areas of future
concern for the region's air quality attainment.
11H.11
Reconciliation of an emission based model and a source
based model via source apportionment of PM2.5 - Part 2.
Trace metals. Jaemeen Baek, Sangil Lee, Bo Yan, Mei Zheng,
ARMISTEAD G. RUSSELL, Georgia Institute of Technology.
In order to improve accuracy of an emission based
model and receptor models, reconciliation of measured
and simulated metal concentrations is performed using
CMAQ and the Southeastern Aerosol Research and
Characterization Study (SEARCH) monitoring data
measured in July 2001 and January 2002 data. One of
main objectives of this study is to investigate the
quantitative correction factors of PM2.5 emissions and to
sources of bias. CMAQ is extended to follow source
tracers for metals and organic carbon compounds
(molecular markers) and is combined with source profiles
in four receptor models to obtain simulated individual
species metal concentrations. For comparison with
receptor modeling, one PMF and three CMB modeling
studies with different fitting species are used. Al and Si
are used as tracers for dust sources, Zn and EC for mobile
sources, K for wood burning, etc.
Preliminary analysis of trace metals suggests that
emission inventory bias factors are needed for urban and
rural sites, and for wintertime and summertime. For
example, correction factors of wood burning at Oak
Grove, one of rural sites, are around two for both seasons,
while correction factors at Jefferson St., which is an urban
site, in winter are 0.15 and close to one in summer.
Different correction factors at different land-use types
suggest that not only quantity of emissions but also spatial
surrogate of emissions need to be modified. For other
sources, correction factors of emissions of mobile sources
ranged from three to eight in both seasons at most of sites
and around 0.5 for soil/road dust.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11I.1
Assessing Traffic Related Exposure to Ultrafine and Fine
Particulate Matter, Particle-bound PAHs, CO and CO2
Across Communities in the Greater Toronto Area. KELLY
SABALIAUSKAS, Greg J. Evans, Elki Tsang, Amy Peers,
University of Toronto; Monica Campbell, Toronto Public
Health; Dave Stieb, Amanda Wheeler,Health Canada; Jeff
Brook, Environment Canada.
Proximity to traffic and time spent in transportationrelated microenvironments has been of recent health
concern; however, monitoring in Canadian cities has so
far been limited. A personal exposure monitoring study
collected measurements of ultrafine (UFP) and fine
(PM2.5) particulate matter, CO, CO2 and particle bound
polycyclic hydrocarbons (PAHs) on roadways and in
communities throughout the Greater Toronto Area (GTA)
during the summer of 2006. Simultaneous in-vehicle and
roadway measurements were conducted while driving on
highways throughout the GTA. A video camera was used
to capture the on road traffic conditions and to assist in
peak identification. With the air conditioning on and air
intake set to re-circulate, the in-vehicle concentrations of
UFP, PM2.5 and PAHs were 6 times lower than those
measured simultaneously on the highway. In contrast, the
concentration of CO and CO2 increased by a factor of 5
inside the vehicle.
Traffic-related pollutant spatial gradients were
investigated across communities by collecting
measurements in parks. A total of 19 parks were visited
on two separate occasions that ranged in distances from
10-70 km from the downtown core. In order to compare
measurements between days, the ratio of measurements
taken at a centralized site in downtown Toronto was
compared to the field measurement in the park. As the
distance from the downtown core increased, the ratios
between the field measurements and centralized
measurement decreased for UFP and PM2.5. In addition,
the UFP and PM2.5 concentrations measured in the parks
were highly correlated to wind direction and proximity to
high traffic roads within the studied communities. A
correlation between distance and CO and CO2
concentrations was not observed.
11I.2
Emissions from an ocean going, crude oil vessel. HARSHIT
AGRAWAL, William W. Welch, Abhilash Nigam, J. Wayne
Miller, David R Cocker III, University of California Riverside,
CE-CERT.
Ocean-going vessels are significant sources of
anthropogenic emissions yet emission data from these
sources remains scarce. Tankers are one of the most
important ocean-going vessels and the goal of this
research was to measure the full slate of criteria and
greenhouse gas emissions from a crude oil tanker for ISO
cycle and also for actual real-time operation of the vessel.
Measurements were made following both the standard
certification cycles normally used for inventories and
during the actual activity of moving crude from the
VLCC to the refinery. Methods for sampling and analyses
of the gases conformed to the requirements of ISO 8178
-1. PM measurement was done by modified ISO 8178-1
cycle. The emissions of particulate matter (PM), criteria
gases (NOx, SO2 and CO) and carbon dioxide are
reported. Additional speciation of gas-phase
hydrocarbons and particulate matter (including elemental
and organic carbon, sulfate, and metals) will be presented
for the main and auxiliary diesel engines and for the
boiler. Real-time measurement of gases and PM
emissions was conducted for the actual operating
activities of the vessel, for example maneuvering in and
out the port, voluntary speed reduction (VSR), and the
lightering of oil.
The main engine and boiler on this vessel operated on
heavy fuel oil (HFO) while the auxiliary engine operated
on marine gas oil (MGO). PM from main engine was
comprised of 70 to 80% sulfate, 15% organic carbon (OC)
and <5% elemental carbon (EC). PM emissions per
kilowatt-hour from the auxiliary engine were about 5% of
the PM emissions from the Main engines. Complete
chemical characterization of emissions for different
engines will be presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11I.3
The Effects of Meteorological Conditions upon Infiltration
of Outdoor Particles into Residential Building with ShelterIn Place. INTAEK HAHN, National Research Council Senior
Research Associate, US EPA, Russell W. Wiener, National
Homeland Security Research Center, US EPA.
A field study to investigate and understand the
mechanisms involved in infiltration processes of outdoor
particles (of 0.02 micrometers to 1 micrometer) into a
residential building has been conducted. Using the
simultaneous and continuous real-time total number
concentration time series data measured at multiple
outdoor and indoor locations, the infiltration flux rates are
calculated for various indoor sites by the time-series
analysis method of cross-correlation. In order to examine
how the meteorological factors may affect the infiltration
flux rates, characterization of the infiltration rates at a
given location and their relationships with a number of
meteorological variables has been performed. In
particular, parameterization of the infiltration rate as a
function of wind speed is achieved. The relationships
between the infiltration rates and other meteorological
parameters such as the outdoor/indoor temperature and
the relative humidity are also investigated. The effects of
the wind speed upon the indoor/outdoor particle
concentration ratios are also presented by modeling the
association between the infiltration velocity and the
concentration ratio. In addition, the effectiveness of a
protective shelter-in place inside the building has been
analyzed by calculating the infiltration times into the
shelter-in place and establishing their relationships with
the same meteorological variables. Some of the major
results are:
(1) The elevated wind speed (u) results in the increase in
infiltration flux rate (Q) in accordance with a power law
a
function; i.e. Q = f (u ) with the power-law coefficient, a,
ranging from 1 to 2.
(2) The above empirical relationships suggest that the
various walls of the building exhibit the characteristics of
different infiltration processes or mechanisms.
(3) The most dominant meteorological parameter that
affects the infiltration flux rate is the wind speed, with the
wind angle, the temperature, and the relative humidity
showing weak or no apparent influence.
11I.4
Personal Exposure to Trace Organics in Fine Particulate
Matter. GREGORY BRINKMAN, Michael P Hannigan, Jana
B Milford
Studies have shown that personal exposure to fine
particulate matter (PM2.5) can differ significantly from
ambient PM2.5. The research described in this
presentation includes personal PM2.5 exposure samples
from seven different individuals for a total of 64 samples.
Subjects wore cyclone filter samplers loaded with quartz
and PTFE filters in the breathing zone for 24-hour
periods. Several of the samples were seeded with extra
time in source-specific microenvironments such as
restaurants and bars with cigarette smoking. Activity logs
were kept by the subjects to keep track of locations and
activities that provide information regarding potential
sources of PM2.5.
The quartz filters were used to quantify bulk elemental
and organic carbon using thermal optical transmittance,
and trace organic species using chemical extraction and
GC-MS. The PTFE filters were used to quantify total
PM2.5 mass and elements using ICP-MS. Results are
presented for the bulk carbon, mass, and trace organic
analysis. Trace organic speciation can be helpful for
determining the sources of PM2.5 due to the presence of
marker species that are emitted predominantly by a single
type of source. Correlations between the subjects'
activities and concentrations of particular species are
presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11I.5
Seasonal variation of ultrafine particle events in ambient
atmosphere at Gwangju, Korea. JIYEON PARK, Jae-Seok
Kim, Jihyun Kwak, Youngju Heo, Gangnam Cho, Kihong Park,
Gwangju Institute of Science and Technology, Gwangju, Korea.
Fine and ultrafine particles in the ambient atmosphere are
of interest because of their effect on the earth's radiation
budget, visibility impairment, and human health. We are
particularly interested in ultrafine particles (<~100 nm)
having higher reactivity and toxicity due to their enhanced
surface area-to-volume ratio. These particles are emitted
directly from various sources or formed in the ambient
atmosphere by gas-to-particle conversion process. In this
study, continuous measurements of particle size
distributions from 3 to 80 nm by the Nano SMPS, and
from 15 to 600 nm by the Regular SMPS have been
conducted in four seasons at Gwangju, Korea.
Simultaneous measurements of nitrogen oxides, carbon
monoxide, ozone, and meteorological data were made.
Furthermore, elemental and morphological analyses of
ultrafine particles were carried out off-line on distinct
ultrafine event days to better understand possible sources
of ultrafine particles and their formation and growth. The
sampling site is located ~1.8 km from Hanam industrial
complex, and the sampling inlet (PM2.5) is placed on the
roof of a four-story building. The site is also influenced
by traffics, residential heating, and agricultural burning
from nearby highway, residential/commercial areas, and
agricultural area. Ultrafine particle events were classified
into four types according to size ranges having high
particle number concentrations and hypotheses on the
possible causes for the enhanced particle numbers
(Watson et al., 2006): (1) 10 to 30 nm nucleation event,
(2) 10 to 30 nm or 50 to 80 traffic event; (3) 10 to 30 nm
photochemical event, and (4) 50 to 80 residential heating
event. In the traffic event, peak times of N(10-30 nm) or
N(50-80 nm) coincided with peak times of NOx, while in
the nucleation or photochemical events, peak times of N
(10-30 nm) were not consistent with those of NOx or CO.
The residential heating event was mostly observed in fall
and winter, while the photochemical event was observed
in summer.
John G. Watson, Judith C. Chow, Kihong Park, and
Douhlas H. Lowenthal (2006) J. Air & Waste Manage.
Assoc., 56,417-430
11I.6
High-time Resolution Observation of Ultrafine Particle Size
and Number Concentrations in an Urban Area. CHEOLHEON JEONG, Greg J. Evans, University of Toronto.
Traffic exhausts influence the number concentration of
ultrafine particles and might cause severe air quality
problems. High time resolution measurements are needed
to measure rapidly changing concentrations of ultrafine
particles on roads and in their vicinity.
The high-time measurements of particle size and number
concentrations have been conducted by using a Fast
Mobility Particle Sizer (FMPS, TSI 3091) and an
Aerodynamic Particle Sizer (APS, TSI 3321) since
January 2006 at a roadside building in downtown
Toronto. The FMPS using multiple low noise
electrometers measured particles from 5.6 to 560 nm in 32
channels at one second. The high-time resolution data
provide the ability to detect rapid changes in particle
number and size distributions. The size and number
concentrations were also measured by a Scanning
Mobility Particle Sizer (SMPS) consisting of a NanoDifferential Mobility Analyzer (Nano-DMA, TSI 3085)
and a Water-based Condensation Particle Counter
(WCPC, TSI 3786). Other co-located measurements
included SO2, NOx, O3, and meteorological variables
during the sampling period. In order to determine the size
distribution of nonvolatile fraction, ambient particles were
also analyzed by the SMPS equipped with a thermal
denuder at temperatures from 80 oC to 300oC.
Significant discrepancies in both particle numbers and
size distributions between the FMPS and the SMPS were
observed mainly due to diffusion losses in the SMPS
system. This result suggests that the use of high-time
resolution FMPS data can be used to better characterize
dynamically evolving particle number and size
distributions in the roadside environment. Based on the
FMPS data over 14 months, seasonal and diurnal
variations in the particle number and size distributions
will be presented. Moreover, local particle nucleation
events and regional nucleation/growth events will be
characterized and compared with gaseous pollutants as
well as meteorological parameters. The results will be
useful to obtain a better understanding of roadside particle
dynamics.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11I.7
Daily Variation in The Properties of Urban Ultrafine
Aerosol: Physical Characterization and Volatility.
KATHARINE MOORE, Zhi Ning, Leonidas Ntziachristos,
Constantinos Sioutas, University of Southern California; James
J. Schauer, University of Wisconsin, Madison, WI.
A summer air quality monitoring campaign focusing on
the evolution of ultrafine (less than 180 nm in diameter)
particle concentrations was conducted at an urban site in
Los Angeles from June through July 2006. Previous
observations suggest that ultrafine aerosol at this site are
generally representative of the Los Angeles urban
environment. Continuous and intermittent gas and
aerosol measurements were made over 4 weeks with
consistent daily meteorological conditions. Monthly
averages of the data suggest the strong influence of
commute traffic emissions on morning observations of
ultrafine particle concentrations. By contrast, in the
afternoon our measurements provide evidence of
secondary photochemical reactions becoming the
predominant formation mechanism of ultrafine aerosols.
The ultrafine number concentration peak occurs in the
early afternoon, before the maximum ozone concentration
is observed. The source of this offset is unknown and
requires further investigation. It is possible that the
chemical mechanisms responsible for secondary organic
aerosol formation evolve as atmospheric conditions
change and/or secondary semi-volatile components of the
aerosol re-volatilize due to the elevated peak temperatures
observed (ca. 30 to 35 degrees C) combined with the
increased atmospheric dilution during that time.
Measurements of the volatility of the ultrafine aerosol are
consistent with this interpretation as overall volatility
increases in the afternoon by up to 50 percent by volume
for selected initial particle diameters. There is also less
evidence of external mixing as the non-volatile particle
fraction decreases in the afternoon. Simultaneous
observations of aerosol composition are consistent with
this interpretation.
11I.8
Seasonal Variability of Aerosol Optical Properties in a
Mediterranean Coastal Zone. AUROMEET SAHA, Texas
A&M University; Marc Mallet, Laboratoire d'A
Aerosol optical properties were continuously measured in
the French Mediterranean coastal zone covering one
complete seasonal cycle in 2005-2006. BC mass
concentration, absorption and scattering coefficients,
columnar aerosol optical depth (AOD), number-size
distributions of fine and coarse particles were measured,
along with the surface meteorological parameters. Large
surface BC concentrations and high values of scattering
coefficients occurred during winter months, followed by
lower values during spring and summer. The columnar
aerosols showed different seasonal behavior, with high
AOD values occurring during the summer months, and
low to moderate values prevailed during the rest of the
period. Monthly mean AOD at mid visible wavelengths
ranged between ~0.1 and 0.34. The Angstroms coefficient
(estimated from the AOD spectra) remained high (<1.2)
during the entire study period, thereby indicating the
relative dominance of fine particles. The surface single
scattering albedo at 525 nm was in the range from 0.7 to
0.8, thereby indicating the dominance of absorbing
aerosols over this region.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
11I.9
Aerosol Light Absorption and Scattering at Four Sites in
and Near Mexico City: Comparison with Las Vegas, Nevada,
USA. GUADALUPE PAREDES-MIRANDA, W. Patrick
Arnott, University of Nevada - Reno and Desert Research
Institute; Nancy A. Marley, Jeffrey S. Gaffney, University of
Arkansas.
Four photoacoustic spectrometers (PAS) for aerosol light
scattering and absorption measurements were deployed in
and near Mexico City in March 2006 as part of the
Megacity Impacts on Regional and Global Environments
(MIRAGE). The four sites included: an urban site at
Instituto Mexicano del Petroleo (IMP); a suburban site at
the University of Tecamac; a rural site at La Biznaga
ranch; and a site at the Paseo de Cortes, on the saddle
between the volcanoes Popocatepetl and Iztaccihuatl. A
similar campaign was held in Las Vegas, Nevada, USA in
January-February, 2003. The IMP site gave in-situ
characterization of the Mexico City plume under
favorable wind conditions while the other sites provided
characterization of the plume, mixed in with any local
sources. The PAS used at IMP operates at 532 nm, and
conveniently allowed for characterization of gaseous
absorption at this wavelength as well. Instruments at the
second and third sites operate at 870 nm, and the one at
the fourth site at 780 nm. Light scattering measurements
are accomplished within the PAS by the reciprocal
nephelometery method. In the urban site the aerosol
absorption coefficient typically varies between 20 and
180 Mm-1 during the course of the day and significant
diurnal variation of the aerosol single scattering albedo
was observed probably as a consequence of secondary
aerosol formation. Comparisons with TSI nephelometer
scattering at the T0 site will be presented. We will present
the diurnal variation of the scattering and absorption as
well as the single scattering albedo and fraction of
absorption due to gases at the IMP site and compare with
Las Vegas diurnal variation. Mexico City breaths more
during the course of the day than Las Vegas, Nevada in
part because the latitude of Mexico City resulted in more
direct solar radiation. Further insight on the
meteorological connections will be discussed.
11I.10
Interactions between boreal wildfire and urban emissions.
KEITH BEIN, Yongjing Zhao, Anthony Wexler, University of
California Davis; Murray Johnston, University of Delaware.
A suite of particulate, gaseous and meteorological
measurements during the Pittsburgh Supersite experiment
were used to characterize the impact of the 2002 Quebec
wildfires on pollutant concentrations and physical and
chemical processes dominant in the region. Temporal
shifts in the size distribution of wildfire particles isolated using single particle mass spectrometry data combined with CO, NOx and O3 mixing ratios identified
two separate periods (Periods I and II) when the
measurement site was directly impacted by plumes of
unprocessed wildfire emissions; e.g. increases in ultrafine
wildfire particles, CO and NOx concomitant with a
decrease in O3 from intra-plume NOx titration.
Carbonaceous particle size distributions predominantly
associated with vehicular emissions, PM2.5 sulfate and
nitrate mass concentration and SO2 mixing ratio resolved
individual components of local and regional sources.
Single particle signatures indicated a period of intense
atmospheric processing following the arrival of
unprocessed wildfire emissions during Period II which
caused rapid growth of the ultrafine mode due to
simultaneous sulfate and secondary organic mass
accumulation. Satellite imagery, HYSPLIT trajectories
and radiosonde data characterized transport of wildfire
emissions from the Quebec fires down into the Pittsburgh
air shed. Although the emissions detected during the first
period were directly transported to the site by strong
northerly flow coupled to high pressure subsidence, the
emissions during Period II were observed to arrive from
the east as a small section of the westerly advected plume
was drawn back inland. Vertical HYSPLIT trajectories
were used to estimate an injection rate on the order of
1.6E18 wildfire particles/s for Period I, roughly
equivalent to emissions from about 50 coal fired power
plants or 500,000 vehicles.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
12E.1
Trends in Black Carbon Concentrations and Emission
Factors from Diesel Vehicles in California. THOMAS W.
KIRCHSTETTER, T. Novakov, Shaheen Tonse, Lawrence
Berkeley National Laboratory; Jeffery Aguiar, University of the
Pacific; David Fairley, Bay Area Air Quality Management
District.
We derived ambient black carbon (BC) concentrations
and estimated emission factors for diesel vehicles using
archived Coefficient of Haze (COH) data that was
routinely collected beginning in 1967 at many locations
throughout the San Francisco Bay Area. COH values are a
measure of the attenuation of light by particles collected
on a white filter and are proportional to BC
concentrations measured using the conventional
aethalometer. Monthly averaged BC concentrations are up
to five times greater in winter than summer, and,
consequently, so is the population's exposure to BC. The
seasonal cycle in BC concentrations is similar for all Bay
Area sites, most likely due to area-wide lowering of the
atmospheric inversion height and thus decreased pollutant
dispersion during wintertime. A strong weekly cycle is
also evident, with weekend concentrations significantly
lower than weekday concentrations, consistent with
decreased diesel traffic volume on weekends. The weekly
cycle suggests that, in the Bay Area, diesel vehicle
emissions are the dominant source of BC aerosol. Despite
the continuous increase in diesel fuel consumption in
California, annual Bay Area average BC concentrations
decreased by a factor of ~3 from the late 1960s to the
early 2000s. Diesel BC emission factors, based on
estimated annual BC concentrations, diesel fuel
consumption data and a study of on-road diesel vehicle
BC emissions, decreased from >10 g per kg in the late
1960s to <1 g per kg after 2000. Reductions in the BC
emission factor correspond to major milestones in
improved engine technology, emission controls and
changes in diesel fuel composition.
12E.2
Reconciling Emission Factors of PM Species Emitted by
Vehicles in Freeways and Roadway Tunnel Environments.
Zhi Ning, Harish C. Phuleria, MICHAEL D. GELLER,
Constantinos Sioutas*, University of Southern California.
Individual organic compounds such as hopanes and
steranes (originating in lube oil), selected polycyclic
aromatic compounds (PAHs) (generated via combustion),
and trace metals found in particulate emissions from
vehicles have proven useful in source apportionment of
ambient particulate matter. Currently, little ambient data
exists for a majority of these species. Three sampling
campaigns have been carried out in four different
environments with similar ambient conditions: a gasoline
only freeway, a heavy-duty diesel influenced freeway
(~20% diesel), a gasoline only tunnel, and a mixedvehicle (~4% diesel) tunnel. Trace organic species in the
size-segregated ultrafine (<0.18 micro-meters) and
accumulation (0.18 - 2.5 micro-meters) particulate matter
(PM) modes were measured with a high volume sampler.
Using the incremental increase of CO2 over the
background as an indication of the dilution ratio of
vehicle exhaust to ambient concentrations, this study
attempts to relate organic and trace metal species
concentrations measured in tunnels with those next to
freeways. Hopanes and steranes as well as high
molecular weight PAHs such as benzo(ghi)perylene
(BgP) and coronene levels are found comparable near the
freeways, while elemental carbon (EC) and lighter
molecular weight PAHs are elevated near I-710 compared
to CA-110. Very good agreement is observed between
CA-110 and LDV tunnel emission factors as well as I-710
measurements and corresponding reconstructed emission
factors from the tunnel for hopanes and steranes as well as
heavier PAHs such as BgP and coronene. In addition to
the organic compounds, emission factors for trace metals
in ultrafine and fine particles were also determined. Good
agreement with the reconstructed emission factors from
other studies was observed. This study demonstrates the
effective use of CO2-estimated dilution to associate
assorted vehicle-emitted PM bound compounds in distinct
vehicle-dominated environments.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
12E.3
On-Road Measurement of Gasoline and Diesel Vehicle
Emission Trends. George Ban-Weiss, John McLaughlin,
ROBERT HARLEY, University of California, Berkeley;
Thomas Kirchstetter, Melissa Lunden, Lawrence Berkeley
National Laboratory; Anthony Strawa, NASA.
Gaseous and particle-phase pollutants were measured
during summer 2006 at the Caldecott tunnel in the San
Francisco Bay area. Measurements were made in two
separate tunnel bores with high traffic volumes: the first
bore was reserved for light-duty (LD) vehicles, and the
second bore carried a mix of LD passenger vehicles and
heavy-duty (HD) diesel trucks. Particulate matter
(measured gravimetrically as PM2.5) and nitrogen oxide
(NOx) emissions decreased for both LD and HD vehicles,
compared to previous measurements made at the same
location in 1997. For NOx, the decrease in LD vehicle
emissions was larger than for HD vehicles (67 vs 27%).
In contrast, HD vehicle emissions of PM2.5 decreased
more than LD vehicle emissions (56 vs 36%). Black and
organic carbon measured using thermal optical analysis
methods accounted for similar fractions of PM2.5 in both
1997 and 2006, with a higher BC fraction in diesel
exhaust as expected. Diesel exhaust is now the dominant
source of NOx emissions, accounting for ~2/3 of total onroad vehicle emissions in California. High time-resolution
(1 Hz) measurements of CO 2, NOx, and black carbon
were used to calculate emission factors from individual
HD truck exhaust plumes. These data allow us to describe
the distributions of HD diesel vehicle contributions to
fleet-average emissions. Light scattering and absorption
of PM emissions were measured simultaneously by cavity
ring-down spectroscopy to assess the optical properties of
vehicle exhaust emissions, in relationship to mass
emission rates described above.
12E.4
Commonalities between Nonroad and Onroad Diesel
Emissions. HARSHIT AGRAWAL, Abhilash Nigam,
Varalakshmi Jayaram, Ajay Chaudhary, Kent Johnson, William
W. Welch, Wayne Miller, David Cocker, University of
California-Riverside, CE-CERT; Aniket Sawant (currrently at
Johnson Matthey Inc.); Sandip Shah (currently at Ford Motor
Company).
The promulgation of increasingly stringent emissions
regulations for on-road heavy-duty diesel vehicles has
brought the relatively unregulated nonroad sources into
sharper focus. These sources include marine engines,
locomotives, yard-tractors, heavy-duty diesel trucks, and
jet aircraft. There exists a need to characterize and
quantify emissions across multiple nonroad sources, with
the objective of understanding the relative contributions
of each to local and regional inventories. Case studies for
PM and NOx emissions from each source type will be
used to explore their similarities and differences. For
example, the relative carbonyl emission (formaldehydeacetaldehyde-acrolein) rates from partial oxidation across
all sources and engine loads were found to be quite
similar. However, elemental carbon and organic carbon
emissions were found to vary widely by engine type and
load.
The transportation industry is a major contributor to
emissions of pollutants that are typically regulated by
state and/or federal agencies. Emission indices for engines
involved in the transportation and distribution of
commercial goods will be evaluated on the basis of per
ton per mile of goods moved. This information,
combined with CO2 emissions, is used to identify the
most efficient transport of goods in terms of overall
emissions inventory.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
12E.5
Megacity Polycyclic Aromatic Hydrocarbon Exposure,
Emissions, and Transformations in Mexico City. LINSEY C.
MARR, Dwight A. Thornhill, Mei Jiang, Virginia Tech; Katja
Dzepina, Jose L. Jimenez, University of Colorado; Janet Arey,
University of California at Riverside; Scott C. Herndon,
Timothy B. Onasch, Ezra C. Wood, John T. Jayne, Charles E.
Kolb, Aerodyne, Inc.; Berk Knighton, University of Montana;
Miguel A. Zavala, Luisa T. Molina, Massachusetts Institute of
Technology.
The rapid proliferation of megacities and their air quality
problems is producing unprecedented air pollution health
risks and management challenges. Through three separate
field campaigns in Mexico City between the years 2002
-2006, we have measured emissions, concentrations,
transformation, and spatial and temporal variability of
particulate polycyclic aromatic hydrocarbons (PPAHs),
which are potent carcinogens. The use of a
photoionization aerosol sensor allows fast, sensitive
quantification of total PPAHs. Median total PPAH
concentrations along Mexico City's roadways range from
-3
60 to 910 ng m . These levels are approximately five
times higher than concentrations measured in the United
States and among the highest ambient values reported in
the literature. Through on-road measurements using the
Aerodyne Mobile Laboratory, we generate the first
estimates of PPAH and black carbon emissions from
motor vehicles in Mexico City, 57 +/- 6 and 1700 +/- 200
metric tons per year, respectively. The ratio of PPAHs to
aerosol active surface area is much higher along roadways
and in other areas of fresh vehicle emissions, compared to
ratios measured at sites influenced more by aged
emissions or noncombustion sources. Ambient PPAH
concentrations exhibit a strong diurnal pattern; they
-3
typically peak at ~100 ng m during the morning rush
hour and then rapidly decrease to a steady daytime level
-3
of <20 ng m . Intercomparison of three PPAH methods photoionization, integrated filters, and Aerosol Mass
Spectrometry - reveals that fresh combustion-generated
particles are rapidly coated by secondary aerosol during
the mid-morning hours. Poor intersite correlations of
PPAHs among six sites suggest that local sources
dominate ambient levels and that a single regional-scale
concentration cannot be used to represent exposure. The
results of this research can be used to help develop control
strategies for PPAHs and to conduct risk assessments of
exposure to ambient particles in megacities.
12E.6
Abrasion Particles Produced by Road Traffic. NICOLAS
BUKOWIECKI, Peter Lienemann, Christoph N. Zwicky,
Matthias Hill, Brigitte Buchmann, Robert Gehrig, Empa Materials Science and Technology; Markus Furger, Andre
Prevot, Urs Baltensperger, Paul Scherrer Institut.
Particle emissions of road traffic are generally associated
with exhaust emissions only. However, recent studies
performed in Switzerland identified a clear contribution
of non-exhaust emissions to the PM10 load of the ambient
air. These emissions are expected to consist of particles
from the abrasion of paving, tires, brakes and clutches and
are predominately found in the coarse mode fraction of
the ambient aerosol (aerodynamic particle diameter 2.5
-10 micrometer). However, quantitative information about
the contributions of the individual abrasion processes is
scarce up to now. It is of particular interest to know
whether abrasion emissions from paving or from vehicles
are dominating the non-exhaust PM10 contribution. This
would be necessary for effective PM10 reduction
scenarios. In Switzerland, the emissions of road traffic
abrasion particles into the ambient air are currently
characterized in the project APART (Abrasion PARTicles
produced by road traffic), funded by the Swiss Federal
Roads Authority (ASTRA) and the Swiss Federal Office
for the Environment (BAFU). The project aims at finding
the contribution of the non-exhaust sources to total trafficrelated PM10 and PM2.5 for different traffic conditions,
by determining specific elemental fingerprint signatures
for the various sources. This is achieved by hourly
elemental mass concentration measurements in three size
classes (2.5-10, 1-2.5 and 0.1-1 micrometers) with a
rotating drum impactor (RDI) and subsequent synchrotron
radiation X-ray fluorescence spectrometry (SR-XRF). To
quantify the different source contributions, the elemental
fingerprint measurements are embedded into a large set of
aerosol, gas phase, meteorological and traffic count
measurements. First results show elevated coarse mode
(2.5-10 micrometer) mass concentrations for a range of
trace elements (in particular Fe, Sn, Sb, and Ba) at a
traffic-rich measuring site in Zurich (Switzerland),
compared to the respective urban background. These
elements are thus likely fingerprints for abrasion particles
and will be used for a source apportionment and emission
factor calculations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
13E.1
Physical And Chemical Characterizatics Of Ultra-Fine And
Accumulation Mode Particles Near The Los Angeles Port.
MOHAMMAD ARHAMI, Andrea Polidori, Constantinos
Sioutas, University of Southern California.
During the spring of 2007 a diffusion charger (DC), a
photoelectric aerosol sensor (PAS), a condensation
particle counter (CPC) and a scanning mobility particle
sizer (SMPS) were operated at an outdoor urban site to
obtain real-time particle surface area, particle-bound
polycyclic aromatic hydrocarbons (PAHs), particle
number concentration and aerosol size distribution data,
respectively. Hourly outdoor fine particulate matter
(PM2.5 ), organic carbon (OC), elemental carbon (EC),
ozone (O3), carbon monoxide (CO) and nitrogen oxides
(NO, NO2 and NOX) concentrations were also measured.
The selected outdoor site was located in Long Beach, CA,
in the Los Angeles port area, about one mile from a major
freeway and right across a major power plant. Integrated
3-hr filter samples were concurrently collected throughout
the study and analyzed for the total concentration of
PAHs by gas chromatography/mass spectrometry (GC/
MS). Thus, the PAS signal was directly correlated to the
total ambient PAH concentration. Diurnal variations of all
measured particulate and gaseous species, important
reduced variables derived from the primary information
provided by the instruments employed in this study (e.g.
PAS/DC and DC/CPC) and correlations between all
measured variables will be discussed in detail to provide
new insights about the chemical and physical
characteristics of the sampled aerosol in both the
nucleation and the accumulation modes. This work was
conducted as part of the Southern California Particle
Center (SCPC) activities. The overall objective of the
SCPC is to investigate the underlying mechanisms that
produce the health effects associated with exposure to
PM, and to understand how toxic mechanisms and
resulting health effects vary with the source, chemical
composition and physical characteristics of PM.
13E.2
A Comparison of Particles at Multiple Locations in Jakarta,
Indonesia and Los Angeles, California. DANE
WESTERDAHL, University of California at Los Angeles; Scott
Fruin, Constantinos Sioutas, University of Southern California;
Manisha Singh, TSI.
INTRODUCTION
Jakarta is one of the most polluted cities in the world.
Air quality in this city of 12 million people is heavily
impacted by intense vehicular traffic whose emissions are
largely uncontrolled. Los Angeles is a US city of similar
size and population with polluted air. The residents of
both cities are likely to suffer health consequences of the
polluted air they breathe. This paper will report on the
nature of particles measured in Jakarta and compare them
with similar measurements made in Los Angeles.
METHODS
Particle monitoring instrumentation, including an
ultrafine particle counter, a Scanning Mobility Particle
Sizer Spectrometer, an Aerodynamic Particle Sizer
Spectrometer, a black carbon analyzer, and a beta
attenuation monitor, were operated in Jakarta during June
and July of 2005. PM 2.5 mass was also reported by a
nephalometer. These instruments produced time-resolved
observations at residential, urban near-roadway locations
and at a site distant from urban traffic. Similar
instrumentation was operated at several locations in the
Los Angeles area as part of the US EPA-funded Supersite
between 2002-2005. Data-Merge software from TSI was
employed to evaluate the number and mass distribution of
particles in the size range from 10 nm to 10 um.
RESULTS AND CONCLUSIONS
The instrumentation produced data with temporal
resolution which allows evaluation of the impacts of
vehicular activity and an opportunity to compare the
results of various methods that report particle mass.
Particles smaller than 1 micron diameter account for most
of the particle mass at the urban Jakarta sites. The site
approximately 20 km from densely urban Jakarta retained
high levels of very fine particles and black carbon,
indicating a regional impact of urban activities while Los
Angeles sites often contain a considerable portion of their
mass in particles larger than 1 micron.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
13E.3
Particle Volatility in the Vicinity of a Freeway with Heavyduty Diesel Traffic. SUBHASIS BISWAS, Leonidas
Ntziachristos, Katharine F. Moore, Constantinos Sioutas,
University of Southern California.
During February-March 2006, a major field sampling
campaign was conducted adjacent to the interstate 710 (I
-710) freeway in Los Angeles, CA. I-710 has high traffic
volume (ca. 11,000 vehicles h 1) and a high percentage
(17-18%) of heavy duty diesel vehicle (HDDV) traffic.
The volatility of ambient particles of 20,40, 80 and
120nm in diameter was investigated using a Tandem
Differential Mobility Analyzer (TDMA) at two locations close to the freeway (10m) and approximately 150m
downwind. The smallest particles (20nm) are largely
volatile at both locations. Larger particles, (e.g., greater or
equal to 40nm) showed evidence of external mixing, with
non-volatile fraction increasing with particle size. Particle
volatility increased with decreasing ambient temperature.
The HDDVs contribute to relatively larger non-volatile
particle number and volume fractions and greater external
mixing than earlier observations at a pure light-duty
gasoline vehicle freeway [ Kuhn et al., 2005, Atmospheric
Environment 39, 7174-7166]. Finally, the fraction of the
externally mixed soot particles decreased as the
downwind distance increased from the I-710, due to
atmospheric processes such as vapor adsorption and
condensation as well as particle coagulation.
Atmospheric Environment, 41 (2007): 3479-3493.
13E.4
The Morphology of Ultrafine Particles on and Near Major
Freeways. Teresa L. Barone, Oak Ridge National Laboratory;
YIFANG ZHU, Texas A&M University - Kingsville,.
Higher total particle number concentrations are present
near major freeways than in community air in Los
Angeles, CA. For particles in different size ranges,
distinct number concentration decay characteristics were
found with distance from freeways. Morphological
analysis of ultrafine particles with distance from a
freeway may give insight on the processes involved in
altering the number-size distribution. The objective of
this study is to systematically investigate ultrafine particle
morphologies in roadway micro-environments and
provide insight into their compositions and associated
mechanisms that affect their transport and transform away
from roadways.
Samples were collected while driving on the I-405 (~5%
diesel trucks) and I-710 (~ 25% diesel trucks) freeways in
April 2006 and were also collected 30, 60, and 90 m
downwind of I-405. Freeway aerosols were size selected
and passed through a nano-aerosol sampler and collected
on a TEM grid for morphology analysis. Typical
observed morphologies included aggregates, spheres,
irregularly shaped particles, and particles with multiple
inclusions. More than 90% of 50 nm opaque particles
were surrounded by a transparent (probably highly
volatile) material. This suggests that much of these
particles were heterogeneously internally mixed. The
fraction of aggregates encapsulated by transparent
material measured 90 m downwind of I-405 was
significantly less than the fraction measured on the
freeway (p < 0.001). Because aggregates are a primary
aerosol (directly emitted), this may indicate that
secondary aerosol (formed in the atmosphere) becomes
more prevalent with increasing distance from the freeway.
The fraction of particles with multiple inclusions
measured 90 m downwind of I-405 was significantly
greater than the fraction measured on the freeway (p <
0.001). The increase in the number of particles with
multiple inclusions with increasing distance from the
freeway suggests that dilution does not prevent particles
from colliding and merging which may alter the particle
size distribution.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
13E.5
Investigation on on-road ultrafine and submicron particles
by combining 1-s time-resolution data obtained from a FastMobility-Particle-Sizer and a Photoacoustic Instrument.
XIAOHONG YAO, Andrew J. Knox, Greg J. Evans, University
of Toronto; Jeffrey R. Brook, Environment Canada.
On-road ultrafine (<100 nm) and submicron particles
mainly originate from nucleation processes, primary
vehicular emissions and background particles. Some
freshly nucleated particles in vehicular plumes rapidly
grow to large particles via gas-particle condensation and
particle-particle coagulation while most nucleated
particles evaporate and/or are scavenged by coagulation.
These processes occur on the order of seconds or even
less than 1-s. Thus, high time-resolution instruments are
demanded for studying these rapidly varying particles.
The Fast-Mobility-Particle-Sizer and the Photoacoustic
Instrument can yield a 1-s particle spectrum and a 1-s BC
concentration, respectively. Combining simultaneous
measurements made by the two instruments allows for the
isolation of the on-road particles freshly emitted and/or
formed from background particles and for studying
transformation of particles and/or the influence of
different vehicle operating conditions on particle
concentration and particle size distribution. Two-months
of continuous measurements made by the two instruments
at a roadside site in Toronto were used to investigate size
distributions and transformation of on-road particles. The
results show that the two instruments simultaneously
detected vehicular emission spikes, which contributed up
to 10% of the total particle number concentration and BC
concentration during daytime on weekdays. The number
and volume size distributions of on-road particles emitted
from various vehicle types operating under different
conditions are presented. Bi-modal number size
distributions of ultrafine particles with two modes at 9-11
nm and 15-50 nm were frequently detected in spikes. The
tri-modal number size distributions with modes at 9-11
nm, 15-50 nm and 60-90 nm were less frequently detected
in spikes. The number and volume concentrations
normalized by BC concentrations are used to investigate
the difference between particle emission characteristics of
different vehicle types. Variations of concentrations and
size distributions of on-road particles in spikes are also
discussed in terms of the transformation of these particles.
13E.6
Relative Toxicity Of Size-Fractionated Particulate Matter
Obtained At Different Distances From A Highway. SeungHyun Cho, James R Lehmann, Q Todd Krantz, John McGee,
Mary J Daniels, Donald L Doerfler, M IAN GILMOUR, U.S.
Environmental Protection Agency, National Health
Environmental Effects Research Laboratory.
Epidemiological studies have reported an association
between proximity to highway traffic and increased
respiratory symptoms. This study was initiated to
determine the contribution of ambient particulate matter
(PM) to these observed effects. Ambient PM was
collected for 2 weeks using a three-stage (ultrafine: < 0.1
micro-meter; fine: 0.1-2.5 micro-meter; and coarse: 2.5
-10 micro-meter) high-volume impactor at two different
locations: 20 meter (Near-Road: NR) and 300 meter (FarRoad: FR) from an interstate highway in Raleigh, NC.
Collected samples were sonicated in methanol, and
resulting PM suspensions concentrated by evaporation,
diluted in physiological sterile saline and analyzed by
ICP-AES. Female CD-1 mice were intratracheally
instilled with saline, 25 or 100 micro-gram of each size
fraction, then assessed for airway reactivity to
methacholine and markers of lung injury and
inflammation at 4 and 18 hours post-instillation. In both
the NR and FR samples, fine particles comprised
approximately 55% of total PM mass while coarse and
ultrafine contributed 30% and 15%, respectively. Total
PM mass was 18% more in NR than FR. Higher
concentration of certain elements [Ba, Ca, Cr, Cu, Fe, Pb,
Sb, SiO2, Ti, Zn (< 50%); Al, Mn, Sr (< 30%)] was
measured in NR than FR suggesting potential for
increased toxicity of NR PM. Pulmonary endpoints
(neutrophils, IL-6, MIP-2, TNF-alpha, methacholine
reactivity) showed that coarse PM was associated with the
greatest effects, which were independent of collection
distance from the highway. By comparison, fine and
ultrafine PM-exposed animals and saline-control animals
exhibited minimal adverse effects. These results support
previous work, which has shown that on a mass basis,
coarse ambient PM produces greater inflammatory
responses than fine and ultrafine ambient PM, and in this
instance distance from production source did not
significantly enhance pulmonary toxicity. (This abstract
does not reflect EPA policy.)
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
14E.1
Intra-community variability in ultrafine particle number
concentrations in an urban mixed environment.
KATHARINE MOORE, Payam Pakbin, Constantinos Sioutas,
University of Southern California; Margaret Krudysz,
University of California at Los Angeles.
Ambient particulate matter concentrations in a community
are typically characterized with the hourly or daily
average PM2.5 mass at a single site. These observations,
while useful, do not capture well the high variability in
ultrafine aerosol particle concentrations due to their short
atmospheric lifetimes and sharp concentration gradients
from local sources. Exposure to ultrafine aerosol particles
is of increasing concern due to the association with
adverse health outcomes. Therefore, we deployed a
network of 13 condensation particle counters (CPCs, TSI
Model 3022A) in the communities of San Pedro,
Wilmington and West Long Beach, California, during
February 2007. These mixed industrial and residential
communities are heavily impacted by the emissions
associated with the activities of the Ports of Los Angeles
and Long Beach and represent a source region for
ultrafine aerosol in the Los Angeles air basin. Most of the
networked CPCs are contained within a circle roughly 8
km in diameter, centered in Wilmington. Site locations
within the network were developed by taking into
consideration known sources and prevailing local wind
patterns, as well as community concerns. The network
will monitor particle number concentration and local
meteorological conditions at one minute resolution
through November 2007. The data collection rate will
allow the investigation of intra-community variability in
particle number concentration as a function of time of
day, weather, seasonality, sources and other factors using
both statistical methods and back trajectory analyses.
Preliminary results from the study will be presented.
14E.2
Spatial and Temporal Trends of Organic and Elemental
Carbon as a Component of PM2.5 from the New York City
Area. Steve Kurian, MONICA A. MAZUREK, Min Li,
Rutgers, The State University of New Jersey; Stephen R.
McDow, National Exposure Research Laboratory, U.S.
Environmental Protection Agency.
Elemental (EC) and organic carbon (OC) ambient mass
concentrations were measured at four Speciation Trends
Network (STN) with collocated PM2.5 collectors from
5/2002 to 5/2003 as part of the Speciation of Organics for
Apportionment of PM-2.5 in the New York City Area
(SOAP) project. The sites were Queens, NYC (high
density urban residential); Elizabeth, NJ (adjacent to NJ
Turnpike); Westport, CT (downwind NYC); and a
regional background site in Chester, NJ (upwind NYC).
EC and OC ambient mass concentrations were determined
independently for the STN (N=58) and SOAP (N=78)
daily filters using NIOSH Method 5040. The two data
sets for EC and OC provided an opportunity to compare
measurements generated by both networks by using
paired daily filters at the sites (N=58). Descriptive
statistics were calculated for the SOAP and STN EC and
OC ambient mass concentrations. Mean, median,
standard deviation, and range values were slightly
different for the independent data sets. The SOAP
network EC and OC results were generally lower than the
STN network, although concentration versus sampling
date patterns were similar at a given site. Linear least
square regressions were performed for STN and SOAP
EC and OC ambient mass versus PM2.5 mass. Highest
correlation of OC mass with PM2.5 mass was seen for the
Wesport CT (R^2=0.74) site and lowest for Chester NJ
(R^=0.22) with intermediate values for Elizabeth NJ (NJ
Turnpike) (R^2= 0.66) and Queens NY (R^=0.46). These
results suggest PM2.5 mass is a moderate to good
predictor of OC mass for highly urbanized sites but
cannot adequately predict background OC mass with
potentially higher levels of secondary OC.
Although this work was reviewed by EPA and approved
for publication, it may not reflect official EPA policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
14E.3
Mobile Measurements as a Powerful Tool for
Characterization of Spatial Variability of Aerosol in Urban
Areas. ANDREY KHLYSTOV, Denina Hospodsky, Duke
University.
We present a new approach to characterize spatial
variability of aerosol on a neighborhood scale in urban
areas using mobile measurements. As examples we use
the results of the recently conducted field studies in
Wilmington, DE and Raleigh, NC. In the first study
mobile measurements were performed over a 4 km by 4
km area of downtown Wilmington for three components:
formaldehyde (representative of volatile organic
compounds and also photochemically reactive pollutants),
aerosol size distribution (representing fine particulate
matter), and hexavalent chromium (representative of toxic
metals). These measurements were used to construct
spatial and temporal distributions of air toxics in the area
that show a strong temporal and spatial variability. The
dynamic behavior of the four pollutants appears to be
influenced by the relative contribution of local (mostly
primary) sources and long-range (secondary) sources. In
the second study the effect of the sound barrier was
investigated using a mobile unit which provided highly
resolved gradients of ultrafine aerosol concentration as a
function of the distance from the highway.
14E.4
Fine-Scale Spatial and Temporal Variability of PM Number
and Size Distributions within a Community. MARGARET
KRUDYSZ, University of California, Los Angeles; Katharine
Moore, Michael Geller, Constantinos Sioutas, University of
Southern California.
Due to their short atmospheric lifetimes and strong
dependence on local sources, ultrafine particle (UFP)
numbers vary significantly on very short spatial and
temporal scales. Therefore, measurements of particle
number concentration are necessary to determine
exposure gradients to UFP. Further, simultaneous
observations of the particle size distributions can help in
identifying the types of aerosols present at different
sampling sites and the effects of photochemistry and
aerosol aging on a local scale. Starting in April 2007, we
monitored both particle number concentration and size
distributions at three sites within 2-5 miles of each other
in the Wilmington/Long Beach, California area using
Condensation Particle Counters and Scanning Mobility
Particle Sizers (SMPS), respectively. This area includes a
complex mix of industrial (refineries, power plants), and
transportation sources (marine vessels, diesel trucks, port
activities), all of which can influence UFP number
concentrations and size distributions. The instruments are
located at sites along well-defined wind trajectories,
which allows for tracking the particles' evolution as they
move downwind. SMPS data from each site are divided
into 10 size intervals (12-50 nm, 50-75 nm, 75-100 nm,
100-200 nm, 200-250 nm, 250-330 nm, 330-400 nm, 400
-450 nm, 450-550 nm, and 550-650 nm) to determine the
relationships between sites as a function of particle size.
In addition to correlation analyses, Coefficient of
Divergence analyses are conducted to investigate the
relationship between spatial variability and particle size.
These methods can provide a measure of source
similarities between the sampling sites. Results from this
study will help to identify UFP source profiles, spatial
variability, and aerosol transformation on a community
scale.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
14E.5
Sources and Causes of Spatial Variability in Coarse
Particulate Matter Concentrations in Detroit, Michigan.
JONATHAN THORNBURG, Charles Rodes, RTI International;
Ron Williams, U.S. EPA NERL.
Sample collection for the 108 participants of the Detroit
Exposure and Aerosol Research Study (DEARS) covering
2004 to 2007 was completed recently. Two primary
objectives of DEARS were to: 1) determine the
associations between concentrations measured at central
site monitors and outdoor residential monitors and 2)
describe the physical and chemical factors affecting these
relationships. Data for coarse particles (between 2.5 and
10 micrometers) were collected successfully during the
final two sampling seasons. This presentation will
examine these two objectives in relation to the coarse
particle concentrations measured across the Detroit metro
region.
A compact, battery powered coarse particle sampler
(CPEM) developed by RTI was deployed during the
summer 2006 and winter 2007 sampling seasons. An
advantage of the CPEM is its ability to collect PM2.5 and
coarse particles simultaneously. Outdoor and indoor
samples were collected in five distinct census areas of the
Detroit plus a central community monitoring site.
Approximately 300 CPEM samples were collected per
season.
Coarse particle concentrations measured by the CPEM at
the central community site agreed well with collocated
Dichotomous sampler concentrations. The regression
slope for summer 2006 data was 0.94 and the intercept
was -0.09. Comparison of average coarse particle
concentrations measured in the five census areas and the
central community site during summer 2006 indicated a
spatial non-uniformity existed. Two census areas, located
in a heavily industrial area of Detroit, had coarse particle
concentrations twice as high as the other census areas and
the central community site. Further analysis examining
the influence of local sources, seasonality, meteorological
conditions, and potentially correlated toxic gas markers
will be conducted to determine the significance and cause
of the spatial non-uniformity in coarse particle
concentrations.
14E.6
Spatial Variability of PM10-2.5 Measured with Passive
Samplers. Darrin Ott, Naresh Kumar, THOMAS PETERS, The
University of Iowa.
Since atmospheric coarse particulate matter, PM10-2.5,
often exhibits high spatial variation in comparison to fine
particulate matter (PM2.5), the use of coarse particles
monitored at a central monitoring station alone may
introduce considerable uncertainty in exposure
assessment and subsequently may result in exposure
misclassification. Therefore, it is critically important to
estimate spatially detailed surfaces of PM10-2.5 for
exposure assessment. This work introduces real-time
aerosol mapping and passive sampling as inexpensive
alternatives to filter-based sampling for coarse particles.
The study was implemented in two stages. In the first
stage, the pilot data on PM10-2.5, PM2.5, and ultrafine
particles were measured in a medium-sized Midwest City
(Iowa City, IA) with real-time particle monitors mounted
in a van. The analysis of these data revealed that 34 sites
identified by an optimal spatial sampling design were
needed to capture 95% of the spatial variability in coarse
particles. In the second stage, PM10-2.5 was measured
over three seven-day sampling periods with passive
samplers deployed at the identified 34 sites across the
city. At a control site, PM10-2.5 and PM2.5 were
measured with a filter-based dichotomous sampler
simultaneous with passive measurements, and wind
speed, temperature, and relative humidity were logged.
PM10-2.5 was observed to range from 5.0 ug/m3 to 31.7
ug/m3. The coefficient of variation of PM10-2.5 across
the city for each sampling period was 23%, 24%, and
30%. The coefficient of variation for collocated
measurement of PM10-2.5 with the passive samplers was
11%. PM10-2.5 measured passively correlated well with
that measured with the dichotomous sampler (r = 0.99).
Although this work was reviewed by U.S. EPA and
approved for publication, it may not necessarily reflect
official Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
16E.1
Regional Transport of Secondary Particulate Matter in
California with Source Contribution Analysis. QI YING,
Michael J. Kleeman
The San Joaquin Valley (SJV) of California has the worst
wintertime particulate air pollution problem in the nation.
Previous studies have show that secondary ammonium
nitrate and organic carbon emitted from residential wood
combustion are the major sources of particulate matter.
The major sources for secondary nitrate are diesel and
gasoline engines. Elevated ammonium nitrate
concentrations are relatively uniform throughout the
Valley, even in regions where sources of nitrate are
scarce, indicating a significant regional transport of
secondary pollutants and/or their precursors. However,
the exact amount of materials transported from one region
to another has never been quantified.
To better understand this regional pollutant transport
phenomena, the source-oriented CIT/UCD air quality
model is applied to determine the significance of the
sources in one sub-region to the primary and secondary
airborne particles in other sub-regions of the SJV. The air
quality and meteorology data collected during the
wintertime California Regional Particulate Air Quality
Study (CRPAQS) is used to evaluate the significance of
regional pollutant transport. The entire Valley is divided
into 9 sub-regions. Emissions of gases and particles from
each region are tracked separately through a full
simulation of transport and photochemical reactions. In
this way, the source contributions from different regions
to the particulate matter concentrations at a given receptor
location are directly determined. During the highly
stagnant periods in the modeled episode, inter-region
transport of pollutants from the southern part of the SJV is
not significant. However, in the period immediately
following the stagnation episode, transport of secondary
pollutants from the southern Valley towards northern
Valley is significant. Primary pollutants have a radius of
influence much smaller than secondary pollutants, and
their impact on air quality is mainly localized. Several
additional runs with different artificial wind fields are
performed to determine the approximate range of
influences for secondary pollutants.
16E.2
Modeling a wintertime PM2.5 episode in the California
Central Valley. BETTY K. PUN, Rochelle T. Balmori,
Christian Seigneur, Atmospheric and Environmental Research,
Inc.
The Community Multiscale Air Quality Model (CMAQ)
and a version with the Model of Aerosol Dynamics,
Reaction, Ionization, and Dissolution (MADRID) are
used to model a wintertime PM2.5 episode that took place
in December 2000 to January 2001. During this episode,
ammonium nitrate was the most abundant component on
average, but carbonaceous aerosols dominated the highest
24-hour average PM2.5 concentrations in urban areas. An
aloft nighttime nitrate formation mechanism has been
postulated based on observations. The modeling results
are used to elucidate further details of such a mechanism,
including the conditions conducive to nitrate formation
aloft, the chemistry involved, the spatial distribution of
the atmospheric chemical species of interest, and the
evolution of vertical concentration profiles as a function
of time resulting from vertical mixing. The models'
ability to reproduce the horizontal and vertical spatial
distribution of carbonaceous aerosols is used to infer the
accuracy of the representation of primary and secondary
aerosols. Wintertime transport of primary aerosols and
formation of secondary aerosols are analyzed based on
modeling results. Discrepancies between observations
and modeling results provide a basis for recommendations
for future research on model development and
measurements.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
16E.3
Reconciliation of an emission based model and a source
based model via source apportionment of PM2.5 - Part 1.
Organic molecular markers. Jaemeen Baek, Bo Yan, Sangil
Lee, Yongtao Hu, Mei Zheng, ARMISTEAD G. RUSSELL,
Georgia Institute of Technology; Sunkyoung Park, North
Central Texas Council of Government.
In order to improve accuracy of source apportionment of
fine PM in an emission-based air quality model and a
receptor model, detailed comparison of simulated and
measured organic molecular markers was performed, by
combining source apportionment results from CMAQ
with source profiles in a chemical mass balance model.
The Southeastern Aerosol Research and Characterization
Study (SEARCH) monitoring data measured in July 2001
and January 2002 were used. Major sources of organic
carbon and their related tracers, such as wood burning
with levoglucosan, mobile sources with hopanes, and
natural gas combustion with PAHs are studied.
Errors in emission inventories were addressed as the
major factors that caused differences between simulation
and observation. Using the detailed speciation,
comparison with other gaseous and particulate species
suggest that the OC fraction of PM emissions in Atlanta
are a factor of 1.2 to 4 low in the summer and by a factor
of seven in winter. Emissions from natural gas
combustion appear to be 60% to 85% high in winter, and
emissions from meat cooking are about a factor of two
high in both seasons. Emissions of soil/road dust appear
low by a factor of two in summer and high in winter.
Decreases in soil/road dust will result in a decrease in fine
PM in CMAQ simulation of more than 4 micro-gram/m 3.
Comparison of simulated with measured levoglucosan did
not match with OC comparison, and suggest that the
current wood burning source profile may not be
applicable for SEARCH sites, or that there is oxidation.
16E.4
A Comparison Study of CMAQ Aerosol Prediction Using
Two Thermodynamic Modules: UHAERO V.S.
ISORROPIA. FANG-YI CHENG, Daewon Byun, Andrey V.
Martynenko, Jiwen He, University of Houston.
The accurate prediction of the gas/particle partitioning of
semi-volatile inorganic aerosol components is a
challenging task. In this study, a new inorganic gasaerosol equilibrium module UHAERO is incorporated
into the U.S. EPA Models-3/CMAQ 3-D air quality
model to assess the aerosol prediction capability. The
CMAQ/UHAERO simulation results are compared with
the one using ISORROPIA module.
The differences between UHAERO and ISORROPIA
modules can be distinguished based on three general
features: (1) methods of computing activity coefficient for
aerosol-phase species, (2) methods of computing the
aerosol water content, and (3) the numerical techniques to
determine the equilibrium state. For instance,
ISORROPIA uses a priori specification of the presence of
solid phases at a certain relative humidity but UHAERO
predicts both deliquescence and crystallization based on
thermodynamics. In general, the UHAERO is developed
with an efficient computational framework and easily
cooperated with different activity coefficient models (PSC
and ExUNIQUAC activity coefficient models are
implemented currently.)
The CMAQ simulations are conducted for a summer
(2001 July) and winter (2002 January) episode. The
resolution is at 36-km regional scale. The observed
CASTNET and IMPROVE datasets over the continental
U.S., and super site programs at Atlanta and Pittsburgh
are used to evaluate the model performance. In this study,
the simulated metastable and deliquescence behavior will
be addressed and compared between two modules
specifically over the low relative humidity and low
temperature regions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
16E.5
Response of Regional and Urban Air Quality to Future
Changes in Climate and Emissions. JOHN DAWSON, Pavan
Racherla, Barry Lynn, Peter Adams, Spyros Pandis, Carnegie
Mellon University.
The Global-REgional Climate-Air Pollution modeling
System (GRE-CAPS) was used to simulate the effects of
future changes in climate and emissions on regional and
urban ozone and PM2.5 concentrations. GRE-CAPS
consists of a general circulation model/chemical transport
model (GISS II'), a regional meteorological model
(MM5), and a regional chemical transport model
(PMCAMx). This system is used to quantify the effects of
changes of changes in global climate and emissions on the
regional and urban air quality in the US.
Five present-day Januaries and Julys were simulated, and
the results were compared to several future scenarios for
the 2050s. A full change set of simulations, based on the
IPCC A2 scenario, was run, including changes in climate,
intercontinental transport, global emissions, and US
emissions. An alternative future scenario, based on IPCC
B1, was also simulated. In addition to these, future
scenarios were run in which either climate,
intercontinental transport, or Eastern US emissions were
kept at present-day values in order to isolate their
individual effects on air quality.
16E.6
Impact of Sea-Salt Aerosol on the Weekend Effect.
ALEXANDER COHAN, Donald Dabdub, University of
California, Irvine.
The weekend effect has become an important issue in
regulation as it may suggest that controlling NOx would
be counter productive to reducing ozone concentrations.
Current hypotheses suggest that the dynamics of NOx
(changes of quantities and timing NOx emissions rates)
explain in part the increase in ozone concentrations. In
the past few years there have been new discoveries of
atmospheric processes such as the chemistry of sea-salt
aerosol in coastal areas. This study quantifies the impact
that sea-salt aerosol has on air quality in urban regions.
The focus area of this study is the South Coast Air Basin
of California. Particular emphasis will be placed to the
impact of sea-salt aerosol to the weekend effect.
The relative importance of changes in climate,
intercontinental transport, and Eastern US emissions are
examined in order to quantify their effects on future
ozone and PM concentrations. The use of both the A2 and
B1 scenarios also allows estimates of upper and lower
bounds, respectively, of emissions and intercontinental
transport changes.
Preliminary results indicate an appreciable effect of
changes in climate and transport on air quality, without
taking into account changes in emissions. Changes in July
daily maximum 8-hour average ozone were minor over
most of the domain, with a large increase (5 ppb) in the
Southeast. Average PM2.5 concentrations decreased by 0.7
-3
micrograms m in January and increased by 2
-3
micrograms m in July, driven mainly by changes in
precipitation, temperature, and transport.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
17E.1
Surface Chemistry Analysis of Urban and Rural Aerosols
During a Night-time High PM Burning Event in Yuma, AZ.
HEATHER A. HOLMES, Bonnie J. Tyler, Richard E. Peterson,
Eric R. Pardyjak, University of Utah.
A field study designed to investigate the spatial and
temporal variability of aerosols during high particulate
matter (PM) events along the US/Mexico border near
Yuma, AZ was run during the week of March 18, 2007.
The experiments were designed to quantify the transport
and determine the chemical composition of aerosols
generated via events such as high wind and burning. The
field study included two \fully instrumented\ monitoring
sites; one rural and one urban, equipped with sonic
anemometers, continuous particulate concentration
monitors and ambient aerosol collection equipment. In
addition to the two main monitoring sites, six additional
urban locations were equipped with TSI DustTrak
monitors to allow for the investigation of the spatial and
temporal distribution of PM2.5 concentrations. All
DustTrak devices were collocated with Beta-Attenuation
Mass (BAM) monitors deployed by the Arizona
Department of Environmental Quality. The focus of the
work to be presented will be on comparing organic
surface chemistry of aerosols collected from the rural and
urban sites during a nighttime high PM burning event that
originated near the border. Aerosol samples were
collected using two Graseby-Anderson eight stage
cascade impactors that collected size segregated samples
on aluminum substrates. The aerosol sample's surface
composition will be analyzed using time-of-flight
secondary ion mass spectrometry (ToF-SIMS) for organic
material. The interaction between local atmospheric
conditions and both concentration distributions and
chemical composition will be discussed. Additional
information regarding organic and inorganic chemical
species using integrated carbon and X-ray fluorescence
analysis will be presented.
17E.2
Characteristics of PAHs in Ambient Nanoparticles Collected
by Nanoparticle Sampler with Inertial Filter. M.
FURUUCHI, Y. Otani, S. Tsukawaki,Kanazawa University,
Japan; N. Tajima, T. Kato, KANOMAX Inc., Japan; P. Hang,
Authority for the Protection of the Site and the Management of
Angkor and the Region of Siem Reap (APSARA), Cambodia; S.
Sieng, Ministry of Industry, Mines and Energy, Cambodia.
The information on the composition of atmospheric
nanoparticles with respect to particle size is the key in
order to investigate the health effects of atmospheric
nanoparticles. In this study, a newly developed device,
which can separate nanoparticles from the larger particles
using fibrous filters, was used to discuss characteristics of
Poly-cyclic aromatic hydrocarbons (PAHs) in the ambient
aerosol sized down to nanosize range. A nanoparticle
sampler, which consists of three stages: PM10-2.5, PM2.5
-0.05, PM0.05. Stainless steel fiber mat (fiber diameter 8
micro-m, 8mm thickness, 4mm diameter, packing density
0.0065) was used as a filter material. 50% cut off size was
0.044 micro-m at 40L/min of flow rate. Samplings of
ambient aerosol were conducted in several locations in
different countries: Japan, Thailand and Cambodia for 24
hours as well as during the daytime and nighttime. Fifteen
diffrent PAHs (Nap, Ace, Phe, Ant, Fle, Flu, Pyr, BaA,
Chr, BaP, BbF, BkF, DbA, IDP and BghiPe) were
analyzed using an HPLC.
The PAHs mass fraction was found to be lager in finer
particles, particularly for PAHS with 4-6 rings (Flu to
IDP). The total PAHs concentration becomes largest in
some locations for PM2.5-0.1. PM0.1 particles contain a
less fraction than PM2.5-0.1 but 2-4rings of PAHs (Nap,
Ace, Phe, Ant Flu) is larger in this range. These are
similar in a different sample. More carcinogenetic
compounds (5-6rings) have larger fractions in finer
particles less than 2.5 micro-m. PM0.05 contains more 4
-5 rings than PM2.5-0.05.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
17E.3
Organic Speciation of Vehicle Exhaust Particulates:
Gasoline and Diesel Light Duty Vehicles. MIN LI, Monica A.
Mazurek, Claire Belisle, Majad Ullah, Rutgers University; Shida
Tang, Robert Whitby, New York Department of Environmental
Conservation.
17E.4
Wintertime nitrate size distribution as an indicator of
regional or local sources during the 2007 Seasonal
Particulate Observations in the Region of Toronto (SPORT)
Campaign. KRYSTAL J. GODRI, Greg J. Evans, Jay Slowik,
Jonathan Abbatt, University of Toronto.
On road vehicle emissions is a major source of particulate
matter pollution, especially in urban areas. In this study,
organic speciation of fine particulate matter (PM2.5,
d<2.5 um) was carried out on 18 light-duty vehicles tested
from July, 2005 to May, 2006 on a chassis dynamometer
over different driving cycles. Twelve gasoline vehicles
were evaluated including a gasoline-electric hybrid. Two
of the vehicles operated on compressed natural gas
(CNG), and four were diesel vehicles. Three gasoline
vehicles were tested with both summer and winter fuels.
Size-resolved soluble aerosol inorganic (Cl , SO4 , NO3 ,
NO2-, NH4+) and organic species were measured with
collocated Aerodyne Aerosol Mass Spectrometer (AMS)
and Dionex Gas Particle Ion Chromatography (GPIC)
systems. Sampling was conducted during a three week
intensive sampling campaign in winter 2007 beside a busy
roadway in downtown Toronto, Canada. The GPIC also
measured the gaseous precursors of the inorganic aerosols
(HCl, SO2, HNO 3, HNO 2, NH3). AMS mass distributions
of particulate nitrate showed two distinct modes, one at
~100 nm and one at ~400 nm, which were attributed to
different sources. The larger nitrate particles are
associated with elevated particulate sulphate
concentrations of the same size and westerly winds
carrying industrial emissions from outside the greater
Toronto area. The smaller nitrate mode correlates with
gaseous HNO3 but not with wind direction, suggesting
these particles originate from local gaseous emissions.
The midday maxima exhibited by the small particulate
NO3- is influenced by the photochemical production of
precursor gases. Subzero temperatures, high relative
humidity, and supersaturated ammonia and nitric acid
concentrations measured during the campaign all favour
formation of the smaller nitrate particles. The likely
source of the small NO3- particles is condensation of
gaseous HNO3 and NH3 species onto preexisting
particulates. Adequately elevated relative humidity also
allows for ammonium nitrate deliquescence
characteristics to be investigated. Observed ammonia and
nitric acid gas concentrations are compared to results
from the Aerosol Inorganic Model (AIM-II) to investigate
partitioning between the particle and gas phase.
Comparison of the ambient measurements and modeling
results helps elucidate the effect of non-nitrate aerosol
components (e.g. organics) on nitrate partitioning. The
fate of the small nitrate particles will also be discussed.
A detailed chemical profile was constructed for the
vehicle exhaust particulates collected from each vehicle
and fuel type. More than 100 organic molecular marker
compounds quantified by gas chromatograph/mass
spectrometry (GC/MS). The major organic components
identified were n-alkanes, PAH (polycyclic aromatic
hydrocarbons), hopane, sterane, n-alkanoic acids and
benzoic acids. These chemical profiles will update
current vehicle emission profiles for motor vehicles
operating in the NY City metropolitan area. The detailed
chemical profiles corresponding to the fine organic PM
will provide new apportionment and modeling tools to
distinguish organic particulate pollution from gasoline
and diesel powered vehicles.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
-
2-
-
Atmospheric Aerosols: Urban
2007 AAAR Annual Conference Abstracts
17E.5
Measurements of nitropolycyclic aromatic hydrocarbons,
polycyclic aromatic hydrocarbons and azaarenes in urban
air particulates in east of France. OLIVIER DELHOMME,
Maurice Millet, Laboratoire de Physico-Chimie de l
17E.6
High Time-Resolved Chemical Mass Closure of Fine
Particles in Helsinki, Finland. SANNA SAARIKOSKI,
Minna Aurela, Kimmo Teinila, Timo Makela, Risto Hillamo,
Finnish Meteorological Institute.
Urban atmospheres contain various kinds of organic pollutants.
Among them, polycyclic aromatic hydrocarbons (PAHs) and
nitropolycyclic aromatic hydrocarbons (NPAHs) are of
particular interest to the environmental analytical community
because of the extraordinary mutagenic and carcinogenic
activities, even at low concentration levels [1]. PAHs and
NPAHs in the atmosphere mainly originate are direct emissions
from combustion processes such as automobile exhaust and
coal- fired power plants emission. NPAHs can also be produced
in ambient air from gas- phase reactions of the parent PAH with
hydroxyl (OH) radicals during the day, and with nitrate (NO3)
radicals at night in the presence of oxides of nitrogen (NOx) [2].
In general the sources of azaarene are similar to those of
polycyclic aromatic hydrocarbons, namely vehicle exhausts,
coal burning and bitumen spreading.
Chemical composition of fine particles was investigated
in Helsinki, Finland at an urban background station
SMEAR III from June 2006 to March 2007. Major
inorganic ions were determined using a Particle Into
Liquid Sampler combined with two ions chromatographs
(PILS-IC), organic carbon (OC) and elemental carbon
(EC) were measured with a semi-continuous OCEC
analyzer (Sunset Lab Inc.) and PM2.5 was measured by a
Tapered Element Oscillating Microbalance (TEOM)
equipped with a Filter Dynamics Measurement System
(FDMS). Time-resolution was 15 min for the PILS-IC, 3
hours for OCEC analyzer and 30 min for the TEOM. Of
all aerosol components measured, EC concentration
varied most significantly during the day. It was strongly
related to the intensity of traffic since a clear workday-toweekend and a day-to-night variation was found for EC.
Also for nitrate the diurnal variation was evident. In
summer and in fall nitrate concentration decreased in the
afternoon, probably because of higher mixing height and
change in gas/particle equilibrium, whereas in winter the
maximum concentration of nitrate was measured in the
morning caused by the morning rush hour combined with
the low mixing height. For OC the diurnal variation was
similar to that of nitrate in summer whereas in fall and
winter the concentration of OC was rather constant in the
course of the day. Chemical mass, the sum of ions, OC
and EC, was compared with the measured PM2.5. The
differences in mass were found to be instrumental, related
to e.g. detection limits, different cut-off dimensions or
factors used in calculations.
The complexity of atmospheric environmental samples and the
low concentration levels of nitro- PAHs require a sensitive and
selective analytical method. The separation and quantification of
NPAHs and HAPs was carried out by reversed-phase high
performance liquid chromatography (HPLC) and fluorescence
detection while detection of azaarenes was made by using GCMS. However, for the NHAPs, the HPLC-fluorescence method
require on-line reduction of the NPAHs to their corresponding
amino polycyclic aromatic hydrocarbons (APAHs) since
generally, NPAHs exhibit only very weak fluorescence signals
[3].
Particle phase concentration of 12 NPAHs, 10 HAPs and 15
azaarenes were quantified in ambient air collected in downtown
Strasbourg (East of France) during four seasons between August
2006 and April 2007. Different ratios were studied for
understand the behaviour of NPAHs and PAHs. To evaluate the
relative contribution of primary sources versus atmospheric gas
phase formation on the distribution of NHAPs, the ratio of 2
nitropyrene and 1 nitropyrene have been applied. The ratios of
the NPAHs and PAHs were compared for the different seasons
for studied the NPAHs formation coming from PAHs.
Atmospheric concentrations measured for azarenes remains
poorly studied in France and first results obtained in Strasbourg
will be detailed.
[1] Jinhui, X. and Lee, F.S.C, Analytical Chimica Acta 416:
111- 115 (2000).
[2] Bezabeh, D.Z., Bamford, H.A., Schantz, M.M and Wise, S.
A., Anal. Bioanal. Chem. 375: 381- 388 (2003).
[3] Schauer, C., Niessner, R. and Poschl, U., Anal. Bioanal.
Chem. 378: 725- 736 (2004).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.1
A Method for Extracting Additional Information on the
Organic, Elemental and Pyrolyzed Carbon from Real Time
Measurements with the Sunset Carbon Aerosol Analyzer.
MIN-SUK BAE, James J. Schwab, Kenneth L. Demerjian,
University at Albany, State University of New York; Oliver
Rattigan, Dirk Felton, New York State Department of
Environmental Conservation.
2M.2
Interference of Organic Signals in Highly-time Resolved
Nitrate Measurements by Aerosol Mass Spectrometer. MinSuk Bae, James J. Schwab, QI ZHANG, Olga Hogrefe, Kenneth
L. Demerjian, University at Albany, State University of New
York; Silke Weimer, Paul Scherrer Institute; Kevin Rhoads,
Doug Orsini, Siena College; Prasanna Venkatachari, Philip K.
Hopke, Clarkson University.
Semi-continuous Organic Carbon and Elemental Carbon
(OCEC) instruments are becoming more widely employed
to measure the carbonaceous fraction of atmospheric
particulate matter. Determining accurate concentrations of
atmospheric OC and EC is necessary for identifying their
sources and predicting their effects on various
atmospheric processes. We have obtained hourly timeresolved measurements of OC and EC at Pinnacle State
Park (PSP) in upstate New York and the South Bronx,
New York City. OC and EC were determined using a
NIOSH-like protocol using a Sunset Real Time ECOC
Analyzer - that is, using four temperature steps to a final
temperature of 840 degrees C for OC and two steps to 850
degrees C for EC.
Highly time-resolved measurements of nitrate in ambient
aerosols were conducted by an Aerodyne Quadrupole
Aerosol Mass Spectrometer (Q-AMS or simply AMS)
and a Particle-into-Liquid Sampler with Ion
Chromatography (PILS) from field intensives at two sites;
an urban site in New York City (Queens College; QC) for
wintertime (January 22 to February 5, 2004) and a rural
site in southwestern New York state (Pinnacle State Park;
PSP) for summertime (July 20 to August 4, 2004).
In this study, the inorganic nitrate signal from Q-AMS
may contain significant interferences from organic
signals, especially in rural atmospheres. Analysis of the
QC data indicates a good agreement between the PILS2
nitrate and AMS-nitrate measurements (R = 0.94; linear
regression slope = 1.05). In addition, the m/z 30 and m/z
46 (two dominant ion fragments in nitrate mass spectrum)
2
signals tightly correlate at QC (R = 0.98) and have an
average ratio similar to that determined in the laboratory
for NH 4NO3 (m/z 30 / m/z 46 = 2.4). In contrast, at the
PSP site the correlation between PILS- and AMS-nitrate
was poor (R2 = 0.34), the AMS reported nitrate values
were substantially higher, and the m/z 30 to m/z 46 ratios
were generally much larger than 2.4. These observations,
together with evaluations by aerosol phase ion balance,
indicate that the AMS m/z 30 signals at PSP have been
strongly influenced by organic compounds that also
produce signals at m/z 30, including organic nitrates (NO
+
), oxygenated organics (CH 2O+), hydrocarbon-like
+
organics (C2H6 ), and nitrogen-containing organic
+
compounds (CH4N ).
There is currently no standard procedure for determining
concentrations of the empirical parameters OC and EC in
the atmosphere, and there is considerable controversy
over measurement methods and parameter designations.
In this study, the contour plots of the time series
carbonaceous concentrations associated with the analysis
temperature steps provide clear evidence that the
carbonaceous nature of sources changes significantly. The
identification of additional "Е“thermal fractions" with the
Sunset ECOC data are compared with organic chemical
species measured by aerosol mass spectrometer (AMS).
In addition, time series of split time variation determined
by optical laser transmittance could illustrate the behavior
of pyrolyzed carbon concentrations and shed light on its
sources.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.3
Single Particle Black Carbon and BC Mixing State
Measurements over Mexico City and Seattle: Results from
the MILAGRO and INTEX-B Studies. R SUBRAMANIAN,
Gregory L Kok, Droplet Measurement Technologies; Darrel
Baumgardner, Universidad Nacional Aut
Black carbon (BC) plays an important role in aerosol
radiative forcing by absorbing light and warming the
atmosphere. Recent studies have shown that the
absorption efficiency of coated black carbon particles
could be 1.5 times that of uncoated BC (Bond et al. 2006).
The DMT single particle soot photometer (SP2) measures
single particle black carbon mass using laser
incandescence, and can also provide information on the
mixing state of BC using scattering measurements. The
instrument was used on board the NSF C130 aircraft
during the March 2006 MILAGRO and April/May 2006
INTEX-B studies over Mexico City and Seattle
respectively, and on the DOE G1 aircraft during
MILAGRO. Data are available for all eleven C130 flights
during MILAGRO and for thirteen flights during INTEXB. The flights over Mexico sampled different conditions
including brown haze, fresh and aged emissions from
Mexico City, and clean areas, often on the same flight.
For the flight on March 18, 2006, about 10% of the
particles incandesced (i.e. contained BC); BC averaged
less than 3% of the particulate mass as estimated by the
SP2 (assuming ammonium sulfate as the non-BC mass).
Incandescent particles were usually composed of 40-60%
BC by volume. For the INTEX-B flight on May 1, 2006,
the incandescent particles contained 10-15% BC by
volume, which correspond well to AMS-reported
measurements of high sulfate and low organic content.
Use of either a core-shell or internally mixed model did
not make an appreciable difference since the aerosol
extinction is dominated by non-incandescing particles.
Complete results from both INTEX-B and MILAGRO
including an intercomparison test between the G1 and
C130 aircrafts will be presented at the conference.
2M.4
Carbonaceous aerosols in the remote free troposphere: A
time series from the Mauna Loa Observatory. STEVEN
HOWELL, Barry Huebert, John Zhuang, University of Hawaii;
Trevor Kaplan, Mauna Loa Observatory.
Since June 2005 we have operated a Sunset Labs
semicontinuous Organic Carbon/Elemental Carbon
(OCEC) analyzer at the Mauna Loa Observatory (MLO)
on Hawaii. At 3397 meters, the site is well above the
usual marine boundary layer and nighttime downslope
winds usually bring minimally perturbed air from the free
troposphere. Concentrations are normally very low, so
we opted to forego the usual organic/elemental split and
instead maximize sensitivity to total carbon (TC).
Monthly averages of nighttime samples (10:00 PM to
8:00 AM local time) range from 0.02 to 0.26 micrograms
-3
C m with maximum values in the spring, corresponding
to the elevated sulfate and calcium associated with
transport of Asian pollution and dust to MLO. At these
low TC levels, the performance of the denuder that
removes organic vapors from the sample stream is critical
and dominates the error calculation and hence the
detection limit. To improve the detection limit, we
installed a second OCEC analyzer in September 2006 so
blanks and samples can be simultaneous.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.5
Organic functional groups in submicron aerosol by FTIR
measurements in the Gulf of Mexico during TEXAQS/
GoMACCS 2006. Lynn M Russell, LELIA N HAWKINS,
Scripps Institution of Oceanography; Tim S Bates, National
Oceanic and Atmospheric Administration Pacific Marine
Environmental Laboratory.
To characterize the pollutants created in and transported
to the Gulf of Mexico and coastal Texas, a multi-platform
campaign was conducted involving ground, air, and
shipboard measurements. Organic compounds in primary
and secondary aerosol particles are of particular interest
due to the complexities associated with CCN activity in
organic-containing particles. We present results of FTIR
spectral analysis from approximately 100 filters collected
over 6 weeks of the 2006 Texas Air Quality Study / Gulf
of Mexico Atmospheric Composition and Climate Study
(TEXAQS/GoMACCS). Quantified functional group
concentrations include aromatic C=C-H, unsaturated
aliphatic C=C-H, saturated aliphatic C-C-H, organic
hydroxyl O-H, organosulfur C-O-S, and carbonyl C=O.
Aerosol organic fraction and functional group speciation
as measured by FTIR analysis is compared with results
from an Aerosol Mass Spectrometer (AMS) operated on
board the Ronald H. Brown. Specifically, carbonyl
concentration as determined by a peak in the infrared
-1
spectrum at 1720 cm is compared to the loading of m/z =
44 which has been established as a CO2 fragment in
previous AMS studies. Time-resolved total organic
concentration as derived from each technique is also
compared. The AMS particulate organic matter (POM)
50th percentile and 95th percentile are 1.27 micro-g m-3
and 14.53 micro-g m-3, respectively.
2M.6
Searching for Evidence of Acid-Catalyzed Enhancement of
Secondary Organic Aerosol Formation Using Ambient
Aerosol Data. ROGER L. TANNER, Kenneth J. Olszyna,
Tennessee Valley Authority; Eric S. Edgerton, ARA, Inc.;
Eladio Knipping, EPRI.
Laboratory experiments suggest that strong acids promote
formation of significant levels of secondary organic
aerosol (SOA), which is problematical since organic
aerosols have been implicated in health impacts of fine
PM. We report efforts examining hourly speciated fine
particle data for evidence of aerosol acidity-catalyzed
secondary organic aerosol formation. The assumption is
that, if SOA formation can be accelerated by acid
catalysis in fine particles, larger increases in the
concentrations of organic aerosol mass should occur on
days and in locations with more acidic aerosol (lower NH4
+
/SO 4= ratios) compared to neutral aerosol days. The
approach used herein is based on examining data sets
from which the hourly acidity of PM2.5 aerosols can be
estimated, and for which hourly organic carbon content
have been measured, then selecting episodes and
statistically relating within-day SOA buildup with aerosol
acidity. Using this approach we have examined (1)
VISTAS Focus Site data from Look Rock, TN, Cape
Romain, SC, and Millbrook, April, 2003-December,
2004; (2)SEARCH data from CTR and YRK for 2003 and
2004; and (3)other hourly TVA data from the Tennessee
Valley.
VISTAS focus site data from Look Rock, for mid-July to
mid-August, 2004, included PILS-IC hourly ammonium,
nitrate data (J. Collett, personal communication).
Surprisingly, nighttime average imputed acidities
generally exceeded daytime values, but data showed no
increases in OC with increasing aerosol acidity. Plots of
OC changes time-delayed from the observed imputed
acidity also show no clear relationship. SEARCH network
data (2003-2004) for rural Centreville, AL (CTR) and
Yorkville, GA (YRK) sites have also been examined for
+
=
nitrate-corrected NH4 /SO 4 - OC relationships. Warm
season acidity levels were higher at CTR than YRK, and
daytime levels exceeded those at night. However, no
consistent evidence of positive correlation between OC
levels and aerosol acidity have yet been found, even with
lags of up to 6 hr.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.7
Investigating the chemical nature of humic-like substances
in atmospheric aerosols using LC-MS/MS. ELIZABETH A.
STONE, Curtis J. Hedman, Martin M. Shafer, James J. Schauer,
University of Wisconsin-Madison and Wisconsin State
Laboratory of Hygiene.
Liquid chromatography coupled with tandem mass
spectrometry (LC-MS/MS) was used to investigate the
chemical nature of humic-like substances (HULIS) in the
atmospheric particulate matter (PM). HULIS has been
observed in the atmosphere and has been suggested to
comprise a significant portion of the unresolved organic
mass. Presently, HULIS is believed to be of secondary
origin, although its chemical structure and formation
mechanisms are not fully understood. Ambient PM
collected at on quartz fiber filters was extracted with ultra
pure water using sonication. Water-soluble organic
carbon (WSOC) in the extracts was measured and the
HULIS portion of WSOC was isolated using reversedphase LC. Mass spectra of the HULIS material under low
fragmentation conditions were used to characterize the
molecular weight range of WSOC and to identify highmolecular weight species. MS/MS was used to generate
product ion spectra of these compounds and important
functional groups were identified based on comparisons
to HULIS-relevant surrogate standard compounds.
Spectral signatures of HULIS were compared and
contrasted between spatially resolved locations in North
America where primary sources were well-understood.
Correlations between primary sources and secondary
HULIS are discussed.
2M.8
Airborne aerosol measurements over West Africa during the
AMMA SOP 1 and 2 field campaign. GERARD CAPES,
Hugh Coe, Paul Williams, Jonathon Crosier, University Of
Manchester, UK; Jennifer Murphy, Claire Reeves, University Of
East Anglia, Norwich, UK; Doug Parker, University Of Leeds,
UK.
In July-August 2006, a large field campaign took place in
West Africa, forming part of the international AMMA
(African Monsoon Multidisciplinary Analyses) project.
Several ground based sites and 5 aircraft were involved in
the project. This paper presents findings from the aerosol
particle measurements made on board the UK Facility for
Airborne Atmospheric Measurements (FAAM), a
BAe146 research aircraft. The operating region covers a
large area of West Africa, and some of the Atlantic Ocean
off the coast of Senegal.
A preliminary analysis of the AMS data has been
performed in conjunction with data from the other
instruments, and a very clean mass loading across the
operating region, though examples of biomass burning,
dust, urban plumes, and areas dominated by biogenic
emissions have been identified. Biomass burning layers
were encountered between 6 and 10N. These showed a
similar spectral signature to those measured in the same
region in the dry season and appear to have originated in
the southern hemisphere. Urban plumes from Niamey and
Lagos were measured, and were characterised by a
significant hydrocarbon signature typical of near-source
fossil fuel emissions. Emission ratios relative to CO have
been estimated. In regions of biogenic emission, where
high levels of isoprene and other VOCs were measured,
there was little evidence of enhanced organic aerosol
loading, typically observed when secondary organic
aerosol (SOA) has been produced. The extent to which
observable organic loading is observed has been
determined on the basis of statistical comparison as
individual data points are rarely above the detection limit
of the instrument. This illustrates that the loadings are low
and that models of SOA formation may predict similar
concentrations to observations in clean, biogenic
environments.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.9
Evaluation of Influences in Ambient Organic Compounds
Levels by the Operations of a Coal-Fired Power Station in
Tong Liang, China. STEVEN SAI HANG HO, Judith C.
Chow, John G. Watson, Desert Research Institute; Deliang
Tang, Frederica Perera, Columbia University.
2M.10
Organic Speciation of Detroit Exposure and Aerosol
Research Study (DEARS) Samples for Source
Apportionment. STEPHEN R. McDOW, John Turlington,
Sania W. Tong Argao, Ronald Williams, National Exposure
Research Laboratory, U.S. EPA.
Non-polar organic compounds, including n-alkanes,
polycyclic aromatic hydrocarbons, (PAHs), hopanes, and
steranes, in PM2.5 were measured in a long-term
epidemiological study in Tong Liang, China where coal
combustion was a major contributing source to the
aerosol. A coal-fired power station in Tong Liang was
operated in winter and spring (November to May) but
permanently shutdown in March 2003. Integrated 72hour samples were collected at three sites from 3/2/2002
to 2/26/2003 and 3/1/2005 to 2/26/2006 to evaluate the
influences by the operations of the coal-fired power
station. In-injection port thermal desorption/mass
spectrometry (TD-GC/MS) was applied for the organic
speciation. At the site closest to the coal-fired power
station, compared with the same period when it was
operated, the concentrations of priority PAHs with
molecular weight ranging from 252 to 302 decreased
14.2% to 36.9% after the station shutdown. Picene, a
known organic tracer for coal burning, was also shown a
decrease of 11.7%. No consistent variations were
determined for other low molecular weight (<228) PAHs.
The concentrations of high molecular weight n-alkanes
(C<27) were 8.8% to 33.7% lower than those measured
before the closure of coal-fired power station. Hopanes,
molecular markers of aerosol emissions from fossil fuel
utilization, were shown an averagely decrease of 4.8% but
no significant changes were determined for steranes. The
concentrations of the organic compounds at other two
sites closed to urban areas did not demonstrate influences
from the coat-fired power station shutdown. Residential
coal combustion and vehicle emissions were major
contributions for the sites.
The Detroit Exposure and Aerosol Research Study is a major
three-year study conducted by the U.S. Environmental
Protection Agency (EPA). Its primary objective is to investigate
the relationship of select air pollutant concentrations and their
sources measured at community air monitoring stations in
comparison to those measured in various neighborhoods in
Wayne County, Michigan. To accomplish this, residential
indoor and outdoor samples were collected daily at 40 locations
for 6 seasons at a flow rate of 10 liters/minute.
Analysis of samples for organic species by gas chromatography/
mass spectrometry (GC/MS) provides valuable data for
determining source contributions from several important sources
of particulate matter, including motor vehicle exhaust, diesel
exhaust, wood burning, and meat cooking. Application of
GCMS analysis to daily low volume samples presents a
considerable analytical challenge because of the large amounts
of material needed for analysis. One approach to addressing this
problem is analysis by selective ion monitoring to achieve
substantially lower limits of detection and quantitation. As a
part of our research effort to reduce uncertainty in source
apportionment, we will report method detection limits for
selective ion monitoring of organic markers of well under 100
pg/m3 for low volume 24-hour samples. These levels are
sufficient for detection of all compounds in most samples
collected. We will also report our initial demonstration of
method proficiency, which thoroughly characterizes method
detection limits, method precision, and recoveries for organic
markers. We will also show results from the first year of
sampling that indicate concentrations of all markers are present
at concentrations well above detection limits, and markers
known to be unique and to originate from the same source
exhibit measurably stronger associations at levels well above
detection limits.
Results to the first year of DEARS samples collected at the
community monitoring station indicate that particulate matter
from motor vehicle exhaust is subject to extensive day to day
variability. Associations among different hopanes, markers for
motor vehicle exhaust, and among different polycyclic aromatic
hydrocarbons, markers of general combustion processes, were
significantly stronger than associations between hopanes and
polycyclic aromatic hydrocarbons, suggesting that other
combustion sources besides motor vehicles may be important
contributors to particulate mass.
Although this work was reviewed by EPA and approved for
publication, it may not reflect official EPA policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2M.11
Investigation of Sources of OC and EC at Rural Sites in the
Northeast US Using Highly Time-Resolved Data. GEORGE
ALLEN, Iyad Kheirbek, John Graham, Gary Kleiman,
NESCAUM; Jeff Emery, ME-DEP.
Highly time-resolved speciated aerosol measurements
combined with back trajectory analysis can provide
enhanced insight into aerosol source attribution. In this
presentation, we analyze 2-hour EC and OC data from the
Sunset Model 3 field carbon analyzer at two sites in the
Northeast US: Acadia National Park in ME, and Mohawk
Mt. in CT. These are ongoing intensive rural sites run
under the MANE-VU Haze RPO RAIN program. Initial
analysis of 10 years of summer IMPROVE EC and OC
data from Acadia using incremental probability plots
showed the source region on high OC days to be from the
WNW for both OC and EC, suggesting the source is longrange transport of wildfire smoke (http://www.nescaum.
org/documents/2006-05-memo8-rain.pdf/).
We expand this analytical approach in this analysis to
include all seasons and a second site, and to include both
third-day 24-h IMPROVE and the two-hour resolution
carbon data from the Sunset method. ATAD hysplit back
trajectories for each two-hour period are calculated using
EDAS meteorological inputs for 2005 and 2006. The
objective of this analysis is to determine to what extent
we can enhance this source attribution approach using
larger and more highly time-resolved data. For example,
at Mohawk Mt. in NW CT, we would expect to see a
range of carbon source signatures based on back
trajectories, including both wildfire and urban mobile
sources (from the New Haven and New York
metropolitan areas for example).
2N.1
Application of Anion Exchange Chromatography with
Pulsed Amperometric Detection for Measurement of
Levoglucosan in Ambient Aerosol Samples. AMANDA S.
HOLDEN, Amy P. Sullivan, Sonia Kreidenweis, Jeffrey L.
Collett, Jr., Colorado State University; Bret Schichtel, William
Malm, National Park Service/CIRA, Colorado State University;
Graham Bench, Lawrence Livermore National Laboratory.
Six day integrated fine particle samples were collected
during winter and summer seasons at 12 IMPROVE sites
throughout the United States using Hi-Vol samplers. The
monitoring sites included urban, near-urban, and rural
14
locations. Measurements of C/C were made by
accelerator mass spectrometry at Lawrence Livermore
National Laboratory and presented by Schichtel et al.
[2007]. Results of these analyses indicate a prevalence of
modern carbon at many of the measurement locations. In
order to examine the aerosol fraction associated with
biomass combustion, either from residential wood
burning or wildland fires, we measured concentrations of
levoglucosan in selected samples. Levoglucosan, a
breakdown product of cellulose, is a widely used tracer
for biomass combustion. Measurements were made using
a new approach involving aqueous filter extraction
followed by direct analysis of levoglucosan and other
carbohydrates using High Performance Anion Exchange
Chromatography. In this method carbohydrates are
separated on a Dionex Carbopac PA-10 column, using a
sodium hydroxide gradient elution, and detected using
pulsed amperometry. A summary of measured
levoglucosan concentrations will be provided. Using
published source profiles, we will estimate the organic
carbon fraction associated with biomass combustion and
compare findings to fractions of modern carbon
determined using the carbon isotope approach.
Bret A. Schichtel, William C. Malm, Graham Bench,
Stewart Fallon, Charles E. McDade, Judy C. Chow
(2007). Fossil and Contemporary Fine Carbon Fractions
at 12 Rural and Urban Sites in the United States, J.
Geophys. Res., in review.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2N.2
Identification of Organic Compounds in Aerosols using
GCxGC TOF-MS. AMY LEITHEAD, Shao-Meng Li,
Douglas Lane, Yu Cheng, Environment Canada.
GCxGC TOF-MS is a powerful tool for the identification
of organic compounds in aerosol samples. Compounds
that are difficult to analyze by traditional methods can be
easily separated and identified using this instrument. The
dual columns, cryofocusing on the 2nd column along with
the highly sensitive detector and powerful de-convolution
software easily separate and identify compounds that are
impossible to analyze by traditional GC-MS. Details of
the method development work will be shown for different
classes of compounds. Methods were tested for both nonpolar compounds such as the alkanes and more polar
compounds such as the ketones, amides, fatty acids and
sugars. To optimize the separation of each class of
compounds, three different column combinations were
tested.
Preliminary results of aerosol samples collected by Hivol at 2 sites, a rural site and a forest site will be
presented. The samples were extracted by ASE and
separated by silica gel chromatography into fractions of
increasing polarity. Even after separation, a large number
of unique compounds were found in each fraction.
Unknown compounds in the aerosol samples will be
identified. This large volume of data provides valuable
information that is difficult to obtain by other methods.
2N.3
A Quantitative Protocol for Highly Polar Organic
Compounds in PM2.5 from the New York City Airshed.
HARMONIE HAWLEY, Min Li, Monica A. Mazurek,Rutgers
University.
This project focuses on the quantitation of highly polar
organic compounds extracted from PM2.5 samples
collected as part of the Speciation of Organics for
Apportionment of PM-2.5 in the New York City Area
(SOAP). The SOAP network operated from May 2002 to
May 2003 at four sites: Queens, NYC (high density urban
residential); Elizabeth, NJ (adjacent to the NJ Turnpike);
Westport, CT (downwind NYC); and a regional
background site in Chester, NJ (upwind NYC).
A key science question was how much of the
carbonaceous PM2.5 is primary versus secondary. To
address this question, a group of highly polar, low
molecular weight organic acids were selected as
secondary species. These compound do not appear as
significant species in the chemical profiles of known
primary sources such as motor vehicle exhaust and
cooking emissions. The target compounds selected were
diacids (oxalic acid and malonic acid) and oxo-carboxylic
acids (glyoxylic acid, pyruvic acid, 2-oxobutanoic acid,
levulinic acid, 5-oxohexanoic acid and oxobutanedioc
acid). A quantitative extraction and gas chromatographic/
mass spectrometric (GC/MS) chemical analysis procedure
was developed and evaluated. Trimethylsilyl (TMS)
derivatives were prepared prior to GC/MS analysis and 5point calibrations and multiple replicates were evaluated
to determine method precision.
Good precision and sensitivity were seen based on the
statistical analysis of the standard compounds evaluated
and in the ambient PM2.5 samples. This procedure
establishes a new identification and quantitation method
for determining highly polar and oxidized secondary
molecular markers in regional fine aerosols samples.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2N.4
Extractability and Determination of Different Polarity
Organic Species in Air Particulate Matter. Tylor J. Lahren,
JOSEF BERANEK, Irina Smoliakova, Steven B. Hawthorne,
Alena Kubatova, University of North Dakota; Artur Braun,
Empa Swtizerland.
Air pollution studies involving characterization of organic
particulate matter (PM) typically determine 20 - 50 wt%
of total organic carbon (OC). This is due to limitations in
extractability by organic solvents and/or elutability
through GC typically used for PM analysis. Therefore, we
have evaluated the extractability using hot pressurized
water and sequential organic solvent (Soxhlet) extraction
from three PM samples of different origin, e.g., diesel
exhaust PM (SRM 2975), urban PM (SRM 1648), and
wood smoke PM. The advantage of sequential extraction
is its ability to differentiate different polarity fractions and
thus estimate the importance of primary (non-polar) vs.
secondary (polar) emissions.
Carbon contents in solid samples and extracts were
evaluated using a thermal optical organic/elemental
carbon (OC/EC) analyzer in transmittance and reflectance
modes and a total organic carbon (TOC) analyzer. The
extraction yields of total OC (40 - 70%) were comparable
for both extraction techniques. Both methods showed
capability to separate organics based on their polarity.
Highly polar species (e.g., water-soluble) represented
significant portions of wood smoke and urban PM (7 and
20%, respectively). Although water-soluble species were
not observed, a significant polar fraction (20 - 30%) was
extracted with higher polarity solvents (e.g., 50 - 150
degrees C pressurized water or methanol) from diesel
exhaust PM.
The distribution of polar organic compounds with
hydroxyl, carbonyl and carboxyl functional groups over
different polarity fractions was confirmed with carbon 1s
near-edge X-ray absorption fine structure (C 1s
NEXAFS) and H1-nuclear magnetic resonance (H1NMR). Furthermore, using gas chromatography with
mass spectrometry carboxylic acids, sugars and aldehydes
were determined in all studied PM samples. These
chemicals were predominantly found in polar fractions of
extracts, which correlated with C 1s NEXAFS and H1NMR analysis.
2N.5
Characterization of Sugars in Fine Particles Collected at
Three Rural and Urban Sites in Texas. YULING JIA,
Shagun Bhat, Matthew Fraser, Rice University.
A total of 174 aerosol samples (<2.5 Вµm) were collected
at two rural sites (San Augustine and Clarksville) and one
urban site (Dallas) in Texas from November 2005 to July
2006 for quantification of sugars. Two extraction methods
were applied to identical samples, and the combination of
3-15 ml aliquots of dichloromethane and 3-15 ml of
methanol was found to extract sugars with greater
efficiency (>75%) than other solvent suites.
Concentrations of total aerosol sugars ranged from 22 to
3
164 ng/m . Levoglucosan, glucose, mannitol, arabitol and
glycerol were the dominant saccharides at all the sites.
Levoglucosan contributed most to the total sugars, and its
concentration was significantly enhanced during local
wildfire periods (Nov 2005 – Apr 2006). Glucose was the
second most abundant sugar (5% to 39%), and was most
prevalent in the growing season, while sucrose was
significant only in spring. In contrast, trehalose and sugar
polyols (except glycerol) increased concurrently in
summer and early autumn during the period of leaf
senescence. A Pearson’s test supported this observation
by showing a strong correlation among the concentrations
of trehalose and major sugar polyols during the period of
study.
The seasonal variations in the sugar composition in the
aerosol samples, concurrent with the seasonal change of
sugar production and utilization by plants and microbiota
in the ecosystem, indicate a strong contribution from
biogenic sources. Total concentrations of sugars were
3
generally higher at the two rural sites (93 ng/m , on
3
average) than at the urban site (55 ng/m , on average).
The seasonal trend of sugar composition was significantly
weaker at the urban site compared to the rural sites. One
possible explanation is a smaller influence of emissions
on sugars in aerosols from local sources at the urban site.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2N.6
Organic Functional Group Composition of Atmospheric
Aerosol During MILAGRO 2006 on the NCAR C130.
STEFANIA GILARDONI, Lynn M.Russell, Satoshi Takahama,
Grag C. Roberts, Scripps Institution of Oceanography,
University of California San Diego; Jose L. Jimenez, Peter F.
DeCarlo, University of Colorado.
Atmospheric aerosol samples were collected during the
Megacity Initiative: Local and Global Research
Observation (MILAGRO) campaign in March 2006 on
board the NCAR C130 to characterize the chemical
composition of the aerosol particles and to investigate the
transformation of the organic functional group properties.
Twelve flights were performed over central Mexico and
the Gulf of Mexico. The organic fraction of submicron
aerosol was analyzed by Fourier Transform Infrared
Spectroscopy (FTIR) and organic hydroxyl O-H, aromatic
C=C-H, unsaturated aliphatic C=C-H, saturated aliphatic
C-C-H, and carbonyl C=O groups were identified and
quantified. FTIR response was calibrated for a set of
standard organic compounds. FTIR absorption signals for
the different functional groups varied linearly with the
concentration of the standards. The functional group
composition was used to determine the concentration of
organic carbon and organic mass. The organic mass
concentration calculated by FTIR data agreed with the
organic mass measured simultaneously by an Aerosol
Mass Spectrometry (AMS). Single particle analysis was
performed on a few aerosol samples by Soft Transmission
X-Ray Microscopy (STXM). The carbon edge structure
showed the presence of saturated aliphatic, unsaturated
aliphatic, aromatic, and carbonyl groups. The detailed
airborne chemical measurements (both single particle
analysis and AMS) allowed an independent assessment of
the aerosol mixing state and the ability of particles to act
as cloud condensation nuclei (CCN). The
physicochemical measurements provide insight to the
chemical evolution of the aerosols and their potential
effect on aerosol-cloud interactions.
2N.7
Rapid Analysis of PAHs in Aerosol Using Desorption
Electrospray Ionization Mass Spectrometry. Hong Chen,
Mei Li, Jinjun Lian, Yaping Zhang, XIN YANG, Jianmin Chen,
Fudan University.
Due to the high carcinogenic property, PAHs in airborne
particulate matter have been paid lots of attention and
studied extensively. The analysis of complex mixtures of
environmental pollutants such as PAHs has in the past
involved the extraction of the target compound from the
sample matrix. These conventional methods to analysis
PAHs usually demand concentration steps and additional
cleanup, which can be associated with additional labor
and extra time spent in the laboratory. The recently
developed technique of desorption electrospray ionization
(DESI) has been applied to the rapid analysis of PAHs in
ambient aerosol samples. Experiments have been
performed using a commercial Finnigan LCQ Advantage
ion-trap mass spectrometer with limited modifications.
Results from the analysis PAHs in ambient aerosol
demonstrate the ability of the DESI technique to detect
the PAHs without any pretreatment and pre-separation.
Full-scan mass spectrometry data provide preliminary
identification by molecular weight determination, while
rapid analysis using the tandem mass spectrometry (MS/
MS) mode provides fragmentation data, which provide
structural information and final identification of the PAHs
when compared to the reported spectra. The analysis of
several different PAHs simultaneously demonstrates the
high throughput and reliability of the DESI technique for
ambient aerosol. The detect limit of PAHs are estimated
down to pico gram. The detecting of several PAHs in
aerosol from different sources is also presented. The
optimization of the operating conditions is in progress.
Other compositions of aerosol will be applied for the next
step.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2N.8
Characterization of Carbonaceous Aerosols Using CCSEM An Update on Analysis Methodology. GARY S. CASUCCIO,
Traci L. Lersch, RJ Lee Group, Inc.
The characterization of carbonaceous aerosols continues
to be of great interest to the scientific community.
Although progress continues to be made in developing
analysis methodologies, the carbon wars continue.
Computer controlled scanning electron microscopy
(CCSEM) offers the potential to provide additional
insight on carbonaceous aerosols, mainly because it offers
the ability to characterize individual carbon particles
based on morphology in addition to chemistry. Thus,
naturally occurring carbon particles (e.g., spores, pollen
and vegetative detritus) can often be distinguished from
other forms of carbon such as diesel emissions and wood
burning. Information of this nature provides additional
insight in the apportionment of carbonaceous material.
However, there are numerous complexities associated
with the characterization of carbon particles using
CCSEM, the most serious of which is detecting carbon
particles on a carbon substrate.
This presentation will provide an overview of the state-ofthe-art associated with CCSEM, with a focus on
characterization of carbon particles and limitations of the
analysis for these particles. Results from recent studies
involving the characterization of carbonaceous particles
will be provided.
2N.9
Detection of Particle-Phase Polycyclic Aromatic
Hydrocarbons in Mexico City using an Aerosol Mass
Spectrometer. KATJA DZEPINA, Jose-Luis Jimenez,
University of Colorado at Boulder; Janet Arey, University of
California at Riverside; Linsey C. Marr, Virginia Tech; Douglas
R. Worsnop, Timothy B. Onasch, Aerodyne Research, Inc.;
Dara Salcedo, Universidad Aut
We present the quantification of ambient particle-bound
polycyclic aromatic hydrocarbons (PAHs) for the first
time using a real-time aerosol mass spectrometer
[Dzepina et al., 2007]. The measurements were carried
out during the Mexico City MCMA-2003 field study. For
the first time two different fast, real-time methods were
used to quantify PAHs alongside traditional filter-based
measurements in an extended field campaign [Marr et al.,
2005]. This poster focuses on the technical aspects of
PAH detection in ambient air with the Aerodyne AMS, on
the comparison of AMS PAH measurements to those
measured with the other two techniques, and gives some
ambient results.
Slowik et al. [2004] demonstrated the capability of the
AMS to detect PAHs in/on particles generated in a
laboratory propane flame. Based on laboratory
experiments with eight PAH standards, we show that their
Q-AMS spectra are very similar to those in the NIST
database and that PAH molecular ions are often the
largest peak in Q-AMS spectra. We have developed a
subtraction method that allows the removal of the
contribution from non-PAH organics to the ion signals of
the PAHs in ambient data. We report the mass
concentrations of all individual groups of PAHs and the
total PAH mass concentration. MCMA-2003 PAH time
series of the Photoelectric Aerosol Sensor and Q-AMS are
well correlated. Comparison of Q-AMS PAH
measurements and GC - MS analysis of filter samples
shows agreement within the uncertainties for several
groups of PAHs, while the Q-AMS measurements are
larger for several others. In the ambient Q-AMS
measurements the presence of ions tentatively attributed
to cyclopenta[cd]pyrene and dicyclopentapyrenes causes
signals at m/z 226 and 250 significantly stronger than the
signals in GC - MS analysis of filter samples, suggesting
the presence of very labile, but likely toxic, PAHs in the
MCMA atmosphere that decayed rapidly due to reaction
during filter sampling.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2O.1
Comparison of Several Secondary Organic Aerosol (SOA)
Models for a Mexico City case study of April 9, 2003.
KATJA DZEPINA, Ingrid Ulbrich, Jose-Luis Jimenez,
University of Colorado at Boulder; Pierre Tulet, Meteo France /
CNRM-GREI; Robert J. Griffin, University of New Hampshire;
Rainer Volkamer, University of California at San Diego; Julia
Lee Taylor, Sasha Madronich, National Center for Atmospheric
Research; Bernard Aumont, Marie Camredon, Universit
Particle-phase organic species in the atmosphere can be
divided into primary and secondary organic aerosols
(POA and SOA, respectively). While POA is emitted
directly into the atmosphere, SOA is produced by
chemical reactions of gaseous organic precursors. Recent
field studies have found large discrepancies in the
measured vs. modeled SOA mass loadings in both urban
and regional polluted atmospheres [Volkamer et al., 2006
and references therein]. The reasons for the large
differences in the measured vs. modeled SOA mass
concentrations are unclear.
In this study, measurements of oxygenated organic
aerosols (OOA) estimated with the multiple component
analysis technique (MCA, Zhang et al., 2007) and
positive matrix factorization technique (PMF, Ulbrich et
al., 2006) are used as a surrogate for SOA and compared
to model results. We focus on a case study from the
Mexico City Metropolitan Area (MCMA-2003) field
campaign when fairly comprehensive measurements of
gaseous and aerosol species were performed [e.g. Salcedo
et al., 2006].
Volkamer et al. [2006] calculated SOA production for this
case study using the measured OH and VOC precursors
and published SOA yields [Koo et al., 2003], and found
an underestimation of the measured SOA by a factor of 8.
Here, we look at the same case study with several
additional SOA models and mechanisms: CACMMPMPO/ORILAM, the NCAR/U.Paris Self Generating
Master Mechanism (SGMM), the oxidation of
semivolatile and intermediate volatility POA proposed by
Robison et al. [2007] and the irreversible uptake of
glyoxal from the gas-phase. The results from the different
models and their sensitivities to important parameters (gas
phase chemistry, pre-existing organic aerosols
concentrations, activity coefficients, temperature,
enthalpy of vaporization) are presented, and their
implications for future studies are discussed.
2O.2
Validation of Soot Aging Models with Particle-Resolved
Simulations. NICOLE RIEMER, Stony Brook University;
Matthew West, Stanford University; Rahul Zaveri, Richard C.
Easter, James C. Barnard, Pacific Northwest National
Laboratory.
To assess the chemical reactivity, cloud condensation
nuclei activity, radiative properties and health impacts of
aerosol particles, the understanding of their mixing state
and the aerosol aging process is of crucial importance.
However, tracking the mixing state in aerosol models
would require treating a multidimensional size
distribution, which is computationally prohibitive.
Therefore current models adopt certain simplifications,
which usually translate into the assumption of an internal
mixture within one mode or size section. The
uncertainties associated with this assumption, which
artificially ages freshly emitted particles instantly, are not
well quantified.
A prominent example where the mixing state of aerosols
is essential to assess their impact in a meaningful way is
the aging process of soot. Many ad-hoc sectional and
modal aerosol models have been proposed to treat soot
aging, but a lack of data has lead to a lack of validation
and hence great uncertainty in the model results.
In this study, we present a new approach, the particle
resolved model PartMC. With PartMC, we explicitly
resolve and track the evolution of individual particles as
they undergo transformations by coagulation and
condensation in the atmosphere. We achieve this by
interleafing a stochastic Monte Carlo algorithm to treat
coagulation with the deterministic treatment of
condensation. In this sense, PartMC serves as a
benchmark model capability to provide baseline
simulations for validating modal and sectional models and
for improving the current aging parameterizations by
providing a mechanistic foundation for the aging process.
For the purposes of this study, we use PartMC coupled to
a 1D version of the meteorological model KAMM/
DRAIS. We will show first results of how the soot aging
time scales derived with PartMC compare to the aging
time scales derived with an approximate model approach,
the modal model MADEsoot.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
2O.3
Tracking organic particulate matter in Europe with the
Polyphemus system. EDOUARD DEBRY, Teaching and
Research Center on Atmospheric Environment (CEREA, ENPC
& EdF). Christian Seigneur, Atmospheric and Environmental
Research (AER), Inc.
Among atmospheric particle components, organics are by
far the least understood. Several hundreds of individual
chemical components contribute to the aerosol
composition, which may constitute up to 30% of PM10
mass (on an average annual basis) over Europe. The
complexity of organic aerosols, which is due in part to a
diversity of contributing sources, variable atmospheric
lifetimes and complex chemical reactions/condensation
pathways, has made their simulation in current 3D
chemical transport models difficult because such models
require lumped representations of chemical species due to
computational time constraints.
In this communication, we will present the new organic
aerosol model developed at the Teaching and Research
Center on Atmospheric Environment (CEREA). This
organic aerosol model is based on the SIREAM aerosol
module (Debry et al., Atmos. Chem. Phys., 7, 1537-1547,
2007) and the AEC organic multi-phase model (Pun et al.,
J. Geophys. Res., 107, 4333-4347, 20022002, JGR), with
additional treatments for oligomerization and aerosol
formation from diesel exhausts.
This aerosol model is incorporated into the Polyphemus
air quality modeling system (Mallet et al., Atmos. Chem.
Phys. Discuss. 2006) and used to simulate PM over
Europe. A comparison with available data shows that this
new PM module produces significantly more organic
aerosol mass than the previous model aerosol module
(SORGAM) and is in better agreement with the annual
average measured organic particle mass.
2O.4
Composition Effects on Secondary Organic Aerosol (SOA)
Partitioning: CMAQ simulations of the southeastern U.S.
Xinlian Chang, Vanderbilt University; FRANK BOWMAN,
University of North Dakota.
Composition effects on SOA partitioning have not been
considered in most air quality models because of limited
SOA component information, uncertainties in component
interactions, and computational complexity, In this study,
the aerosol module in CMAQ was modified to account for
composition effects on SOA partitioning. An updated set
of lumped SOA products with new partitioning
parameters is used to represent the major identified SOA
components corresponding to each precursor. Chemical
structure information for semivolatile products has been
added and UNIFAC is used to calculate organic aerosol
phase activity coefficients. The updated CMAQ was used
to simulate aerosol formation in the southeastern U.S.
during July 3-17, 1999. Simulation results from the
updated CMAQ are evaluated against field measurements
from SOS99, IMPROVE, and SEARCH sites and are also
compared with predictions from CMAQv4.4.
Model predictions were found to depend strongly on the
assumed POA composition. POA modeled as polar wood
smoke produced results similar to assuming an ideal
mixture with no composition effects. Modeling POA as
less polar diesel soot, however, reduced predicted SOA
concentrations by over 50%. Significant discrepancies
exist between model predictions and ambient
measurements, with the model underpredicting fine
organic aerosol concentrations by 40-60%. Model results
were shown to be highly sensitive to the temperature
dependence of SOA component vapor pressures, water
uptake by organics, and the number of SOA components
used in the model.
Because important parameters of the organic model
remain poorly characterized, it is useful to investigate the
sensitivity of the model predictions to those parameters.
For example, we investigate the sensitivity of the model
to the enthalpy of vaporization of semi-volatile organic
compounds, which greatly influence the diurnal organic
mass variation, and to the deliquescence relative humidity
of hydrophilic organic species, which may influence in
turn the liquid water content of particles.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
4E.1
Measurements of Smoke Aerosol Size Distributions and
Refractive Indices During a Series of Laboratory Biomass
Burning Experiments. GAVIN MCMEEKING, Christian
Carrico, Ezra Levin, Sonia Kreidenweis, Jeffrey Collett, Jr.,
Colorado State University; Hans Moosmuller, Patrick Arnott,
Desert Research Institute; Cyle Wold, Wei Min Hao, United
States Forest Service, William Malm, National Park Service.
Wildland fire emissions have a large impact on air
quality, visibility, and radiative forcing on local, regional
and global scales. The size, shape, and optical properties
of particles emitted from fires control how they scatter
and absorb light, which affect visibility, radiative forcing,
and tropospheric photochemistry. We present
measurements of dry (relative humidity < 40%) aerosol
size distributions for a series of controlled laboratory
burns featuring a variety of North American fuels. The
experiments were conducted in 2006 and 2007 at the US
Forest Service's Missoula Fire Science Laboratory. A
correction method was applied to the size distributions to
account for highly aspherical particles that were emitted
by the majority of the burns. Smoke refractive index was
retrieved from size distribution measurements using a
differential mobility analyzer and an optical particle
counter. We compare optical properties calculated from
dry aerosol size distribution data and retrieved dry aerosol
refractive indices to simultaneous measurements of dry
aerosol scattering and absorption. Retrieved refractive
indices are compared to values calculated from
composition measurements. We also investigate the
impact of strongly-absorbing aerosol on the uncertainty in
optical sizing measurements. Aerosol optical properties
were a strong function of combustion phase. Burns
dominated by flaming combustion emitted more strongly
absorbing aerosols than those dominated by smoldering
combustion. We also explore the implications of our
results on efforts to estimate the visibility impact of
wildland fires and attempts to model direct radiative
forcing from biomass burning emissions.
4E.2
Diversity of Biomass Burn Aerosols Based on Fuel. Rebecca
J. Hopkins, Zi Wang, A.V. Tivanski, MARY K. GILLES,
Lawrence Berkeley National Laboratory; Kirsten Lewis, W.P.
Arnott, University of Nevada; Yury Desyaterik, Alexander
Laskin, Pacific Northwest National Laboratory.
The atmospheric radiation budget is strongly coupled
with aerosols produced during natural and anthropogenic
biomass burns. These aerosols consist of particulate
organic material, black carbon (BC or soot) and inorganic
species. As BC is strongly light absorbing and organic
carbon mostly scatters radiation, biomass burn aerosols
both scatter and absorb light. They can indirectly affect
the atmospheric radiation budget by serving as cloud
condensation nuclei and as a condensation surface for
photochemically produced organics. To estimate the
radiative contributions of these aerosols, a range of
chemical and physical properties must be determined,
including chemical composition, particle size, shape and
hygroscopicity. These properties may vary with biomass
fuel, flaming versus smoldering fires and subsequent
atmospheric processing. The laboratory combustion of
biomass fuels during the Fire Lab at Missoula Experiment
(FLAME) provided an opportunity to explore the
fundamental relationships between the chemical, physical
and optical properties as a function of fuel.
During FLAME ~20 biomass fuels were burned and the
resulting particulate matter collected for later
microspectroscopic analysis. Numerous in situ techniques
measured chemical and physical properties, which appear
to correlate with our results. Complementary
microspectroscopy techniques were used to elucidate
spatially resolved local chemical bonding, carbon-tooxygen atomic ratios, percent of sp2 hybridization
(graphitic nature), and elemental composition. These
parameters are compared directly with in situ
measurements of optical properties. The biomass
combustion products could be divided into three
categories based on chemical, physical and optical
properties. Only materials displaying a high degree of sp2
hybridization, with chemical and physical properties
characteristic of 'soot' or black carbon, exhibit single
scattering albedos and Angstrom coefficients that indicate
a high light absorbing capacity. Current work is focused
on the correlation of these spectral categories with plant
types.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
4E.3
The chemical and physical characteristics of biomass
burning particulate emissions studied at the Fire Science
Laboratory. TIMOTHY B. ONASCH, Achim Trimborn, Jesse
Kroll, Doug Worsnop, Ingrid Ulbrich, J. Alex Huffman, Jose
Jimenez, Sonia Kreidenweis, Wei Min Hao
Biomass burning emissions may be an important
contributor to secondary organic aerosol generation and
the overall PM2.5 loadings in the US. Biomass burns
emit significant amounts of primary particulate and gas
phase compounds. However, insufficient research has
focused on the volatility of these emissions, the
partitioning of emitted compounds into the particulate
phase, and the evolution of these emissions down wind.
Biomass burns at the USDA Fire Science Laboratory are
analyzed to begin to address these issues.
Controlled biomass burns were conducted in 2006 and
2007. Particulate emissions were characterized using
real-time instrumentation including two WTOFAMS
(electron impaction and Li ion attachment ionization), a
scanning mobility particle sizer, and a thermal denuder
located in front of these instruments. The high-resolution
AMSs measure the nonrefractory particulate chemistry
and size and enabled separation of the organic
composition into oxygenated and non-oygenated
components.
The fuel burned during these two studies focused on
typical western and southeastern biomass that represent
the major fire affected areas of the United States and were
combusted in several experimental formats, including
stack burns that lasted several minutes and chamber burns
that were conducted inside a large room and the resulting
smoke was allowed to reside inside the mixed chamber
for 2-12 hours.
The biomass burns typically combusted 200 grams of
material and ~80% was consumed during the flaming
stage. The resulting biomass aerosol loadings varied by a
factor of ~30 depending on the fuel, and the compositions
varied from 50-100% organic by mass. The organic
particulate emissions were dominated by the nonoxygenated components, indicating that the primary
particulate emissions were not highly oxidized. Evidence
for the repartitioning of the organic particulate mass from
the particulate phase into the gas phase as a function of
chamber dilution, transforming the particle chemistry,
was observed.
4E.4
Determination of Particle-phase Organic Compounds as
Wood Burning Tracers in a Residential Site of Germany.
MD. AYNUL BARI, Guenter Baumbach, Bertram Kuch,
Guenter Scheffknecht, Universitaet Stuttgart.
Particle-phase PM10 samples were collected with low
volume sampler from November 2005 to March 2006 at a
residential site surrounded by forests near Stuttgart in
Germany. Samples collected on pre-baked glass fibre
filters were extracted using toluene with ultrasonic bath
and analysed by gas chromatograph/mass spectrometry
(GC/MS). Deuterated standard compounds are added to
the ambient samples prior to extraction to determine
analyte recoveries in each sample. Twenty-five different
organic wood smoke tracer compounds, primarily 18
species of syringol (i.e. syringaldehyde) and guaiacol
derivatives which result from the pyrolysis of wood
lignin, retene, resin acids (pimaric, iso-pimaric,
sandarancopimaric, abietic and dehydroabietic acids) and
levoglucosan, a thermal degradation product of wood
cellulose were detected and quantified in 33 samples in
this study. The concentrations of these compounds were
compared with the fingerprints of emissions from wood
combustion carried out in test facilities. It was found that
the collected airborne particles contain different amounts
of syringaldehyde. Syringaldehyde and other syringol and
guaiacol derivatives were also found in higher amounts in
particulate emissions from hardwood burning as well as
ambient samples. Thus syringaldehyde seems to be an
ideal tracer for particulate matter from hardwood burning.
Levoglucosan was found as the most abundant organic
compounds detected in all samples. The characterization
of different wood smoke tracers allows to better assess the
contribution of residential wood smoke to the PM10
loadings in residential sites.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
4E.5
Characterizing of smoke properties from laboratory
combustion of forest fuels using an aerosol mass
spectrometer. TAEHYOUNG LEE, Jeffrey L. Collett, Sonia
M. Kredenweis, Colorado State University; Jose L. Jimenez,
Joel Kimmel, University of Colorado; Jesse H. Kroll, Timothy
B. Onasch, Achim M. Trimborn, Aerodyne Research
Incorporated; William Malm, National Park Service/CIRA; Wei
Min Hao, Cyle Wold, US Forest Service, RMRS Fire Sciences
Laboratory.
Aerosols play important roles in many biogeochemical
cycles. They also contribute to adverse health effects,
indirect and direct forcing of Earth's climate, and
visibility degradation. Carbonaceous aerosols remain
poorly understood. Biomass combustion sources make a
significant contribution to carbonaceous aerosol in many
locations and seasons. Little is known, however, about
the chemical, physical and optical properties of smoke
particles, especially those produced by wildfires and
prescribed fires. In order to better understand the
chemical properties of particles produced by combustion
of wildland fuels, a study characterizing primary smoke
emissions from laboratory combustion of vegetation fuels
was conducted at the U.S. Forest Service, Fire Sciences
Laboratory (FSL) located in Missoula, Montana. Particle
compositions were measured using an Aerodyne highresolution time-of-flight aerosol mass spectrometer (HRToF-AMS) equipped with a thermal denuder inlet. Small
quantities of various fuel types were burned in a large
combustion chamber and the smoke produced was
sampled by the AMS. A variety of fuel types from the
southeast and western U.S. were studied. We will
examine the size distribution and chemical composition of
the primary smoke particles produced by biomass
combustion as a function of fuel type and monitor
evolution of these properties over a few hours in the
chamber. Differences will also be examined for selected
fuel types between flaming and smoldering emissions.
4E.6
Dual-wavelength Photoacoustic Measurements of Light
Absorption and Scattering by Wood Smoke. KRISTIN A.
LEWIS, William P. Arnott, University of Nevada, Reno; Hans
Moosmuller, Desert Research Institute.
Quantification of light absorption by wood smoke is not
simple due to absorption by poorly-characterized organic
species. We report simultaneous measurements of aerosol
optics for a wide range of biomass smoke using the
photoacoustic method for absorption and reciprocal
nephelometry for scattering at 405 nm and 870 nm.
These first of a kind measurements were made using a
single photoacoustic instrument operating concurrently at
two wavelengths. Angstrom exponents for absorption
were found to range from 1 to 3. The measurements show
conclusively that light absorbing organic material is
present in wood smoke. Spectral properties of this
organic material, which preferentially absorbs light at
lower wavelengths, indicate that casual use of the inverse
wavelength dependence of aerosol light absorption in
remote sensing and modeling applications can introduce
errors as large as a factor of 6 in the UV and a factor of 2
at visible wavelengths. Traditional filter-based
Aethalometer measurement comparisons will be
presented along with measurements by a photoacoustic
instrument operating at a wavelength of 532 nm.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
5B.1
A Method for Smoke Marker Measurements for
Determining Air Quality Impacts of Biomass Burning.
AMY P. SULLIVAN, Amanda S. Holden, Lynn R. Mazzoleni,
Sonia M. Kreidenweis, Jeffrey L. Collett, Jr., Colorado State
University; William C. Malm, National Park Service/CIRA,
Colorado State University; Wei Min Hao, Cyle E. Wold, USDA
Forest Service, Fire Sciences Laboratory.
5B.2
Time-resolved Levoglucosan and Polar Organic Compound
Measurement for a Winter-time Episode by In-situ
Silylation TD-GCMS. Mark Meiritz, University of WisconsinMadison, Wisconsin State Laboratory of Hygiene; REBECCA J
SHEESLEY, James J Schauer, David C Snyder, University of
Wisconsin-Madison; Michael J Kleeman, Walter Ham,
University of California - Davis.
Smoke from wild and prescribed fires can have a
significant impact on airborne fine particle (PM2.5)
concentrations, leading to formation of regional haze
(visibility impairment) and affecting the earth's radiation
balance (global climate change). However, current
monitoring technology is not capable of apportioning
anthropogenic emissions such as mobile sources or other
industrial related activity from wild or prescribed fire
emissions. In order to routinely determine contributions
of wild and prescribed fires to fine particle organic carbon
concentrations there is a need for an inexpensive and
robust technique for measuring concentrations of smoke
tracer compounds. Information about the mass emission
rates (relative to PM2.5 organic carbon) of those tracers
from relevant fire types is also necessary. Therefore, a
new technique to measure levoglucosan was developed
using high-performance anion-exchange chromatography
with pulsed amperometric detection. This approach offers
numerous advantages over traditional methods, including
being less expensive and utilizing a much simpler filter
extraction procedure.
Analysis of organic tracers has recently seen major
advances including the development of the thermal
desorption gas chromatography/mass spectrometry
technique which significantly reduces the required
organic carbon loading and enables molecular marker
quantification in personal exposure samples and higher
time-resolved filter samples. However, without the
quantification of levoglucosan and related carbohydrates
and polar compounds, the TD-GCMS method is limited to
assessment of motor vehicle contribution and PAH
quantification. To address this need, an in-situ silylation
derivatization technique has been developed which
enables the direct measurement of levoglucosan, simple
carbohydrates and polyols, sterols and monoglycerides.
This technique has been applied to a set of time-resolved
samples collected over 10 days in Fresno, CA in the
winter of 2007. Atmospheric particulate matter was
collected using a 92 lpm medium volume sampler with
90mm quartz fiber filters. Four, 6 hour samples were
collected each day to give 40 total samples. Ambient
levoglucosan concentrations were very high for this
wintertime episode, while the ambient concentrations of
simple carbohydrates were quite low. The 6 hour samples
provide a unique look at the diurnal trends in
levoglucosan concentrations during the day, which
indicates the trends in biomass burning source emissions
in wintertime Fresno.
A study at the USFS Missoula Fire Science Lab was
conducted in 2006 and 2007 to collect information on
emissions of levoglucosan and water-soluble potassium,
another possible tracer. Results of our analysis of
levoglucosan, water-soluble potassium, organic and
elemental carbon in fine particles produced from open
burning of fuels characteristic of the western and
southeastern U.S. will be presented. The results have
shown that levoglucosan is highly correlated with organic
carbon, however it is a fairly low fraction of the organic
carbon and the ratio of levoglucosan to organic carbon is
fairly similar across all fuel types. Additionally, the
carbohydrate measurements appear to provide a
fingerprint of fuel types, as we have observed that there
are unique peak patterns across different types of fuels.
Elemental carbon and water-soluble potassium are not
correlated with levoglucosan.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
5B.3
Field Investigation of Sources and Processes of Organic
Aerosols with High-Resolution Aerosol Mass Spectrometry
and Positive Matrix Factorization. JOSE L. JIMENEZ, Ingrid
Ulbrich, Kenneth Docherty, Peter DeCarlo, Edward Dunlea,
Allison Aiken, Joel Kimmel, J. Alex Huffman, Donna Sueper,
University of Colorado-Boulder; Qi Zhang, SUNY-Albany;
Douglas Worsnop, Manjula Canagaratna, Aerodyne Research.
Inc.
There are large gaps in our current understanding of
organic aerosol (OA) sources and processes. The field
deployments of the Aerosol Mass Spectrometer (AMS)
combined with component analysis techniques have
resulted in the recognition of the dominance of
oxygenated organic aerosols (OOA) even in urban areas
(Zhang et al., ACP, 2005; Zhang et al., GRL, 2007). The
development of the high-resolution time-of-flight AMS
(HR-ToF-AMS) allows for increased OA chemical
characterization in ambient air with high time and size
resolution. Data from several field campaigns will be
presented, including the first ground deployment of the
HR-ToF-AMS (SOAR-1 campaign in Riverside, CA) and
the first two aircraft deployments of this instrument
(MILAGRO and INTEX-B in the NCAR C-130). OA in
Asian pollution over the US is severely underestimated by
global models. A gradient in OA oxidation (O/C) is
observed between regional and urban aerosols during
aircraft studies, and also in between mornings and
afternoons in urban areas. Real-world SOA is more
oxygenated that SOA formed in chambers and is much
less volatile than current models assume. OOA is watersoluble at high dilution but not very hygroscopic. Positive
Matrix Factorization (PMF) analysis is applied to
synthetic AMS datasets and the pitfalls of the technique
(mixing and splitting of components) are highlighted. OA
spectra derived from PMF reveal strong similarities
between the source profiles at all locations. In addition to
one or several hydrocarbon-like organic aerosol (HOA)
component at urban sites, all sites have multiple
oxygenated organic aerosol (OOA) components that
resemble spectra from chamber studies of secondary
organic aerosol (SOA) formation. Biomass burning
organic aerosols (BBOA) can also be separated in some
studies. (PMF) analysis of HR-ToF-AMS spectra
indicates that most of the organic aerosol in Riverside is
SOA, which is consistent with results from the EC tracer
technique during the same study.
5B.4
Spatial and Seasonal Variations of Secondary Organic
Tracers in the Southeastern United States. XIANG DING,
Liping Yu, Rodney Weber, Mei Zheng, Georgia Institute of
Technology; Eric Edgerton, Atmospheric Research and
Analysis, Inc.; Armistead Russell, ;Georgia Institute of
Technology.
A total of 126 PM2.5 samples collected during one-year
period (May 2004 to April 2005) at four Carbonaceous
Aerosol Characterization Experiment (CACHE) sites in
the southeastern US were analyzed for secondary organic
tracers derived from isoprene and pinene by gas
chromatography-mass spectrometry (GC-MS). The
CACHE project deploys high volume samplers at four out
of the eight Southeastern Aerosol Research and
Characterization (SEARCH) air quality monitoring sites.
The major compounds quantified in this study included 2mtheyltetrols (2-methylthreitol and 2-methylerythritol),
pinonic acid and pinic acid. It is the first time that the
spatial and seasonal variations of these important
secondary organic tracers were reported in the
southeastern US. Distinct spatial variations were observed
with all tracers exhibiting higher concentrations at the
rural site (Centreville, AL) and lower levels at the urban
sites (N. Birmingham, AL; Jefferson St. Atlanta, GA; and
Pensacola, FL). Based on monthly average, significantly
higher concentrations of 2-mtheyltetrols occurred from
June to August 2004; while the lowest levels were found
from December 2004 to April 2005. However, pinonic
acid and pinic acid did not show such a distinct seasonal
variation during the one-year period. The analysis of trace
gas and meteorological data revealed that 2-mtheyltetrols,
pinonic acid, and pinic acid were closely correlated with
temperature, solar radiation and O3. The poor correlation
between isoprene- and pinene-derived secondary organic
tracers suggested their different sources and formation
pathways or that the source strengths did not vary in a
similar fashion.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
5B.5
Temporal and Spatial Variations of Primary Organic
Carbon Sources and Biogenic SOA Impacts. BO YAN, Mei
Zheng, Amy Sullivan, Rodney Weber, Sangil Lee, Charles Evan
Cobb, Santosh Chandru, Hyeon Kook Kim, Armistead G.
Russell, Georgia Institute of Technology; Eric S. Edgerton,
Atmospheric Research & Analysis, Inc.
Fine particulate matter (PM2.5) ambient samples were collected
with Hi-Vol samplers in summer 2005 and winter 2006 in
Atlanta, GA. Three sampling sites were utilized: : a roadside
highway site (directly beside the I-75/85 connector in the
midtown Atlanta); a more typical urban site in the Georgia Tech
campus (approximately 450 meters away from the highway) and
a rural site in Yorkville, GA (impacted primarily by biogenic
emissions and regional transport). At the urban and rural sites,
12-hr sampling and 24-hr sampling were performed,
respectively.
Solvent-extractable compounds in carbonaceous aerosols were
identified and quantified by gas chromatograph/mass
spectrometry (GC/MS), including alkanes, hopanes, steranes,
polycyclic aromatic hydrocarbons (PAHs), fatty acids, and resin
acids. A few important organic tracers of biogenic secondary
organic aerosols (SOA) were also measured here such as 2methyltetrol (2-methylthreitol and 2-methylerythritol), pinonic
acid and pinic acid.
Temporal and spatial variations of primary emission source
contributions to organic carbon (OC) were apportioned using
molecular marker-based chemical mass balance (CMB-MM)
modeling. Contributions of biogenic SOA to carbonaceous
aerosols were investigated, and the correlation with unidentified
OC in CMB-MM modeling was examined. Preliminary results
indicated that seasonal variations exist for 2-methyltetrol, a
photochemical oxidation product of isoprene, and levoglucosan,
a typical tracer of wood burning.
5B.6
Source apportionment of fine organic aerosol in Mexico City
during the MILAGRO-2006 field campaign. ELIZABETH
A. STONE, David C. Snyder, Rebecca J. Sheesley, and James J.
Schauer, University of Wisconsin-Madison.
Organic carbon (OC) comprises a large fraction of fine
particulate matter (PM2.5) in Mexico City. The
carbonaceous fraction of PM2.5 was collected in urban
and peripheral Mexico City from March 17-30, 2006. It
was analyzed for OC and elemental carbon (EC) using
thermal-optical filter-based methods and water-soluble
organic carbon (WSOC). Organic compounds,
particularly molecular markers, were quantified by
soxhlet extraction with methanol and dichloromethane,
derivitization, and gas chromatography with mass
spectrometric detection (GCMS). A chemical mass
balance model (CMB) based on molecular marker species
was used to determine the relative contribution of major
sources to ambient fine organic aerosol. Motor vehicles,
including diesel and gasoline, consistently accounted for
approximately half of PM2.5 OC in the urban area and
one-third on the periphery. The daily contribution of
biomass burning to PM2.5 OC was highly variable, and
ranged from 10-50% over the two sites. The remaining
OC unapportioned to primary sources shows a high
correlation with WSOC and is likely secondary in nature.
Comparison of temporally resolved PM2.5 OC indicates
that the residence time of the urban air mass is less than
12 hours and reveals trends that suggest that the
atmosphere is photochemically active during daylight
hours. This study provides quantitative understanding of
the important sources of PM2.5 OC in Mexico City and
its surroundings during the MILAGRO-2006 field
campaign.
In summer, the average ambient 2-methyltetrol and
-3
levoglucosan concentrations are 167.6 and 26.8ng.m at the
-3
rural site and 208.9 and 61.1ng.m at the urban site. 2methyltetrol accounts for 2.6% and 2.0% of OC at the rural and
urban sites, respectively, while levoglucosan only contributes
0.4% and 0.7% of OC. In winter, the average ambient 2-3
methyltetrol is 0.77ng.m at the rural site, but close to the
detection limit in urban samples. However, levoglucosan
concentrations were very high with an average value of 217.6ng.
-3
-3
m at the rural site and 249.9ng.m at the urban site in winter.
Much higher 2-methyltetrol concentrations in summer imply
strong photochemical activities and higher isoprene emission,
which lead to accumulated secondary organic aerosol (SOA) in
PM2.5. Higher levoglucosan concentrations in winter indicate
that wood burning is a seasonally significant source to OC.
Both pinonic acid and pinic acid, condensable oxidation
products of monoterpenes, were detected in most samples from
summer and
winter.
However,
no Association
clear seasonal
was
Copyright
В© 2007
by the
American
for trend
Aerosol
Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
8D.1
Quinone Emissions from Gasoline and Diesel Motor
Vehicles. CHRIS JAKOBER, M. Judith Charles, Michael
Robert, Peter Green, Michael Kleeman, Sarah Riddle, Cort
Anastasio, University of California - Davis.
Gas- and particle-phase emissions from gasoline and
diesel vehicles operated on chassis dynamometers were
collected using annular denuders, quartz filters and PUF
substrates. Quinone species were measured using O(2,3,4,5,6-pentafluorobenzyl)hydroxylamine
derivatization in conjunction with gas chromatographymass spectrometry and high performance liquid
chromatography-mass spectrometry. Nine quinones were
observed, ranging from C6 to C16. New species identified
in motor vehicle exhaust include methyl-1,4benzoquinone, 2-methyl-1,4-naphthoquinone (MNQN),
and aceanthrenequinone. Gas-phase motor vehicle
emissions of quinones are reported for the first time. Six
gas-phase quinones were quantified with emission rates of
2-28,000 micro-grams per Liter fuel consumed. The most
abundant gas-phase quinones were 1,4-benzoquinone
(BQN) and MNQN. The gas-phase fraction was greater
than 69% of quinone mass for light-duty gasoline
emissions, and greater than 84% for heavy-duty diesel
emissions. The large gas-phase quinone emissions are
consistent with 2- and 3-ring PAHs, which have similar
vapor pressures and are predominantly observed in gasphase vehicle emissions. Eight particle-phase quinones
were observed between 2-1600 micro-grams per Liter fuel
consumed, with BQN the most abundant species followed
by 9,10-phenanthrenequinone and 1,2-naphthoquinone.
Current particulate quinone measurements agree well with
the few available previous results on a mass concentration
basis. Given the large gas-phase emissions current efforts
are focused on determining the photoreactivity of the
quinones observed. The ultimate ambient partitioning of
gas-phase quinone emissions and their atmospheric
residence time will influence associated health effects of
motor vehicle quinones as gas-phase species will be
scavenged by upper airways of the respiratory system
while particulate quinones will be efficiently transported
to distal regions of the human lung.
8D.2
Determination of Aldehydes and Carboxylic Acids in Diesel
Exhaust Particulate Matter. JOSEF BERANEK, Tylor J.
Lahren, Alena Kubatova
The polar fraction of diesel exhaust particulate matter
(PM) contains aldehydes and carboxylic acids. Besides
their contribution to cloud formation, they may play an
important role in the formation of higher molecular
weight compounds (e.g., oligomers). In addition,
aldehydes and acids exhibit harmful effects on living cells
and some of them are suspected carcinogens, even at low
concentrations.
We have optimized trace analysis methods for aldehydes
and carboxylic acids occurring in air PM. In order to
efficiently recover these compounds from the diesel
exhaust PM, we have developed an extraction method
using sequential hot pressurized water extraction and
compared these results to the traditional solvent
extraction. Identification and quantification were
accomplished using two different chromatographic-mass
spectrometric methods. Due to the complex sample
matrix, liquid chromatography mass spectrometry with
electrospray ionization encountered problems with
chromatographic separation and ion suppression and
would require additional sample purification. Gas
chromatography electron ionization-mass spectrometry
(GC-EI-MS) with prior derivatization was found to be a
more efficient and sensitive technique.
Acids were derivatized with N-Methyl-N-(trimethylsilyl)
trifluoroacetamide (MSTFA) and injected onto GC-EIMS. Among the detected compounds, dicarboxylic acids
(fumaric, succinic, adipic, glutaric, pimelic, suberic,
azelaic), linear saturated and unsaturated acids (palmitic,
stearic, oleic), acids containing hydroxyl groups
(citraconic, malic), and aromatic acids (toluic,
hydrocinnamic, phthalic) were identified and quantified.
Derivatization of aldehydes was performed with O
-2,3,4,5,6-(pentafluorobenzyl)hydroxylamine
hydrochloride (PFBHA). Sensitivity of the analysis was
further improved using solid phase micro-extraction
(SPME) of derivatized aldehydes. Preliminary
experiments showed the presence of linear saturated
aldehydes (formaldehyde, acetaldehyde, propanal,
butanal, pentanal, hexanal, heptanal) and aromatic
aldehydes (benzaldehyde, tolualdehyde). The most
effective SPME method involved derivatization followed
by SPME from liquid phase allowing for the
determination of dialdehydes. Using the optimized SPME
method we were able to detect aldehydes in
concentrations as low as 1 ppb in air PM.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
8D.3
New Chemical Tracers for Diesel Source Emission
Apportionment in Ambient Fine Particulate Matter. JEANPIERRE CHARLAND, Gianni Caravaggio, Penny MacDonald,
Tony MacPhee, Natural Resources Canada, CANMET Energy
Technology Centre-Ottawa; Lisa A. Graham, Environment
Canada.
Specific organic compounds often referred to as
molecular markers are used in the chemical mass balance
(CMB) model to apportion sources of primary organic
aerosol. The main markers used in CMB to apportion
organic carbon (OC) from motor vehicle (MV) exhaust
are hopanes and steranes, also known as petroleum
biomarkers. They are found in engine lubricating oil and
trace amounts are released during engine combustion.
Because of their inherent specificity and resistance to
biodegradation, they became tracers for transportation
source apportionment. Since both gasoline and diesel
engines utilize lubricating oil, the distribution and
abundance of hopanes and steranes relative to OC in
exhaust particulate matter (PM) cannot be used to
distinguish between these sources. These biomarkers
have high boiling points and, consequently, are in low
concentrations in middle distillate petroleum products
(diesel and No. 2 fuels) and are negligible in gasoline.
High molecular weight, pyrogenic, polynuclear aromatic
hydrocarbons (PAH) have been reported to be specific to
gasoline vehicles and may be used to distinguish between
gasoline and diesel vehicles in source apportionment.
However, there is limited knowledge on markers specific
to diesel fuel. Previously, diesel vehicles were identified
as significant contributors to ambient PM, based on
\elemental carbon/organic carbon\ (EC/OC) ratio
measurements. But with the newer generation diesel
engines with reduced emissions, EC can no longer a
unique tracer for diesel exhaust. The purpose of this study
is to find molecular markers specific to diesel fuel that
can be used to assess the contribution of diesel vehicles
exhaust to ambient PM.
8D.4
Can satellite fire detections improve the emission inventories
from forest fires in the southeastern United States? TAO
ZENG, Yuhang Wang, Georgia Institute of Technology; Yasuko
Yoshida, NASA Goddard Space Flight Center; Di Tian, Georgia
Department of Environmental Protection; Amistead G. Russell,
Georgia Institute of Technology; William R. Barnard, MACTEC
Engineering and Consulting, Inc.
Biomass burning is a major contributor to organic carbon
aerosols in the southeastern United States. Uncertainties
in the fire emission inventory are a major problem in our
ability to assess the air quality impacts of fire emissions.
A recent fire inventory for 2002 has been developed by
the Visibility Improvement - State and Tribal Association
of the Southeast (VISTAS) Program over the Southeast
with the focus on prescribed fire. We compare this
inventory to fire counts by Terra Moderate Resolution
Imaging Spectroradiometer (MODIS) and Geostationary
Operational Environmental Satellite (GOES). General
seasonal consistency is found between fire products by
two satellites. However, further comparison reveals a
geographic overlap of <50 % between the two remote
sensing products. Large discrepancies were found
between the seasonal trends of satellite-based fire
products and the bottom-up VISTAS inventory. The
VISTAS inventory shows a maximum fire activity in
spring while MODIS and GOES detected a maximum in
summer. CMAQ model results and PMF source
apportionment analysis using surface measurements point
to active spring burning not summer. Current MODIS fire
detection algorithm appears to have large false positives
in summer. Further analyses are carried out to study the
impacts of two factors, cloud fractions and canopy
coverage, on fire detection.
PM filter samples were collected from gasoline and diesel
vehicles. In parallel, samples of fresh and used enginespecific lubricating oils were also collected along with
gasoline and diesel fuel for organic speciation. All
samples were analyzed by thermal desorption (TD)-GC/
MS. Ambient air PM samples were also collected and
analyzed for the presence of these newly proposed tracers.
We will present data on bicycloparaffins ubiquitous in
crude oil, show that their detection in PM can provide
new insight into diesel emissions and demonstrate how
they can be used in source apportionment.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
8D.5
Top-down correction of 2004 black carbon emissions
inventory in the United States by inverse modeling using
CAMQ-DDM. YONGTAO HU, M. Talat Odman, Armistead
G. Russell, Georgia Institute of Technology.
Black carbon (BC) is emitted into earth lower atmosphere
mainly through fossil fuel combustion and bio-mass
combustion. BC is a significant component of fine
particulate matter in ground level atmosphere, which
associates with adverse human health. BC in atmosphere
also contribute to regional haze which impair the
atmospheric visibility and may harm growth of crop and
plantation. Moreover, BC is found influential on climate
radiation forcing budget and its influence is found highly
regional variable and may be more significant in regional
scale than in global scale. BC emissions in the United
States is estimated about 5% of the global totals (8.0 Tg)
in year 1996, as the third largest country after China
(~20%) and India (~9%).
Inverse modeling techniques reconcile the gap between
modeled and observed pollutant concentrations by multistep adjustments of emissions. Here we apply an inverse
method that using DDM-3D with the Community
Multiscale Air Quality model (CMAQ) at a 36-km
resolution in horizontal on continental North America, to
calibrate the current estimation of U.S regional BC
emissions. The full operational BC monitoring networks,
i.e. STN, IMPROVE, SEARCH and ASACA at both
urban and rural locations are utilized to constrain the BC
emissions in the U.S. for the year 2004. The difference
between the CMAQ simulations and observations, along
with the DDM-3D derived sensitivities of BC
concentrations to each individual source, are used to
estimate how much BC emissions from a specific source
should be adjusted to optimize the CMAQ BC
performance through ridge regression. A set of scaling
factors are calculated for five source categories, i.e. onroad mobile, non-road mobile, fire, woodfuel combustion
and \other\ sources, from five RPO regions, for each
month of 2004. A posteriori BC emissions inventory is
then developed by implementing those scaling factors
obtained from the optimization.
8E.1
On-Line and Off-line Product Studies From Biogenic and
Anthropogenic Aerosol Precursors Under High, Low, UltraLow, and No NOx Conditions. QUENTIN G. J. MALLOY, Qi
Li, Bethany A. Warren, David R. Cocker III, University of
California-Riverisde and CE-CERT; Hiroyuki Hagino, Japan
Automobile Research Institute; Wentai Luo, James F. Pankow,
Oregon Health and Science University.
Recent studies have found that hydrocarbon to NOx ratios
have a large effect on aerosol formation rates from
aromatic precursors (Johnson et al 2005, Song et al 2005).
Possible reasons for this effect range from aldol
condensation reactions to condensed phase reactions
between organic hydroperoxides and HO2, but have yet to
be identified. Experiments conducted in the UCRiverside/CE-CERT environmental chamber facility have
identified this NOx effect extends beyond the aromatic
precursors to biogenic systems as well. In this study we
examine product compositions from multiple systems
including alpha-pinene, beta-pinene, d-limonene, 1,3,5trimethylbezene, benzene, and toluene with an emphasis
on development of the differences in aerosol mechanisms
at high, low, ultra-low, and no NOx conditions.
Particulate and gas phase composition was monitored by
on-line high resolution aerosol mass spectrometry and
proton transfer mass spectrometry respectively. Off-line
analysis was performed by liquid chromatography-high
resolution time of flight mass spectrometry and twodimensional gas chromatography-time of flight mass
spectrometry.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
8E.2
Recent Results in Molecular Speciation of Secondary
Organic Aerosol. JASON D. SURRATT, Jesse H. Kroll, Shane
M. Murphy, Armin Sorooshian, Puneet S. Chhabra, Nga L. Ng,
Arthur Chan, Richard C. Flagan, John H. Seinfeld, California
Institute of Technology; Tadeusz E. Kleindienst, Edward O.
Edney, John H. Offenberg, Michael Lewandowski, U.S.
Environmental Protection Agency; Mohammed Jaoui, Alion
Science and Technology, Inc.; Magda Claeys, Yadian Gomez,
Rafal Szmigielski, Reinhilde Vermeylen, Katarzyna
Szmigielska, University of Antwerp; Willy Maenhaut, Ghent
University.
Despite recent advances in understanding secondary organic
aerosol (SOA) formation mechanisms, such as particle-phase
reactions, much uncertainty remains owing to the lack of
detailed molecular speciation data available. In this study, a
collective force of online and offline analytical techniques have
been employed to characterize the chemical compositions of
SOA formed from the photooxidation of selected biogenic (i.e.
isoprene, monoterpenes, sesquiterpenes) and aromatic (i.e. mxylene and toluene) VOC precursors under low- and high-NOx
conditions, and under various seed aerosol acidities. These
analytical techniques include the use of gas chromatography/
mass spectrometry (MS) with prior derivatization, high
performance liquid chromatography/electrospray ionization
(ESI)-MS, ESI-ion trap MS, ultra performance liquid
chromatography/ESI-high resolution MS, matrix-assisted laser
desorption ionization (MALDI)-MS, aerosol MS, and a particleinto-liquid sampler with subsequent offline analysis with ion
chromatography. By using these techniques, we have been able
to identify SOA formation mechanisms that are not only
applicable to chamber studies but also to ambient aerosol, and
may be common to most VOC precursors. For example, we
have found that particle-phase sulfate ester formation occurs
when isoprene and alpha-pinene are oxidized in the presence of
acidified seed aerosol. Many of these compounds identified in
our chamber studies have now been observed in ambient aerosol
collected from the southeastern U.S and from K-puszta,
Hungary. Surprisingly, we have not observed sulfate ester
formation from aromatic VOC precursors at this time; however,
this finding is consistent with the lack of an observed
enhancement in the SOA mass when acidified seed aerosol is
present. In addition to sulfate ester formation, we shall discuss
the chemical nature of other identified SOA products formed
from the selected biogenic and aromatic VOC precursors in
order to show similarities and differences in SOA formation
mechanisms. Preliminary data suggests the possibility that
particle-phase organic esterification may be applicable to most
VOCs studied.
8E.3
Is the Gas-Particle Partitioning in alpha-Pinene Secondary
Organic Aerosol Reversible? ANDREW GRIESHOP, Neil
Donahue, Allen Robinson, Carnegie Mellon University.
Current models treat gas-particle partitioning of
secondary organic aerosol (SOA) as a reversible process,
but this assumption has not been rigorously evaluated.
Previously, phase partitioning has been studied
quantitatively via SOA production experiments or
qualitatively by perturbing temperature and observing
particle evaporation. In this work, two methods were used
to investigate the effects of dilution on gas-particle
partitioning. Dilution allows us to experimentally travel
back down the SOA partitioning curve generated by
traditional SOA studies. Initial SOA concentrations (COA )
of 200 - 500 micro-grams m-3 were generated via
ozonolysis of alpha-pinene in a laboratory reaction
chamber; the fresh SOA was then diluted by a factor of 25
- 80.
Essentially no evaporation was observed for experiments
conducted with an external dilution sampler with a
residence time of ~30 seconds, suggesting that gasparticle partitioning is not reversible. However,
substantial evaporation was observed during in-chamber
dilution experiments in which the SOA was diluted
isothermally inside the chamber via rapid air exchange
with scrubbed outside air and then allowed to equilibrate.
These experiments show that repartitioning occurs on a
time scale of tens of minutes to hours, and that, given
sufficient time, alpha-pinene SOA does repartition
reversibly. The evaporation rate is consistent with uptake
coefficients on the order of 0.001 - 0.01 versus the
commonly assumed value of 1. While mass-transfer
limitations are a possible explanation for the slow
evaporation rate, it also could be due to decomposition of
weakly bound oligomers. Data from an aerosol mass
spectrometer (AMS) indicate that the composition of
SOA varies with partitioning. For example, larger
contribution from m/z 44, associated with oxygenated
organics such as carboxylic acids, indicates that the
aerosol is composed of more polar components at low
COA .
Disclaimer: Although this work was reviewed by EPA and
approved for publication, it may not necessarily reflect official
Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
8E.4
Secondary Organic Carbon Contributions to Ambient
PM2.5 in the Midwestern United States. MICHAEL
LEWANDOWSKI , Tad E. Kleindienst, John H. Offenberg,
Edward O. Edney, National Exposure Research Laboratory, US
EPA; Mohammed Jaoui, Alion Science and Technology;
Rebecca J. Sheesley, James J. Schauer, University of WisconsinMadison.
Secondary organic aerosol formed from biogenic and
anthropogenic hydrocarbons is believed to be a significant
contributor to ambient PM2.5 in many parts of the United
States. However, it has often proven difficult to
determine the relative contributions of specific
hydrocarbons to the total organic carbon concentrations
measured in ambient PM2.5. Recently, a series of singlehydrocarbon laboratory studies were conducted at the
EPA to measure the concentrations of a number of tracer
compounds as a fraction of the total secondary organic
carbon generated. Laboratory-based mass fractions
obtained for isoprene, alpha-pinene, beta-caryophyllene,
and toluene SOC were used to estimate the contributions
of these hydrocarbons to ambient PM2.5 collected in
Research Triangle Park, NC, USA during 2003.
In the present study, this technique has been applied to
ambient samples collected in five Midwestern US cities
throughout 2003: East St. Louis, Illinois; Detroit,
Michigan; Cincinnati, Ohio; Bondville, Illinois; and
Northbrook, Illinois. Monthly composites were analyzed
using chemical derivatization with BSTFA coupled with
GC-MS analysis in order to estimate the concentrations of
a series of organic tracer compounds. The previously
developed mass fractions were then used to estimate the
contributions of SOC from isoprene, alpha-pinene, betacaryophyllene, and toluene to total ambient organic
carbon in these five cities. Total SOC contributions from
these sources made up approximately 28% of the total
measured organic carbon in the spring, 56% in the
summer, 35% in the fall, and 16% in the winter.
Estimated SOC concentrations during the summer ranged
from 1.2 to 2.1 micrograms C per cubic meter, depending
on the sampling location.
8E.5
Comparison of Health Effects and Composition of
Secondary Organic Aerosols Formed With and Without
Sulfur Dioxide. MELANIE DOYLE, Matt Campen, JeanClare
Seagrave, Jake McDonald, Lovelace Respiratory Research
Institute; John Seinfeld, California Institute of Technology;
Annette Rohr, Eladio Knipping, EPRI.
We present initial findings from the Secondary Particulate
Health Effects Research (SPHERES) program, whose
main objective is to define the composition and resulting
relative health hazard of secondary organic aerosol (SOA)
synthesized under varying reaction conditions. These
initial findings will focus on the comparisons among
atmospheres generated with an alpha-pinene hydrocarbon
precursor. Reactions were conducted in a 11 m3
continuous flow stir reactor (irradiation chamber), which
permitted constant production of SOA for the conduct of
inhalation exposures in laboratory rodents. We will
report on the comparison of results of SOA atmospheres
produced with alpha pinene:nitrogen oxides and alphapinene:nitrogen oxides:sulfur dioxide. With both sets of
precursors, SOA atmospheres were produced to yield 200
micro-gram/m3 particulate material. Direct comparisons
of composition and toxicity were conducted with
normalization of atmospheres to either total particulate
matter or total organic particulate matter. Exposures were
conducted down-stream of honeycomb denuders
employed to remove the gas-phase precursors and
reaction products. Nose-only exposures were conducted
with either rats (pulmonary effects) or mice (pulmonary
and cardiovascular effects). Chemical composition was
optimized to ensure that organosulfur compounds formed
in the chamber (when SO2 was present) matched
compounds that have previously been reported in ambient
air. SOA had an approximate 100 nm particle size.
Further analysis, along with the comparison of pulmonary
and cardiovascular effects associated with inhalation of
these two SOA, will be reported. Work supported by the
Electric Power Research Institute.
Disclaimer: Although this work was reviewed by EPA
and approved for publication, it may not necessarily
reflect official Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
9E.1
Optical Properties and Hygroscopicity of Fresh Biomass
Aerosols Generated from Various Combustion Conditions.
CHRISTOPH RODEN, Tami Bond, University of Illinois Urbana-Champaign.
9E.2
Cloud condensation nucleus activity of secondary organic
aerosol particles mixed with sulfate. STEPHANIE KING,
Thomas Rosenoern, John Shilling, Qi Chen, Scot Martin,
Harvard University.
This work focuses on determining the factors that govern
the climate-relevant properties of aerosols emitted from
biomass combustion. Open biomass burning emits 3300
and 25,000 Gt of black carbon (BC) and organic carbon
(OC) respectively, equivalent to about 40% and 75% of
all BC and OC emissions respectively. Biofuels are
responsible for an additional 20% of BC and OC
emissions. The optical properties, size, and
hygroscopicity of these emissions are among the major
factors determining their climatic impacts. The
combustion conditions such as temperature and fuel size
determine how much and what type of carbonaceous
aerosols are produced. Low temperature smoldering of
fresh fuels can generate large amounts of organic carbon,
while hot flaming combustion in wood cookstoves often
generates a large percentage of black carbon. These
combustion conditions also control the initial
hygroscopicity of the particulate emissions.
The cloud condensation nucleus (CCN) activity of
organic-sulfate particles was investigated using a steadystate environmental chamber. The organic component
consisted of secondary organic aerosol (SOA) generated
in the dark from 24 +/- 2 ppb alpha-pinene at conditions
of 300 +/- 5 ppb ozone, 40 +/- 2% relative humidity, and
25 +/- degrees C, with the organic mass loading in the
chamber ranging from 23 to 37 micrograms per cubic
meter. CCN analysis was performed for 80- to 150-nm
particles having variable organic-sulfate volume fractions,
estimated from the diameter of the organic-sulfate particle
relative to that of the seed as well as independently from
mass spectra. Critical supersaturation, which increased for
greater SOA volume fraction and smaller particle
diameter, was well predicted by a Kohler model having
two components, one for ammonium sulfate and another
for SOA. The entire data set could be successfully
modeled by a single suite of effective chemical
parameters for SOA. These results suggest that limited
organic solubility, at least for the range of conditions
studied, may be reliably omitted in the treatment of cloud
droplet formation in global climate models.
We generated carbonaceous aerosols under a variety of
combustion conditions. We varied the type of wood fuel,
the combustion temperature, the wood surface area to
volume ratio, and fuel moisture in an effort to vary the
properties of the generated carbonaceous aerosols. We
measured the mass emission, EC and OC factions,
scattering, absorption, and size distributions of the
resulting emissions. The combustion varied from slightly
smoldering to vigorous flaming, and the resulting
emissions had a wide range of single scatter albedos
(SSA), OC/EC fractions, and absorption Angstrom
exponents.
We also used a differential mobility analyzer (DMA) to
first determine the size distribution of the fresh aerosols.
We then used a humidifying tandem differential mobility
analyzer (H-TDMA) to determine the growth of fresh
aerosols under different humidity conditions (from 20%
to 90% RH). We summarize the different growth rates
and optical properties observed and identify key
differences between combustion conditions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
9E.3
Characterizing the CCN characteristics and Droplet Growth
Kinetics of Ageing Secondary Organic Aerosol from Betacaryophyllene. AKUA ASA-AWUKU, Athanasios Nenes,
Georgia Institute of Technology; Gabriella Engelhart, Byong
Hyoek Lee, Spyros Pandis, Carnegie Mellon University.
Aerosols can affect climate directly via scattering or
indirectly through their interactions with clouds. Organic
aerosols constitute a significant mass fraction of aerosols
and are known to be viable precursors for cloud droplet
formation. However, there is limited information on the
thermodynamic properties and droplet formation kinetics
of carbonaceous CCN, especially for secondary organic
aerosol during its initial stages of formation and ageing.
Our experiments indicate that SOA volatility can impact
activation parameters and thus we explore how droplet
kinetics change with the aging of aerosol generated from
dark ozonolysis of beta-caryophyllene (a highly efficient
sesquiterpene SOA precursor). In addition we infer
thermodynamic properties and surface tension depression
characteristics of the aged aggregate organic chemical
composition from off-line filter experiments.
9E.4
Hygroscopic Growth and Cloud Condensation Nuclei
Activity and Chemical Composition of Primary Biomass
Smoke. CHRISTIAN M. CARRICO, Markus D. Petters, Sonia
M. Kreidenweis, Anthony J. Prenni, Paul J. DeMott, Gavin R.
McMeeking, Amy Sullivan, Lynn Rinehart, Jeffrey L. Collett,
Colorado State University; William Malm, U.S. National Park
Service; Cyle Wold, Wei-Min Hao, USDA/USFS Fire Sciences
Laboratory.
The Fire Lab at Missoula Experiment (FLAME) is an
ongoing collaboration investigating biomass smoke
properties important to air quality and aerosol-cloudclimate interactions. In laboratory experiments, we
compared aerosol hygroscopic growth and cloud
condensation nuclei (CCN) properties in relation to
chemical composition for fresh smoke from fuels
common to wildland fires. Particle diameter growth
factors (GF) were measured with a hygroscopic tandem
differential mobility analyzer for 5 < RH < 95%. A CCN
counter measured the relationship between particle dry
diameter and critical supersaturation at s ~ 0.3, 0.5, 0.7,
and 0.9%. Aerosol carbonaceous and ionic composition
were determined with thermal-optical and ion
chromatographic techniques, respectively. Depending on
fuel and combustion conditions, water uptake by smoke
ranged from weakly to strongly hygroscopic (1.04 < GF at
RH = 90% < 1.70). For several fuels, we observed
apparent particle shrinkage, likely due to collapsing
agglomerates. The hygroscopicity parameter, kappa,
compares HTDMA and CCN measurements on a
common scale. Prior laboratory experiments confirmed
that kappa were comparable within ~20% uncertainty for
selected organic-inorganic mixtures. During FLAME, we
observed 0.02 < kappa < 0.6 with the ratio of inorganic
ions to carbonaceous material primarily determining this
variability. Other predictors, such as the elemental carbon
fraction of total carbon, were less important. During
FLAME, we found consistent agreement between
HTDMA and CCN kappa for some cases, particularly
where the inorganic-ionic fraction was large. For most
cases, particularly those dominated by organic carbon, the
hygroscopicity inferred from GF data was considerably
less than required to explain CCN activity. One potential
contributor to this anomalous behavior is the presence of
sparingly soluble material. Our hypothesis is that smoke
must be exposed to very high humidities (<95%) before
some particulate material, likely organic, is fully
dissolved and contributes to observed CCN activity.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
9E.5
Investigation of Thermodynamic Properties, CCN Activity
and Droplet Growth Kinetics of Carbonaceous Aerosol in
Mexico City. LUZ TERESA PADRO, Chris Hennigan,Terry
Lathem, Athanasios Nenes, Rodney J. Weber, Georgia Institute
of Technology.
In this study, we characterize Mexico City aerosols
collected during the MILAGRO campaign conducted in
during March 2006. We focus on the CCN-relevant
properties of the water-soluble fraction of the
carbonaceous aerosol and their evolution as the aerosol
chemically ages. Characterization of organics was
performed by measuring CCN activity, surfactant
properties, and (WSOC) and inorganic carbon (IC)
concentration of the particles. The effect of aging is
inferred by comparing properties at two sites, one in the
city (T0 site) and one downwind of it (T1 site). In order to
perform the organic characterization, particles were
collected on HiVol filters, subsequently extracted in water
and sonicated with heat to extract the water soluble
component of the aerosols. The samples are then
atomized to produce aerosol, the CCN properties of which
are studied with a Streamwise Thermal Gradient CCN
Chamber (STGC) by Droplet Measurement Technologies
(DMT). Surface tension measurements were performed
with a CAM 200 Optical Contact Angle Meter by KSV
Inc. Ion Chromatography and a Total Organic Carbon
(TOC) analyzer were used to measure the ions and
WSOC concentration, respectively. Pure samples, as well
as mixtures with ammonium sulfate (50 and 90 % weight)
were studied. Kohler theory analysis (KTA) was used to
infer the molar volume of the organics by coupling CCN,
surface tension, and chemical composition measurements
with Kohler theory. We find clear evidence for chemical
ageing of the particles, where both molecular weights,
surfactant characteristics of the organics change between
the two sampling sites. Remarkably, these changes are
consistent for the whole sampling period, suggesting that
a first-order parameterization of the process is possible.
9E.6
Water-Aerosol Interactions Downwind of Mexico City:
Inferences about Mixing State, Droplet Growth Kinetics and
Aging of Ambient Aerosol. SARA LANCE, Luz Padro,
Athanasios Nenes, Georgia Institute of Technology; Eben Cross,
Boston College; Tim Onasch, Douglas Worsnop, Aerodyne
Research Inc; Xiao-Ying Yu, Lizabeth Alexander, Pacific
Northwest National Laboratory; James N. Smith, National
Center for Atmospheric Research.
We describe observations of size-resolved cloud droplet
activation of partially aged Mexico City aerosol obtained
during the MIRAGE campaign. These measurements
provide unique insight into the integrated chemical
properties and mixing state of the aerosol population, in
addition to the water uptake kinetics of the particles as
they grow into droplets, all of which are crucial
constraints for understanding aerosol-cloud-climate
interactions and chemical aging of polluted aerosol.
A DMT Cloud Condensation Nucleus counter (CCNc)
was operated in parallel to a particle counter with an
upstream Differential Mobility Analyzer (DMA) to obtain
the size-resolved activation fraction for a given water
vapor supersaturation (SS). Activation spectra as a
function of SS were obtained for the period of March 16
-31, 2006. From the activation spectra, we determine the
fraction of particles that act as CCN and the mean soluble
mole fraction for these particles. We also estimate the
variability in soluble mole fraction for these particles by
analyzing the slope of the activation spectra, eliminating
the effect of particle size variability inherent with the use
of a DMA. We then perform \chemical closure\ using the
Aerosol Mass Spectrometer (AMS) dataset.
The AMS data shows a large fraction of organic
constituents, which may alter the cloud droplet growth
kinetics. We monitor the droplet size distribution at the
exit of the CCNc column, and, given the controlled
conditions within the CCNc instrument, we can compare
the observed droplet growth with the expected growth for
classified, pure ammonium sulfate particles of the same
critical supersaturation. A numerical model is used to
parameterize the droplet growth kinetics in terms of a
water vapor uptake coefficient.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11K.1
Absorption Enhancement of Sulfate-Coated Black-Dyed
PSL Particles. JEONGHOON LEE, Arthur J. Sedlacek III,
Brookhaven National Laboratory.
Black carbon (BC) typically ages through the
development of a nonabsorbing transparent coating that
can start develop within hours of BC generation. Recent
numerical analyses show that light absorption by aged BC
could be 1.5 times greater than that of fresh aerosol.
(Bond et al., 2006) However, experimental verification of
this is difficult because two opposing processes are
present: (i) collapse of the fractal structure of the nascent
soot that causes a decrease in the mass absorption
coefficient and (ii) the growth of a transparent coating
that increases the absorption coefficient. In order to help
elucidate and to quantify these effects we have started to
carry out experiments using a model system that enables
us to separately study the coating effect on absorption
without the complication of fractal collapse. By utilizing
commercially available black dyed Polystyrene Latex
(PSL) particles (Dp=200 nm and up) experiments are
carried out where a transparent suflate coating can be
grown on these particles and the change in the absorption
coefficient measured as a function of coating thickness
using Photothermal Interferometry [PTI]. Recently,
photothermal interferometric technique has been
successfully applied to the direct measurement of ambient
aerosol absorption without interference from scattering.
(Sedlacek and Lee, 2007; Sedlacek, 2006) The ability to
conduct aerosol absorption measurements in situ make the
PTI technique ideal for quantifying the degree of
absorption enhancement caused by the sulfuric coating.
A discussion of the PTI technique, along with the results
of an intercomparison with a PSAP and the
aforementioned coating experiments will be presented.
Bond, T. C., Habib, G. and Bergstrom, R. W., Limitations
in the Enhancement of Visible Light Absorption Due To
Mixing State, J. of Geo. Res., 111, D20211, 2006
Sedlacek, A. J., Realtime Detection of Ambient Aerosols
using Photothermal Interferometry: Folded Jamin
Interferometer, Rev. Sci. Instrum. 77 064903 (2006).
11K.2
Formation of highly hygroscopic soot aerosols by
atmospheric processing with sulfuric acid vapor. ALEXEI
KHALIZOV, Renyi Zhang, Dan Zhang, Huaxin Xue,Texas
A&M University; Joakim Pagels, Peter H. McMurry, University
of Minnesota; Jianmin Chen, Fudan University.
Carbon soot is produced from a variety of anthropogenic
and biogenic sources. Once emitted into the atmosphere,
soot particles are subjected to several aging processes,
including condensation of low-volatile gaseous species,
coagulation with existing aerosols, and surface oxidation.
The changes in morphology, hygroscopicity and optical
properties during the aging likely alter the atmospheric
effects of soot aerosols, including interference with solar
radiative transfer and alteration of cloud formation. For
example, coating of hydrophobic freshly emitted soot
particles by water-soluble material can make them
hydrophilic. Hence, aged soot aerosols can potentially
serve as an efficient source of cloud condensation nuclei
(CCN), affecting the radiative balance indirectly - by
changing the cloud formation mechanism. Absorption of
the visible solar radiation by soot, which directly affects
the earth-atmosphere radiative balance, can be further
amplified by transparent sulfate and organic coatings
developed on particles upon atmospheric processing. We
have investigated the effect of atmospheric processing on
mixing state and hygroscopicity. Airborne soot particles
exposed to sub-ppb (part per billion) concentrations of
sulfuric acid vapor exhibited significant changes in their
morphology and became highly hygroscopic as measured
by tandem differential mobility analyzer (TDMA) and
differential mobility analyzer - aerosol particle mass
(DMA-APM) techniques. Critical supersaturations
measured directly and estimated from hygroscopic
growth-based Kohler curves for H2SO4-coated soot
particles were in the range of 0.1-0.5%. Thus, sulfuric
acid processed soot can readily activate to cloud droplets
under typical cloud conditions. Extrapolation of our
results to atmospheric H2SO4 concentrations indicates
that internal mixing with sulfuric acid is likely a major
mechanism of soot aging with profound implications on
visibility, human health, and climate forcing.
Sedlacek, A. J. and Lee, J., Photothermal Interferometric
Aerosol Absorption Spectrometry, Aero. Sci. Tech.,
(submitted for publication, 2007).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11K.3
A Novel Optical Absorption Approach for Black Carbon
Measurement in Snow. MARTIN SHAFER, Brian Majestic,
James Schauer, University of Wisconsin-Madison.
Snow is potentially an excellent archive of aerosol
deposition history to the polar regions of the world.
Elemental/black carbon (BC) is a useful tracer of
combustion processes, and is likely to be relatively stable
in snow for periods of time relevant for the reconstruction
of climate and atmospheric aerosol histories. The BC
content of snow may also influence albedo. Particulate
(insoluble) organic carbon (POC) content is also of
interest from source reconciliation and photochemistry
angles. However, levels of BC (0.5-1 micrograms per
kilogram) and POC (5-20 micrograms per kilogram) in
pristine snow are very low and measurements quite
challenging. Artifacts (e.g. filter sorption and colloid
passage) in methods that are currently being applied may
result in significant biases. We are addressing these
issues on two fronts. In an effort to eliminate the need for
a filter altogether we are developing a direct optical
method for BC measurement in melted snow. We're
applying liquid wave guide long path (1 meter) optical
cells and low-noise spectrophotometry to capture the
absorption spectrum of suspended BC. After correction
for absorption of dissolved species, one can isolate the BC
signal. Calibrations with a variety of BC containingstandards are linear and reproducible, and detection limits
appear adequate for quantifying trace levels of BC in
snow. Comparisons between BC levels determined by
this novel optical approach and that of filter-based
methods are in-progress. In parallel work we are
examining the potential of 0.1 and 0.2 micron porosity
Anotec alumina filters as replacements for traditional
glass fiber filters. These filters are more efficient in
trapping colloids and therefore my recover more BC/EC
from snow or rain water solution. Anotec filter blanks are
comparable to those of GF filters, and filters may be used
directly in Sunset Labs Thermo-Optical EC/OC
instruments.
11K.4
Humidification Factors (f(RH)) for Fresh Biomass Smoke
from Laboratory Controlled Burns. Derek Day, JENNY
HAND, CIRA, Colorado State University; Gavin McMeeking,
Sonia Kreidenweis, Jeff Collett, Jr., Colorado State University;
Cyle Wold, Wei-Min Hao, USFS Missoula Fire Science
Laboratory; William Malm, National Park Service.
During the 2006 FLAME experiment (Fire Lab at
Missoula Experiment), a series of laboratory experiments
were performed to investigate the physico-chemical,
optical, and hygroscopic properties of fresh biomass
smoke. As part of this experiment, two nephelometers
simultaneously measured dry and humidified light
scattering coefficients (bsp(dry) and bsp(wet),
respectively) in order to explore the role of relative
humidity (RH) on the optical properties of biomass smoke
aerosols. Results from burns of several biomass fuels
showed large variability in the humidification factor (f
(RH) = bsp(wet)/bsp(dry)). For instance, values ranged
from f(RH)~1.02 at RH=80% for Ponderosa pine duff to f
(RH)~1.58 at RH=80% for southern California chamise.
We use measured chemical composition to model the
hygroscopic growth to investigate the role of inorganic
and organic compounds on water uptake for these
aerosols. Understanding the hygroscopic properties and
the compounds responsible for water uptake by biomass
smoke is very important for accurately assessing its role
in climate change studies and visibility regulatory efforts.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11K.5
Laboratory Investigation of the Photochemical Oxidation of
Organic Aerosol from Wood Fires. ANDREW GRIESHOP,
Allen Robinson, Carnegie Mellon University.
Experiments were performed to investigate the effects
of photo-oxidation on organic aerosol (OA) in dilute
wood smoke. The wood smoke was produced by burning
a mixture of hardwoods in a small wood stove then
3
injected into a 10 m Teflon smog chamber filled with
clean air; initial wood smoke OA concentrations were
-3
between 30 - 200 micro-grams m . The diluted wood
smoke was exposed to UV light to initiate photooxidation; changes in particle and gas-phase composition
were monitored continuously with an Aerosol Mass
Spectrometer (AMS), a Proton Transfer Reaction Mass
Spectrometer (PTR-MS) and a Scanning Mobility Particle
Sizer (SMPS). The data were analyzed to determine the
production of SOA, investigate the potential decay of
levoglucosan, and characterize the change of the OA mass
spectra.
The composition and properties of the fresh wood
smoke showed significant inter-experiment variability.
The spectral signature of the biomass-burning marker
levoglucosan is always present, but at varying levels.
Photochemical oxidation can produce substantial
secondary organic aerosol (SOA), increasing the OA
concentration in one experiment by a factor of 2.5 after
several hours. Lower production was observed in other
experiments. In all experiments, AMS spectra of the OA
under UV radiation indicate formation of progressively
more oxidized material. To investigate the evolution of
OA composition, AMS spectra were decomposed into two
components: a spectrum corresponding to the fresh
primary emissions and a residual spectrum which evolves
over time. In experiments with significant SOA
production, the residual spectra are very similar to those
similarly derived from photo-oxidation of diesel exhaust,
oxidized organic aerosol (OOA) spectra derived from
factor analysis of ambient data sets, and aged OA
measured in remote rural areas. This suggests that the
contribution of potentially unaccounted-for SOA
precursors emitted from biomass burning (among other
sources) may make a considerable contribution to the
ambient OA loading.
11K.6
Effect of Hydrophobic Primary Organic Aerosols on the
Yield of Secondary Organic Aerosol from Ozonolysis of
alpha-Pinene. CHEN SONG, Rahul A. Zaveri, Mikaela L.
Alexander, Pacific Northwest National Laboratory; Joel A.
Thornton, University of Washington; Sasha Madronich,
National Center for Atmospheric Research; John V. Ortega,
Alexander Laskin, Xiao-Ying Yu, Alla Zelenyuk, Matt
Newburn, David A. Maughan, Jerome Birnbaum, Pacific
Northwest National Laboratory.
Semi-empirical secondary organic aerosol (SOA) models
typically assume a well-mixed organic aerosol phase even
in the presence of hydrophobic primary organic aerosols
(POA). This significantly enhances the modeled SOA
yields as additional organic mass is available to absorb
greater amounts of oxidized secondary organic gases than
otherwise. Here we investigate the applicability of this
critical assumption by measuring SOA yields from
ozonolysis of alpha-pinene (a major biogenic SOA
precursor) in a smog chamber in the absence and presence
of dioctyl phthalate (DOP) and lubricating oil particles,
which simulate the urban hydrophobic POA. The results
show that these POA had no detectable effect on the SOA
yields, suggesting that while the SOA species do
condense onto pre-existing hydrophobic POA, they may
form a separate phase rather than a single well-mixed
organic phase with the POA species. If these results are
applied to other biogenic and anthropogenic SOA
precursors, then the available semi-empirical models will
predict even less SOA than the previous estimates, which
are already too low compared to observations in the urban
and upper tropospheric environments.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11K.7
Analysis of PM2.5 Speciation Network Carbon Blank Data.
Max Peterson, JAMES FLANAGAN, Larry Michael, and R.K.
M. Jayanty, RTI International.
PM2.5 chemical speciation involves the measurement of a
number of analytical parameters including organic carbon
(OC) and elemental carbon (EC). Air samples for OC/EC
analysis are collected on quartz fiber filters, which are
analyzed using an adaptation of NIOSH Method 5040, a
thermal-optical transmittance method. This method,
which relies on thermal desorption, oxidation of the
carbonaceous material to CO2, reduction of the CO2 to
CH4, and quantitation by FID, does not identify specific
compounds, but reports aggregate concentrations of the
OC and EC.
This paper will describe a statistical analysis of field and
laboratory blank OC/EC data for the PM2.5 Chemical
Speciation Monitoring Network, which is managed and
funded by EPA/OAQPS, and for which Research Triangle
Institute is the contract laboratory. Data from over six
years of network operation have been used for this
analysis.
Quartz filters received from the manufacturer must be
thoroughly cleaned before they can be used for sampling
OC/EC. This is accomplished by heating the filters at
900C for three hours under a slow stream of dry air.
Analysis of laboratory blank data shows that OC
background levels begin to rise immediately after
cleaning, which is consistent with adsorption of volatile
and semi-volatile organic compounds onto the freshly
treated quartz filters. In the field and trip blank data, a
strong dependency on sampler type was found in the OC
levels. This finding may be related to differences
between filter module designs for the different sampler
types. Trip and field blank data were also evaluated to
detect any dependency on site, time of year, or geographic
location. Other variables potentially affecting OC and/or
EC blank levels include impactor grease, sampler module
dead volume, and sampler flow rate.
11K.8
Observations of hygroscopic and optical properties of
biogenic secondary organic aerosol generated using a simple
continuous flow reaction chamber. Markus D. Petters,
GAVIN R MCMEEKING, Taehyoung Lee, Sonia M.
Kreidenweis, Christian M. Carrico, Jeffrey L. Collett, Jr.,
Colorado State University; Paul J. Ziemann, University of
California, Riverside.
Biogenic secondary organic aerosols (BSOA) are a major
component of atmospheric particles that impact air quality
and climate via radiative and hygroscopic processes.
Direct measurements of BSOA properties in the
atmosphere are usually complicated by the presence of
other aerosol species. Here we examine BSOA properties
using a simple continuous flow reaction chamber. We
present measurements of the optical and hygroscopic
properties of BSOA generated through the reaction of
alpha-pinene and ozone at two controlled relative
humidity values (<5 and 90%) and selected
concentrations and ratios of alpha-pinene to ozone. We
measured aerosol size distributions using a differential
mobility analyzer and optical particle counter sizing
system. Based on optical and mobility size we retrieved a
refractive index, n ~ 1.55, which is within the range of
values assumed by models, and similar to values retrieved
previously using the same method for other oxygenated
carbonaceous aerosols. BSOA hygroscopicity was
determined using a humidified tandem differential
mobility analyzer (60 < RH < 90%) and a cloud
condensation nuclei (CCN) counter (0.3 < s < 1.0%). The
results are interpreted using the hygroscopicity parameter
kappa, which compares the two measurements on a
common scale. Hygroscopicity inferred from growth
factor data (0.008 < kappa < 0.045) was considerably
lower than required to explain CCN activity (0.08 < kappa
< 0.16), consistent with previous studies of BSOA
generated in smog chambers. High relative humidity
inside the reactor increased the overall hygroscopicity of
the BSOA. Total aerosol mass loading, concentrations of
ozone and alpha-pinene and their ratios did not strongly
affect observed hygroscopicity.
Implications of these findings for blank-correcting the
STN OC/EC data will be discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11L.1
A Comparison of Thermal-Optical Carbon Measurement
Methods for Aerosols Emitted by a Series of Controlled
Biomass Burning Experiments. GAVIN MCMEEKING, Amy
Sullivan, Sonia Kreidenweis, Jeffrey Collett, Jr., Colorado State
University; Thomas Kirchstetter, Melissa Lunden, Lawrence
Berkeley National Laboratory; Antony Chen, Daniel Obrist,
Hans Moosm
Accurate determination of the organic and elemental
carbon content of atmospheric aerosols is necessary for
composition-based calculations of aerosol properties. A
large number of measurement techniques are available for
determining the carbonaceous content of aerosol and
these techniques often give significantly different values
for organic carbon (OC) and elemental carbon (EC)
concentrations. Several studies have compared different
thermal-optical measurements for a wide range of urban
sources of carbonaceous aerosols, but only a limited
number of biomass burning aerosol samples have been
studied. Here we compare carbon measurements
performed on biomass burning samples using three
protocols: IMPROVE Thermal Optical Reflectance,
NIOSH Thermal Optical Transmittance, and LBNL
Evolved Gas Analysis. The biomass burning aerosol
samples were collected during a series of controlled
laboratory burns of approximately 20 different North
American fuels at the US Forest Service's Fire Science
Laboratory. Overall, burns dominated by flaming
combustion had EC to total carbon (TC) ratios
significantly higher than observed for burns dominated by
smoldering combustion. However, the different
measurement protocols yielded EC/TC ratios that differ
by as much as a factor of two. These differences were due
largely to the method (transmittance or reflectance) used
to correct for OC that was pyrolized or 'charred' during
filter heating. We explore the optical properties of this
'char' using multi-wavelength attenuation measured as
part of the Evolved Gas Analysis method. We also
investigate the impact of carbon measurement
uncertainties on calculated optical and hygroscopic
properties of biomass burning aerosol.
11L.2
Volatility of Organic Materials from Quartz Filters. CHIN
H. PHUAH, Ann M. Dillner, University of California - Davis.
Carbonaceous aerosols are a major component of ambient
particulate matter and are important because they
adversely affect human health, visibility and radiation
balance of the atmosphere. Organic and elemental carbon
concentrations are monitored in rural and urban
environments by the Interagency Monitoring of Protected
Visual Environments (IMPROVE) and Speciation Trends
Network (STN), respectively. Both networks collect
aerosol samples on quartz filters that are subsequently
analyzed using thermal-optical methods. Deposits on
filters include some semi-volatile organic materials which
may volatilize due to temperature fluctuations at the
sampling site and during shipping and handling. Our
research objective is to study the extent of volatility of
organic material from quartz filters. Portions of collocated
PM2.5 samples are immediately analyzed and the rest
heated at 40C for a period of time prior analysis. The
difference in carbon concentrations of the heated filters to
immediately analyzed filters is our measure of volatility.
A Sunset Thermal Optical Analyzer was used for carbon
analysis using the IMPROVE_A temperatures (not the
protocol) and a transmittance laser to determine organic
pyrolisis. IMPROVE_A temperatures were used in this
project to determine the impact of volatilization on the
IMPROVE carbon measurements. A temperature
calibration method was developed and performed on the
analyzer to correct for the discrepancies between the
actual sample temperature and the analyzer reported
temperature, as IMPROVE temperatures are actual
sample temperatures but the analyzer measures
temperatures about 20mm from the filter sample.
Preliminary result shows that OC1 is the fraction that
decreased the most, from 30% to 54% when heated at
40C. In one test, filter samples are heated for 24, 48, and
96 hours and the corresponding OC1 fraction decrease is
41, 52 and 64% respectively. All other fractions changed
by less than 10%. In another test, filters are heated for 2,
4, 6, and 10 hours. OC1 decreases 21% after heated for 4
hours.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11L.3
Real-time analyzers for routine measurement of HNO3,
NH3, NO3- and NH4+. ERIC EDGERTON, Ben Hartsell,
Atmospheric Research & Analysis, Inc.; D. Alan Hansen, Eladio
Knipping, EPRI.
11M.1
PM2.5 composition of several woodsmkoe events observed
durig the winters of 2005-2007. Eric Edgerton, BEN
HARTSELL, Atmospheric Research & Analysis, Inc.; Justin
Walters, John Jansen, Southern Company.
Continuous measurements of aerosol components can
yield better understanding of the dynamics of particulate
matter, expand opportunities to infer source-receptor
relationships, improve the analysis of how aerosols
impact human health and ecosystems, and allow for rapid
dissemination of measurement data to interested parties.
Despite many advances, continuous aerosol
measurements other than those for total PM2.5 mass,
carbon and sulfate are generally limited to short-term,
intensive campaigns operated by highly trained
individuals. A notable exception involves the
Southeastern Aerosol Research and Characterization
(SEARCH) network. SEARCH is compriised of 8 sites,
each of which measures a wide array of aerosols and
aerosol precursors including the important nitrogen
species HNO3, NH3, particulate-NO3- and particulateNH4+. SEARCH techniques involve conversion of target
species to nitric oxide (NO) following extensive
modification of commercial trace gas analyzers to obtain
5-minute average concentration data. Less extensive
modifications are required to obtain 60-minute averages,
such as might be required for NAMS and SLAMS
monitoring objectives. This presentation will describe the
EPRI-sponsored inorganic Continuous Aerosol
Measurement System (iCAMS) project, the objective of
which is to simplify SEARCH analyzers so they can be
used in routine monitoring networks. Modifications of
SEARCH techniques will be discussed and comparisons
between continuous measurements and discrete (i.e.,
annular denuder system) measurements will be presented.
Accuracy, bias and detection limits for modifed
analyszers will also be presented.
Carbonaceous aerosols represent a significant percentage
of PM2.5 mass across the southeastern U.S. Recent
inventories show that biomass burning (including
prescribed burns, agricultural burning and wildfires) is an
important, and perhaps growing, source of carbonaceous
aerosol. This presentation will analyze biomass burning
events observed at Southeastern Aerosol Research and
Characterization (SEARCH) sites during the winters of
2005-2007. Several types of events will be explored. The
first type occurs at rural sites in the immediate vicinity of
a burn. For these events, aerosol from the burn
overwhelms the local background and PM2.5 composition
approximates that of wood smoke. As an example, one
such event observed at the Oak grove, MS site in January
2005 produced 24-hour PM2.5, OC and EC
concentrations of 131, 75 and 16 micrograms/cu. meter,
respectively. Another type of event occurs when smoke
from one or more fires accumulates under a shallow
boundary layer, is tranported into an urban area and
contributes to a multi-day PM2.5 episode. In this case,
the resulting aerosol composition is a complex mixture of
wood smoke, urban primary emissions and products of
gas-particle reactions within the smoke plume. An
example of this type of event occurred at the North
Birmingham, AL SEARCH site in early March of 2007
and produced several days of 24-hour PM2.5, OC and EC
in excess of 40, 15 and 5 micrograms/cu. meter,
respectively. Meteorological data, trace gas
concentrations and continuous PM2.5 composition will be
used to classify and compare these and other biomass
burning events observed at SEARCH sites.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11M.2
Experimental and theoretical closure experiments for
biomass smoke using extinction cells, photoacoustics and
nephelometry. LAURA MACK, Daniel Obrist, Hans Moosm
The FLAME I (2006) and II (2007) experiments were a
series of laboratory studies of the chemical, physical, and
optical properties of fresh smokes from the combustion of
a variety of fuels that would affect air quality in the
western and southeastern U.S. The burns were conducted
in the combustion chamber of the USFS Fire Sciences
Laboratory in Missoula, Montana. This presentation
focuses on attempts at optical closure using
nephelometer-measured scattering coefficients,
photoacoustically-measured aerosol absorption, and
extinction coefficients measured using a cavity ring-down
extinction cell operated at a wavelength of 532 nm. We
first show results for expected uncertainties, estimated
from the various instrument characteristics and from
instrument calibration studies. We then present selected
case studies from FLAME I and II that illustrate the
degree of closure for smokes having widely varying
morphologies and single scattering
albedos.
11M.3
Chemistry of Air Toxics Emitted from In-use Heavy Duty
Vehicles Equipped with DPF and SCR Retrofits. M.-C.
OLICER CHANG, Paul Rieger, Jorn D. Herner, Alberto Ayala,
William H. Robertson, Keshav Sahay, and Mark Fuentes,
California Air Resources Board.
The California Air Resources Board (CARB)
promulgated the new particulate matter (PM) and NOx
emission standards for the heavy-duty vehicles. The new
standards require 90% reduction of PM emissions in 2007
(comparing to 2006), and 90% reduction of NOx
emissions between 2007 and 2010. While diesel engine
manufacturers have demonstrated these standards with
advanced engine modification, and/or equipped with
advanced after-treatment devices such as diesel particle
filters (DPFs) and selective catalyst reduction technology
for NOx, very little is known about the chemistry of
emissions of air toxics.
The CARB HD air toxic emission program, in
collaboration with University of Southern California and
The South Coast Air Quality Management District
(SCAQMD), takes place at the CARB's Heavy Duty
Vehicle Emissions Laboratory in Los Angeles. Vehicle
testing and emission measurement protocols follow
40CFR Part 86 on the chassis dynamometer. Test cycles
include vehicle 50 mph (cruising), UDDS (transit), and
idling (0% engine loading). In addition to real time
measurements of criteria pollutants, PM size distributions
in mass and number concentrations, time integrated
samples are collected in the end of each test cycle,
including volatile organic carbonaceous (C2-C12),
carbonyls, ammonia, and multiple PM samples for water
soluble ions, organic and elemental carbon (OC/EC,
IMPROVE_A), and polynuclear aromatic hydrocarbons
(PAHs).
In this presentation, we will compare the impact of 1)
vanadium- and zeolite-based SCRT, and 2) test cycles on
the emissions of VOC and PM compositions. Meanwhile,
the VOC and PM speciation profiles from HD with the
new after-treatment systems, in addition being used for
correlating to the relative toxicity of air pollutants in this
program, will be used for estimating the emission rates for
these vehicle fleet that are needed for air quality
management.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11M.4
Diesel Engine Emissions Detection Using a Photoelectric
Tandem Differential Mobility Analyzer. MICHAEL A.
HILL, Suresh Dhaniyala, Clarkson University; Brian Frank,
Thomas Lanni, New York State Department of Environmental
Conservation.
Aerosol particles composed of photoemissive substances
such as carbon can be selectively detected and
distinguished from background concentrations using the
technique of photoelectric charging (Matter et al., 1995).
Here, we present measurements using a photoelectric
tandem differential mobility analyzer (P-TDMA) which
uses the combination of photoemissive response and
mobility measurement to obtain real-time compositional
information about combustion particles. In the P-TMDA,
two differential mobility analyzers are operated in tandem
with a photoelectric charger (222 nm KrCl excimer lamp)
inbetween. Laboratory experiments with a range of
intermixed aerosols confirm that carbon particles can be
quantitatively identified from the mixture. For
deployment of the P-TDMA for ambient source-selective
measurements, the photoelectric charging profiles of
different emission sources are required. Towards that
end, emission measurements of a diesel generator, and SI
and diesel vehicles, were made at the New York State
Department of Environmental Conservation's s mobile
sources test facility. The engine loads were varied and
the photoelectric profiles of the emissions were obtained
as a function of particle size. Preliminary analysis
suggests that the photoelectric response of particles in the
100 nm size range can be used to identify different mobile
source contributions to urban ambient particle
populations.
References
Matter, D, M Mohr, W Fendel, A Schmidt-Ott, and H
Burtscher. ''Multiple Wavelength Aerosol Photoemission
by Excimer Lamps.'' Journal of Aerosol Science 26
(1995): 1101-1115.
11M.5
The Impact of Primary Aerosol from Ocean-going Engines
on Air Quality in the Southern California Air Basin.
DAVID R. COCKER III, Harshit Agrawal, Abhilash Nigam, J.
Wayne Miller, University of California Riverside, CE-CERT;
William W. Welch, CE-CERT; Solomon Teffera, South Coast
Air Quality Management District.
Marine ports in the United States are major hubs of
economic activity and major sources of pollution.
Enormous ships with engines running on the fuel with
heavy components, variety of cargo handling
equipments, thousands of diesel truck, trains with diesel
locomotives, and other polluting equipment and activities
at marine ports cause an array of environmental impacts
that can seriously affect local communities and the
environment. These impacts range from increased risk of
illness, such as respiratory disease or cancer, to increases
in regional smog, contamination of water, and the blight
of local communities and public lands.
Accurate assessment of impact of emissions from marine
ports necessitates the need for an emission marker
specific to ship emissions. Potential unique markers are
the elements in the crude oil and lubricating oils.
Specifically the bunker fuel oil contains high amounts of
Sulfur, and Vanadium and Nickel. The lubricating oil
contains performance additives like phosphorus, calcium
and zinc.
A number of source emissions tests were performed on
various ship engines operating on heavy fuel oil on the
ports of Southern California. The ship emission studies
show a constant ratio of vanadium to nickel over the
different tests performed. Emission indices for Ni, V, and
PM are reported. The measured emission rates of V and
Ni relative to total PM are used to directly estimate the
influx of primary particulate from marine vessels into
Southern California using the V and Ni data from ambient
air quality monitoring station across the South Coast Air
Basin.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
11M.6
Using Multi-Wavelength Aethalometer Measurements to
Characterize and Quantify Wood Burning Versus Traffic.
ANDRE S.H. PREVOT, Jisca Sandradewi, Ernest Weingarnter,
Martin Gysel, Nolwenn Perron, M. Rami Alfarra, Urs
Baltensperger, Paul Scherrer Institute, Switzerland; Soenke
Szidat, University of Bern, Switzerland.
In an Alpine valley, we found very distinct diurnal cycles
of the wavelength dependence of the aerosol light
absorption. We suggest to use an exponential fit
separately for the lower and higher wavelenghts to be able
to match the observed wavelength dependence. In the
evening, we found very high absorption in the UV
concurrent with very high contribution of wood burning.
14
This was corroborated by concurrent C analyses and
aerosol mass spectrometer measurements with distinct
wood burning aerosol mass spectra. The combination of
the datasets is used to evaluate the possibilities to quantify
the contribution of wood burning and traffic to the
particulate matter using the aethalometer data. Results
will be shown for the location in the Alpine valley but
also at urban and highway sites in Switzerland.
13C.1
Particle Soot Absorption Photometer (PSAP) Noise and
Averaging. Stephen R. Springston, Jeonghoon Lee, ARTHUR
J. SEDLACEK III, Brookhaven National Laboratory.
The Particle/Soot Absorption Photometer (PSAP), as
manufactured by Radiance Research (Shoreline, WA),
provides a measure of particle absorbance based on the
changing optical transmittance of a filter as particles are
continuously deposited. The time differential of the
resulting transmittance signal is related through Beer's
law to provide a time series of optical absorption
coefficient. Relative simplicity, low cost, and small
footprint have led to wide deployment of these units in
both ground- and aircraft-based sampling systems.
Limitations of the digital optical absorption coefficient
signal as produced by the instrument firmware include:
roundoff, internal truncation and a summation time set
during operation. Another artifact of the instrument
firmware is a high autocorrelation in the output. This
results in data which is not well-suited to normal boxcar
averaging. The noise in this signal does not have the
normally expected square root of averaging time
dependence. The instrument output also includes raw
intensity values which can be externally processed with
different summation (averaging) times and without the
other firmware shortcomings.
The limitations of the absorption coefficient signal are
illustrated. Some of these limitations can be
circumvented by externally processing the raw intensity
data. Examples of simulated data processing, laboratory
tests to measure instrument noise and the application of
different averaging methods to aircraft-based
measurement data are shown. Guidelines for achieving
optimum signal-to-noise as a function of averaging time
are suggested. Because the measurement is based on a
time differential of the intensity, the noise in absorption
coefficient is shown to decrease with the averaging time
to the 1.5 power in theory. Experimentally, the exponent
was 1.3, but still much different from the 0.5 associated
with methods that directly measure absorption coefficient
(Photoacoustic Spectroscopy or Photothermal
Interferometry).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
13C.2
Emperical Evaluation of the Aethalometer Spot Matrix
Effect on Ambient Air Using A Thermodenuder. GEORGE
ALLEN, NESCAUM; Jay Turner, Washington University at St.
Louis.
The Magee Scientific Aethalometer (TM) is a widely used
monitor to continuously measure black carbon (BC) in the
atmosphere by measuring optical attenuation on a quartz
filter tape (no scattering measurements are made).
Several articles in the recent peer-reviewed and grey
literature have shown that the Aethalometer BC response
can be a strong function of the deposit (\spot\) loading
when the aerosol matrix is not massively scattering, with
the reported BC concentration decreasing substantially
with increased spot loading for a constant aerosol. We
present an experimental examination of this effect at a site
near Boston MA during the summer, an environment
where the spot loading dependence of reported BC has a
distinct seasonal pattern and is generally not observed in
the summertime. A TSI 3065 thermodenuder, operated
upstream of the Aethalometer at 220C, is used to remove
a substantial fraction of the scattering aerosol (OC, SO4,
NO3), presumably without pyrolizing OC. Temporal
differences in the spot loading to BC relationship are
quantified for collocated instruments operating with and
without the upstream conditioning. All instruments are
run with a PM2.5 inlet size cut, and BC response across
all instruments is normalized. Information on hourly
sulfate and thermal EC/OC from an urban-scale site in
Boston is available to assess the overall aerosol matrix
composition. Strategies for data compensation methods
in the absence of allied data streams such as scattering
measurements are examined. All Aethalometers are
operated in configurations optimized for the application
of this algorithm.
13C.3
Albedo Measurements and Optical Sizing for Single Aerosol
Particles. TODD SANFORD, David Thomson, Earth System
Research Laboratory NOAA and Cooperative Institute for
Research in the Environmental Sciences University of Colorado;
Daniel Murphy, Earth System Research Laboratory NOAA;
Richard Fox, National Institute of Standards and Technology.
Atmospheric particles affect the radiative balance
of the Earth. A metric for this effect is the albedo, which
is the ratio of scattering to extinction (extinction =
scattering + absorption). Measurements of albedo can
then be used to determine the sign and relative magnitude
of the radiative forcing by atmospheric particles. The
mixing states of black carbon particles also play a role in
the determination of the radiative forcings. Bulk
measurements of albedo do not provide detailed analysis
on a particle-by-particle basis, which is where particle
mixing state information would be found. With these
concerns in mind an instrument has been developed to
carry out scattering and extinction measurements
simultaneously on single particles. These two
measurements carried out in a single instrument allows
for sources of error such as particle position in the laser
beam and relative humidity differences to cancel out.
Also, errors associated with combining measurements
from different instruments and different techniques to
determine albedo are alleviated. The instrument consists
of a diode laser frequency locked to an external triangular
ring cavity. Particles are introduced into the cavity and
the light scattered by the particles is collected by a
spherical/ellipsoidal scattering cell. Also, the light
scattered in the forward and backward directions is
collected in independent measurement channels. The
forward scattering to total scattering ratio is used for
particle sizing. Simultaneously, extinction due to the
particles is obtained via either a cavity ringdown
measurement or by measuring the decrease in the cavity
leakage signal as the particle moves through the beam.
Both the scattering and extinction measurements are
triggered by individual particles introduced into the beam
and are combined for the albedo measurement of each
particle. Instrument characterization is carried out with
laboratory-generated particles of known sizes and
scattering/absorption properties with modifications being
made for eventual field deployment.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
2007 AAAR Annual Conference Abstracts
13C.4
A Comprehensive Temperature Protocol for ThermalOptical Transmission Analysis Optimized for Atmospheric
Black Carbon. JOSEPH M. CONNY, National Institute of
Standards and Technology; Gary Norris, National Exposure
Research Laboratory, U.S. EPA.
We present an optimized temperature protocol for the
measurement of black carbon (BC) in particulate matter
<2.5 micrometers (PM2.5) by thermal-optical
transmission analysis (TOT) that is based on the BeerLambert Law and response surface modeling.
In TOT, the transmission of laser light through a particleladen filter is monitored over time while carbonaceous
material is removed in several heating steps and measured
by flame ionization detection. In helium, the laser signal
is attenuated by the pyrolysis of organic carbon (OC).
Later, while carbon is removed in an oxidizing
atmosphere, the laser signal returns to its value prior to
pyrolysis (split point), whereupon the amount of carbon
equivalent to BC is measured. As BC and pyrolyzed OC
oxidize during analysis, they may not separate physically.
Since pyrolyzed OC may evolve beyond the split point,
the specific absorption cross sections of pyrolyzed OC
and BC must be equivalent.
Optimizing the thermal protocol relies on establishing
equivalent absorption cross sections for pyrolyzed OC
and BC. In addition, the optimization requires sufficient
pyrolysis so that unpyrolyzed OC is not measured as BC
beyond the split point. The convergence of response
surfaces for the pyrolyzed-OC and BC cross sections as
well as a plateau in the BC cross-section surface revealed
the thermal conditions that satisfied the optimization
criteria. From models based on extensive analyses of
PM2.5 samples from Atlanta, Los Angeles, and Seattle,
the following protocol was determined: Steps 1-4 in He:
200 C for 60s, 400 C for 60s, 600 C for 60s, 785 C for
150s; Steps 1-6 in O2-He: 550 C for 60s, 620 C for 60s,
690 C for 45s, 760 C for 45s, 830 C for 45s, 900 C for
90s.
Although this work was reviewed by EPA and approved
for publication, it may not necessarily reflect official
Agency policy.
13C.5
Single-Particle Size, Shape, and Carbon Composition of
Ambient Aerosols by Scanning Transmission X-Ray
Microscopy Analysis. SATOSHI TAKAHAMA, Stefania
Gilardoni, Lynn Russell, Scripps Institution of Oceanography University of California at San Diego; David Kilcoyne,
Lawrence Berkeley National Laboratory.
Organic aerosols play a critical role in air quality and
climate change. Improving our understanding of their
interaction with other constituents of the atmosphere
requires better constraints on the types of organic
compounds present in the aerosols, as well as their mixing
state, morphology, and emission/production source class
(Kanakidou et al., 2005; Fuzzi et al., 2006). In this study,
we present a summary of 595 ambient particles collected
between 2000 and 2006 and analyzed for organic
functional group abundance and morphology by Scanning
Transmission X-Ray Microscope. These particles ranged
between 0.1 and 12 micro-meters and represent aerosols
found in a large range of geographical areas, altitudes,
and times, and include samples from seven different field
campaigns: PELTI, ACE-ASIA, DYCOMS II, Princeton,
MILAGRO (urban site), MILAGRO (aloft), and INTEXB. We show the presence of at least fourteen different
classes of organic particles based on spectroscopic
signatures; different particle types are found within the
same region while the same particle types are also found
in different geographical domains. Combustion-derived,
humic-like, and carboxylic-dominated particles have also
been identified based on comparison of sample spectra
describing the character of carbon-bonds with reference
spectra published in the literature.
References
Fuzzi, S., M. O. Andreae, B. J. Huebert, M. Kulmala, T.
C. Bond, M. Boy, S. J. Doherty, A. Guenther, M.
Kanakidou, K. Kawamura, V. M. Kerminen, U.
Lohmann, L. M. Russell, and U. Poschl (2006), Critical
assessment of the current state of scientific knowledge,
terminology, and research needs concerning the role of
organic aerosols in the atmosphere, climate, and global
change, Atmos. Chem. Phys., 6, 2017
Kanakidou, M., J. H. Seinfeld, S. N. Pandis, I. Barnes, F.
J. Dentener, M. C. Facchini, R. V. Dingenen, B. Ervens,
A. Nenes, C. J. Nielsen, E. Swietlicki, J. P. Putaud, Y.
Balkanski, S. Fuzzi, J. Horth, G. K. Moortgat, R.
Winterhalter, C. E. L. Myhre, K. Tsigaridis, E. Vignati, E.
G. Stephanou, and J. Wilson (2005), Organic aerosol and
global climate modelling: a review, Atmos. Chem. Phys.,
5, 1053-1123.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
2007 AAAR Annual Conference Abstracts
13C.6
Optical Analysis of Coated and Uncoated Soot Particles:
Data for Global Climate Change Models. PEDRO BUENO,
Stephanie Winter, Jeffrey Stehr, George Mulholland, Russell
Dickerson, Michael Zachariah, University of Maryland College Park.
Atmospheric aerosols play a fundamental role in Earth’s
atmospheric chemistry and climate. Soot is an absorbing
aerosol, though the magnitude of that absorption has
largely been determined by measuring the optical
properties of uncoated soot. It has been proposed that
coated soot might absorb radiation more efficiently than
uncoated soot, thus warming the climate more than
previously suspected. For this study, soot is generated in a
well-controlled Santoro-Style diffusion flame burner with
ethylene as the fuel, and has been successfully coated
with dibutyl phthalate (DBP). DBP has a refractive index
of 1.490 (real part), which is similar to the refractive
index of sulfuric acid (n=1.426) at 589 nm. DBP is
substituted for the commonly found sulfate coated
particles for several reasons including safety and
instrument integrity. By changing the temperature of the
DBP, the vapor pressure of the DBP is changed and
consequently, the coating thickness can be changed. The
aerosols are measured with a differential mobility
analyzer (DMA). From the DMA the aerosols are sent to
a condensation particle counter for size distributions or
sent to be analyzed. Optical analysis performed with a
multi-pass extinction cell (MPEC) where the total
scattering and extinction cross-sections are measured.
The MPEC uses a 632nm laser and a 100 m path length.
The absorbance cross-section is calculated from those
measurements. This study improves upon previous
optical studies of soot by using soot particle sizes that are
representative of primary soot particles in the atmosphere
providing a more solid and viable result.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Carbonaceous Aerosols
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.1
Development and Validation of a Model to Predict Aerosol
Breathing Zone Concentrations During Common Outdoor
Activities. JONATHAN THORNBURG, G. Gordon Brown,
RTI International; John Kominsky, Environmental Quality
Management, Inc.
2G.2
Demonstrating the Benefits of a Technician Training
Program for a Successful Longitudinal Research Study.
Jerermy Seagraves, Andrew Dart, JONATHAN THORNBURG,
Jeff Portzer, Charles Rodes, RTI International; Don Whitaker,
Ron Williams, U.S. EPA.
Research has been conducted on aerosol emission rates
during various activities as well as aerosol transport into
the breathing zone under idealized conditions. However,
there has been little effort to link the two into a model for
predicting a person's breathing zone concentration. This
research developed a model to calculate the breathing
zone concentration produced by common outdoor
activities. The model combined aerosol physics and fluid
dynamics principles to develop a set of mass balance
equations applicable to activities of varying intensity and
duration.
A longitudinal research project like the Detroit Exposure
and Aerosol Research Study (DEARS) requires a
comprehensive technician training program to assure
collection of high quality data to address study
hypotheses. More than 40,000 exposure samples
(representing surveys, passive gas monitors, and active
particulate matter samplers) were collected over six
sampling seasons. Additionally, the training had to
flexibly address a range of technician skill levels and
adapt to staff turnover. Accordingly, the training program
was upgraded seasonally to keep pace with changing
project and personnel requirements.
The model consisted of four distinct modules. The first
module defined the characteristics of the person, their
activity, and their location. The second and third modules
calculated the aerosol emission rate generated and the
aerosol transport efficiency to the breathing zone,
respectively, for a specific activity. Activities simulated
included motorcycle riding, yard work, and child's play.
The equations used in modules 2 and 3 depended on
whether the turbulence intensity generated by the activity.
The fourth module calculated the resulting breathing zone
concentration.
Whenever possible, published data were used as model
input independent parameters. If data were not available,
a range of input values were assumed. Monte Carlo
simulation applied to the model equations generated a
probability output distribution for comparison with
experimentally measured results and to perform a
sensitivity analysis on the independent parameters.
Scenario model output distributions were either normal or
log-normal, depending on the distribution of the most
influential input variables. The modeled output
distributions successfully bracketed the range of
experimentally measured breathing zone concentrations.
However, the modeled breathing zone concentration
range during certain activities was 1000 times broader
than the other scenarios because of the positive interaction
between highly significant input variables.
The DEARS training program followed four stages. The
introductory stage consisted of self-guided review of a
comprehensive training manual. This manual contained a
project overview and abridged versions of all Research
Operating Procedures (ROPs). Technicians subsequently
were taught equipment and survey use, sample archival,
and basic aerosol physics in an interactive class
environment. The third phase was collection of samples
and surveys at two residences for three days. This stage
allowed new staff to practice with all equipment and
witness common sample collection problems under the
guidance of experienced personnel. Training culminated
with new and veteran technicians working together during
the first week of real sample collection. This multi-stage
approach reinforced critical thinking skills and reduced
technician errors when collecting real samples during the
remainder of the sampling season.
Comprehensive training for all technicians contributed
substantially to the success and overall cost-effectiveness
of DEARS. Technicians quickly adapted to increases in
workload without extending the sampling periods or
increasing the number of personnel. Most importantly,
increased data capture rates correlated with refinement of
the training program without an increase in project cost.
This framework also proved very effective for training
technicians with disparate backgrounds and experience.
Although this work was reviewed by U.S. EPA and
approved for publication, it may not necessarily reflect
official Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.3
DNS of Aerosol Motion in a Model Room. XINLI JIA, John
B. McLaughlin, Goodarz Ahmadi, Clarkson University; Jos
Derksen, Delft University of Technology.
2G.4
Resuspension of Dust Particles in a Chamber and the
Associated Factors. JING QIAN, Andrea R. Ferro, Clarkson
University.
This talk will present results for the computed trajectories
of aerosols in a model room with a displacement
ventilation system. The fluid velocity fields used in the
trajectory computations were obtained from a DNS using
the Somers formulation of the lattice Boltzmann method
(LBM). The Somers formulation is significantly more
complex than the BGK formulation of the LBM, which is
much more commonly used. The Somers formulation has
two primary advantages over the BGK formulation that
make it more suitable for simulations of indoor air
pollution. First, it is more stable. It has been used by one
of the co-authors (J.D.) to perform LES of stirred tanks at
Reynolds numbers, based on the impeller size and speed,
up to . Second, the Somers formulation permits one to
perform DNS of natural convection for a broad range of
Prandtl numbers. Simulations can be done only for Pr=0.5
for the BGK formulation, and the set of lattice vectors
must be expanded to solve the energy equation. This talk
will focus on isothermal conditions, but a few results will
be shown that include natural convection. Using the
computed flow fields, the trajectories of aerosols ranging
in size from 2 to 20 micrometers are computed. The
equation of motion includes gravity and nonlinear drag.
The particles are uniformly dispersed in the inlet vent and
then tracked until they either deposit on a solid surface or
exit through the outlet vent. The spatial distributions of
deposited particles will be discussed and the roles of
gravity and particle inertia will be assessed. Although the
size of the model room is of order 2 meters, the Somers
formulation permits DNS with realistic inlet velocities at
Reynolds numbers of order 104. We will discuss the use
of DNS to test LES and RANS simulations of indoor air
pollution.
Resuspension experiments were conducted in a full-scale
experimental chamber to investigate particle resuspension
from human activities. Three types of flooring (vinyl
tiles, new carpet, and old carpet) and two ventilation
(mixing and displacement) configurations were tested
during the experiment. The floorings were seeded with
0.1-10 micro-meter Arizona Road Test particles. Sizeresolved resuspension rate was estimated from an airsurface compartment model using real-time particle
concentration data. Resuspension rates are in the range of
-5
-2
-1
10 -10 hr for particles in size ranges of 0.8-10 micrometer, with higher resuspension rates associated with
larger particles. Resuspension via walking activity varied
from experiment to experiment. A \heavy and fast\
walking style was associated with a higher resuspension
rate than a less active style. The ventilation condition
could affect the resuspension on the same scale as the
person-to-person variability. Given the same floor loading
of the test particles, resuspension rates for the carpeted
floor were on the same order of magnitude but
significantly higher than those for the hard floor.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.5
Exposure to Indoor PM: Effects of Climatic and Cultural
Influences. VIVIANA ACEVEDO-BOLTON, Lynn
Hildemann, Stanford University.
2G.6
Silver-deposited Activated Carbon Fibers for Bioaerosol
Control. KI-YOUNG YOON, Jeong Hoon Byeon, Jae-Hong
Park, Chul-Woo Park, Jungho Hwang, Yonsei University.
Air quality standards designed to protect human health are
based on outdoor levels, yet we spend almost 90% of our
time indoors. The concentrations and chemical
characteristics of outdoor particulate matter (PM) are
often not representative of what we are exposed to
indoors - penetration losses reduce the infiltration of PM
from outdoors, while indoor sources elevate indoor PM
levels.
Bioaerosols are airborne particles of biological origins,
including viruses, bacteria, fungi, and all varieties of
living materials. In suitable hosts, bioaerosols are capable
of causing acute or chronic diseases that may be
infectious, allergenic, or toxigenic. Bioaerosols from
outdoor air accumulate on filters of heating, ventilating
and air conditioning (HVAC) system in large quantities
and are able to multiply there under certain conditions.
Activated carbon fiber (ACF) filter is widely used in air
cleaning to remove hazardous gaseous pollutants because
of their extended surface area and high adsorption
amount. However, the ACF filters have good
biocompatibility and bacteria may breed on the ACF
filters, so that the ACF filters themselves becomes a
source of bioaerosols. In this study, silver, traditionally
well-known as antimicrobial material, was deposited on
ACF filters by an electroless deposition method and their
efficacy for bioaerosols removal was tested. Physical
filtration and biological antimicrobial test were performed
and various surface analyses such as scanning electron
microscopy, inductive coupled plasma, and X-ray
diffraction were used to characterize the prepared ACF
filters. Silver-deposited ACF filters showed antimicrobial
effects whereas pristine ACF filters did not. Electroless
silver-deposition did not influence the physical
characteristics such as pressure drop and filtration
efficiency of ACF filters. Gas adsorptive ability of silverdeposited ACF filter decreased compared to the pristine
one because of the blockage of the micropores of ACF by
silver particles. Therefore silver contents on ACF filters
need to be optimized to avoid the excessive reduction of
adsorptive characteristics of the ACF filter and show
effective antimicrobial activity. [This work was supported
by grant No. R01-2005-000-10723-0 from the Basic
Research Program of the Korea Science & Engineering
Foundation.]
The concentration and composition of indoor PM,
including both chemical and biological measures, have
not been well characterized in homes located in different
climates. This project focuses on the differences found
between a temperate climate (San Francisco Bay Area,
CA) and a tropical climate (Singapore), and the effects
that both outdoor climate and building design and
operation may have on indoor air quality. We
hypothesize that we will see higher concentrations in
Singapore of: (1) bioaerosols, (2) elemental carbon, and
(3) sulfate; and elevations in Bay Area homes of (4)
resuspended dust and (5) indoor/outdoor (I/O) ratios of
particles and their components.
The levels and composition of PM in a home can be
influenced by many factors, such as outdoor diesel use (in
Singapore), ventilation conditions (HVAC in Singapore
vs. natural ventilation in the Bay Area), floor type (carpet
in Bay Area vs. no carpet in Singapore), shoe removal (in
Singapore), and different indoor sources. Filter samples
will be collected to analyze for mass, elemental and
organic carbon, water-soluble anions, trace metals,
protein, endotoxin (a tracer for molds), and glucan (a
tracer for bacteria). This presentation, which focuses on
how the study will be designed to test our hypotheses,
will include some preliminary results. For example,
preliminary indoor results show higher sulfate/nitrate
ratios in Singapore, consistent with higher diesel use.
Endotoxin levels in Singapore were similar to those found
in Bay Area homes.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.7
Personal and Indoor Exposure to PM2.5 and Polycyclic
Aromatic Hydrocarbons from Traditional Cooking Practices
in Njombe, Tanzania, East Africa. MARI TITCOMBE, Matt
Simcik, University of Minnesota.
2G.8
Correction of Sampler-to-Sampler Comparisons. PATRICK
T. O'SHAUGHNESSY, The University of Iowa; Vijay Golla,
Western Kentucky University; Jason Nakatsu, Stephen
Reynolds, Colorado State University.
Exposure to indoor smoke from traditional cooking
practices in impoverished countries is responsible for
2.6% of global ill health in human populations, and 1.6
million deaths annually (Desai et. al., 2004). In this
study, personal and indoor exposure to PM2.5 and
Polycyclic Aromatic Hydrocarbons (PAHs) were
measured in households in Njombe, Tanzania using open
wood fires, charcoal, a mix of charcoal and kerosene, and
Liquid Petroleum Gas (LPG) as cooking fuels. Due to
cool local climate and often heavy rainfall, cooking in this
region is conducted indoors, often in small, poorly
ventilated rooms. Results represent work day exposures,
or time spent cooking in the home. Preliminary results
show PM2.5 personal exposures for open wood fire use
roughly two orders of magnitude greater than those of
LPG users. Personal exposures for charcoal, and
charcoal/kerosene users were roughly one order of
magnitude greater than those of LPG users, with pure
charcoal use having roughly a factor of 5 greater
exposures than charcoal/kerosene mix. Significant
differences in PAH exposure were also observed.
Households using wood fires were shown to have the
greatest exposure, followed by charcoal and charcoal/
kerosene users, with the lowest exposures for LPG users.
A comparison is made between exposure levels and
socioeconomic status of the households tested. In
addition, the use of \fuel efficient\ wood stoves for the
reduction of PM 2.5 and PAH exposure was measured in a
local secondary school, boarding approximately 800
students. Proper use of \fuel efficient\ wood stoves was
shown to sharply decrease personal and indoor exposure
of both PM 2.5 and PAHs for the kitchen sampled.
Field studies have been conducted which involve side-byside comparisons of old and new personal aerosol
samplers in order to establish a proportional relationship
needed to relate findings of studies involving any two
samplers. However, the ratios obtained are directly related
to the size distribution of the aerosol during the sampling
episode and will therefore be different if applied to
another setting with a different size distribution. A simple
method for correcting a sampler-to-sampler ratio for
changes in size distribution was developed by computing
a bias factor that relates the measured ratio with a ratio
determined from equations that describe the collection
efficiency curves of the samplers while taking size
distribution into account. Laboratory trials were
conducted to determine whether the resulting bias factor
is independent of aerosol size distribution. During these
studies a 3-piece cassette and respirable cyclone were
compared to an inhalable sampler in both a still-air
chamber and a moving-air chamber operated at 0.2 and
1.0 m/s. An ISO test dust of various size fractions was
generated to produce an aerosol with mass median
aerodynamic diameter ranging from 1.4 to 10.1 micrometers. An organic dust consisting of ground grain
material was also applied to the still-air chamber to
demonstrate differences between dust types. Results
showed that the bias value was significantly different
between dust types for both the cyclone/inhalable (p =
0.001) and cassette/inhalable (p = 0.033) comparisons but
was not different between wind conditions for either
comparison. All but one comparison had insignificant
slopes when comparing the bias value relative to median
diameter indicating that the bias value could be used to
correct for size distributions in most conditions. However,
bias values determined when comparing the cyclone to
the inhalable sampler in the still-air condition produced a
positive slope for median diameters less than 4 micrometers (p = 0.008).
REFERENCES
Desai, A. M., Mehta, S., Smith, K. R. (2004). Indoor
smoke from solid fuels, Assessing the environmental
burden of disease at national and local levels. Geneva,
World
Health Organization, 2004 (WHO Environmental Burden
of Disease Series, No. 4).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.9
Use of Synthetic-Jet-Based Active Flows to Control Particle
Dispersion. JENNIFER ZIEGLER, Michael Amitay, Lupita D.
Montoya, Rensselaer Polytechnic Institute.
The present threat of chemical or biological attack from
terrorists combined with the current American lifestyle,
spent predominantly indoors, presents a real opportunity
for developing smart systems for indoor air quality. The
goal of the current research is to control the dispersion
and movement of a plume of particles, from submicron to
supermicron sizes, using synthetic-jet-based active flow
control. To test this concept, a dedicated, two foot square,
closed chamber was constructed. The chamber
configuration includes a particle point source in the center
of the floor, synthetic jet actuators on both sides of the
source, and a ventilation system along the sidewalls.
The synthetic jet actuators are zero-net-mass-flux in
nature but they provide momentum; thus can alter the
particles field. The time- and phase-averaged velocity
fields of the air and the particles were measured using
Digital Particle Image Velocimetry and Particle Tracking
Velocimetry. When a pair of synthetic jets was activated,
the particle plume was vectored either towards or away
from the jets. The vectoring was controlled by changing
the phase between the two synthetic jets. When the jet
farther from the plume was leading in phase, the plume
was vectored towards the jets. When the jet closer to the
plume was advanced in phase, it resulted in vectoring of
the plume away from jets. Therefore, by controlling the
phase and strength of the synthetic jets, the plume was
controlled and directed to the chamber's ventilation
system to be removed quickly and efficiently. The
decrease in concentration was validated using an aerosol
sampling probe, through the top of the chamber and
connected to an Aerodynamic Particle Sizer.
Measurements were taken throughout the chamber at set
times after release to generate both spatial and temporal
particle distributions.
2G.10
Spatial and Temporal Variability of Particulate Pollutants in
Diesel-Powered School Buses. Maxwell A. Martin, Xiaodong
Zhou, Ryan LeBouf, Emily L. MacWilliams, Alan Rossner,
Peter A. Jaques, ANDREA R. FERRO, Clarkson University.
The characterization of human exposure to diesel
particulate matter (DPM) on school buses is an important
step in understanding the risks to respiratory disease that
children may face throughout the day, both acutely and
chronically. To characterize the potential exposures, DPM
was measured by mass, particle count and composition in
unoccupied buses. Three school buses from the Potsdam,
NY school district were instrumented and monitored
while the buses drove prescribed routes. Self-pollution
and spatial distribution of self-pollution was determined
by adding sulfur hexafluoride (SF6) directly to the
exhaust and monitoring the SF6 at 8 locations inside the
bus using an Innova 1312 Multigas Analyzer and a Mark
3 8-Point Sampler. Composition and size distribution of
the DPM was determined using a suite of semi-continuous
instruments. The impact of idling and traveling at various
speeds was determined by keeping records of bus
operation and analyzing the semi-continuous exposure
data. An aerosol mapping technique was used to provide
the temporal and spatial relationship of DPM on the bus
during various operational modes. The mapping is used
to optimize the monitoring protocol for occupied bus
scenarios to best estimate children's exposures while they
are riding, boarding and deboarding the bus and to
correlate these real-time exposures with acute health
endpoints.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.11
Study of Evaporating Droplet Transport and. Mazyar
Salmanzadeh, Goodarz Ahmadi, Clarkson University.
Transmission of diseases from one person to another
due to coughing and sneezing by emission of droplets that
carrying viruses and bacteria in hospital patient rooms is
studied. A computational model for simulating the
airflow, the thermal and the humidity condition in room
was developed and the distributions of evaporating
droplet in the hospital room with two beds were studied.
The turbulence model was used for continuous fluid
phase calculations and the trajectories of the evaporating
droplets were evaluated with a Lagrangian method. The
particle equation of motion included the viscous drag, the
Brownian, the Saffman lift and the gravity forces. Mixing
and displacement air distribution systems were considered
and trajectories of particles in the range of 1 to 500
microns were simulated. The simulation results suggested
that the chance of disease transmission was higher when
the mixing ventilation system was used. In addition, the
air distribution system does not affect the large particle
trajectories.
2G.12
Resuspension of Particulate Matter by the Human Foot.
JACKY ROSATI, U.S. EPA, National Homeland Security
Research Center (NHSRC); Alfred Eisner, Alion Life and
Environmental Sciences.
Resuspension of particulate matter from flooring surfaces
is a little understood process, yet is thought, along with
tracking, to contribute significantly to the movement of
materials inside a home or office. To investigate what
happens when a human foot steps on a flooring surface
contaminated with particulate matter, an automated
stepping system comprised of a prosthetic foot controlled
by electric actuators was developed. The actuators
control the speed of the �footstep’, as well as the pressure
loading. Thus, the turbulence surrounding the foot as
well as the pressure exerted on the test surface can be
controlled.
Short-pile carpet was seeded with silicon particles. Once
loaded, the carpet was placed under this prosthetic foot,
and a single step was taken. Particle Image Velocimetry
(PIV) was used to study the particle flows under the foot.
These PIV images showed that the visible resuspended
particles are not released from the carpet as the foot is
lifted, but actually fall off the foot after it has moved up
off the flooring surface. These falling particles form a
cloud under the foot that moves rapidly to the front of the
foot during the stepping motion. Particle mean velocities
observed under the foot were about 0.5 m/s. Tests were
performed using both a cotton sock and a rubber soled
shoe. The cotton sock both collected from the carpet and
released from the sole of the foot a greater number of
particles than the rubber-soled shoe. Further investigation
into other types of footwear, flooring surfaces and
materials is ongoing.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
2G.13
The Effectiveness of an Integrated Energy Recovery
Ventilator on the Air Quality in the Bedroom of Asthmatic
Children, 5-14 Years, and Their Improved Respiration and
Restfulness. PETER A. JAQUES, Andrea R. Ferro, Philip K.
Hopke, Clarkson University; Cheryl Gressani, Larry E. Wetzel,
Air Innovations, Inc.
Recent estimates show that about 6.4 million children
under the age of eighteen years suffer from some form of
asthma. Airborne contaminants or allergens exacerbate
the illness. Moderate-to-severe asthmatics that obtain
relief during restful sleep may gain strength and recover.
An efficient prototype air conditioning and filtration
system, Integrated Energy Recovery Ventilator (IERV),
was used to remove particles and climatize the air in the
bedrooms of 45 asthmatic children (5-14 years old).
IERVs were deployed in the homes of children diagnosed
by the respiratory clinic of a local hospital. The children
were split into two equal sized groups for a 3 part case
crossover study. In part 1, group A had the cleaners
turned on and group B did not. During part 2, both had
the cleaners on, and in part 3, Group A turned the cleaners
off and Group B kept theirs running. Each experimental
period averaged about 6 weeks. Each subject serves as
their own control, with an overall 6 weeks of particle
exposure and 12 weeks without. The period of the IERV
being on, followed by it off, is to evaluate whether the
children's reduced inflammation persisted.
2G.14
Relationships Between Indoor And Outdoor Particulate And
Gaseous Species In Two Retirement Homes: Implications
For Particulate Matter Exposure Assessment. ANDREA
POLIDORI, Mohammad Arhami, Constantinos Sioutas,
University of Southern California; Ryan Allen, Simon Fraser
University; Adam Reff, U.S. EPA; Ralph Delfino, University of
California, Irvine.
Hourly indoor and outdoor fine particulate matter (PM 2.5 ),
organic carbon (OC), elemental carbon (EC), particle
number (PN), ozone (O3), carbon monoxide (CO) and
nitrogen oxides (NO, NO 2 and NOX) concentrations were
measured at two different retirement communities in the
Los Angeles basin between July 2005 and February 2007
as part of the cardiovascular health and air pollution study
(CHAPS), a multi-disciplinary project designed to
investigate the effects of micro-environmental exposures
to PM on cardiovascular outcomes. These data were used
to study the relationships between indoor and outdoor
PM2.5 , its components, their seasonal variations, and their
association with gaseous co-pollutants. In particular, the
infiltration factor F inf ; the equilibrium fraction of the
ambient concentration that penetrates indoors and remains
suspended) for all measured particulate and gaseous
species was determined using four techniques: 1) indoor/
outdoor concentration ratios, 2) regression methods, 3) a
recursive mass balance model and 4) a customized multilinear engine model. These methodologies were compared
and their effect on the resulting Finf estimates was
analyzed. Preliminary correlation and regression analyses
showed that 24-h outdoor PM2.5 concentrations were
highly correlated (p< 0.0001) with indoor concentrations
of PM2.5 , but not with indoor particle components or
gases. In contrast, outdoor gas concentrations (especially
CO, NO2 and NOX) were consistently correlated (p<0.05)
with both indoor gas concentrations and indoor
concentrations of PM constituents (EC, OC, and PN). We
hypothesize that these outdoor gases are a good exposure
surrogate for indoor concentrations of combustionderived particles, but not for PM 2.5 mass. Since the
CHAPS retirees spend most of their time indoors, it is
likely that indoor concentrations are a good proxy for
personal exposures. Health effect studies that consider the
ambient concentrations of particle and gaseous species as
independent variables must be analyzed carefully, since
both parameters may be related to components of the
PM2.5 exposure mixture.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
7B.1
Spatial and Compositional Relationships of Indoor Aerosols
in the Detroit Exposure and Aerosol Research Study
(DEARS). ALAN VETTE, Carvin Stevens, U.S. EPA; Charles
Rodes, Jonathan Thornburg, RTI International; Carry Croghan,
Ron Williams, U.S. EPA.
The factors influencing residential and human exposures
to air pollutants of outdoor origin were assessed in the
DEARS by collecting central site and outdoor/indoor
residential PM2.5 samples. These data indicated that the
composition of PM2.5 in the Detroit airshed is similar to
other Midwest and East-coast U.S. cities with organic
carbon (OC) and sulfate being the primary PM2.5
components, especially during summer. A dramatic shift
in PM2.5 composition occurred from summer to winter,
however, with nitrate concentrations increasing by a
factor of three to four. Indoor nitrate concentrations were
only about 50% and 15% of outdoor concentrations
during summer and winter, respectively. Although the
composition of indoor and outdoor PM2.5 was similar, the
relative abundance of components differed considerably,
especially across seasons. In general, indoor and outdoor
PM2.5 consisted primarily of OC, sulfate, nitrate and
elemental carbon (EC) with silicon (Si) and iron (Fe)
constituting the most abundant elements. Potassium (K),
calcium (Ca), manganese (Mn), copper (Cu), zinc (Zn)
and lead (Pb) were also found at lower levels.
Relationships between PM2.5 components concurrently
measured at a central outdoor monitoring site (Allen Park,
MI), and indoors/outdoors at the residences were assessed
using linear mixed effects models. The results of these
analyses on log-transformed data showed that residential
outdoor measurements were generally significantly
related to central site measurements (p<0.05). The mixed
model slopes varied considerably with slopes near unity
for more regional components such as sulfate and OC,
with considerably lower slopes (<0.5) for crustal elements
(Si, K, Ca) and PM2.5 components (Mn, Fe, Cu and Zn)
possibly impacted by local sources. Several of these
elements measured indoors were not significantly related
to central site measurements, particularly Si, Mn, Fe, Cu,
and Zn.
7B.2
Indoor Air Monitoring in Day-Care Centers. Pei-Shih Chen,
YI-LIEN LEE, Ting-Yu Huang, Yu-Han Zhang, Kaohsiung
Medical University.
It was reported that the incidence of many infectious
diseases in day-care center children were greater than
those in homes. Thus, there is a need to assess indoor air
quality, especially influenza virus in day-care center.
Therefore, the aim of this study is to monitor the airborne
influenza A and B virus in day-care centers with different
ventilation form. To our knowledge, this is the first study
to estimate influenza virus concentration in day-care
centers in the world.
Environmental monitoring was held during August 18 to
September 18 in 2006. Day-care center A was near a
busy traffic street and with natural ventilation and air
conditioning while day-care center B was only with air
conditioning system and located in a small lane. An air
sampler (MAS100) was used to collect cultivable airborne
bacteria and fungus. Airborne influenza virus was
sampled by three-piece plastic cassette and analyzed by
real-time qPCR. Temperature, relative humidity, CO, CO
2, and particles were also measured.
The average concentration of influenza A virus in daycare center A and B was 1.2x106 copy-cubic meter and
5.1x10 2 copy-cubic meter, respectively, and the positive
rate was 6.7 percent and 66.7 percent, respectively. The
concentration of influenza B virus in day-care center A
and B was 3.6 x104 copy-cubic meter and 4.7 x10 5 copycubic meter, and the positive rate is 10 percent and 30
percent, respectively. It also found out that bacteria
concentration have positive correlation with wind velocity
and relative humidity in day-care center A, and the fungal
concentration have positive correlation with temperature
and influenza B virus in day-care B. According to the
Indoor Air Quality Recommended Values of Taiwan
Environmental Protection Administration, the failure rate
of bacteria, CO and CO 2 concentration in day-care center
B were all higher than day-care center A.
Although this work was reviewed by EPA and approved
for publication, it may not necessarily reflect official
Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
7B.3
Indoor and Outdoor Concentration of Fine Particles at
Control Site in Mumbai City : A Case Study. ABBA
ELIZABETH JOSEPH, Seema Unnikrishnan National Institute
of Industrial Engineering; Rakesh Kumar, National
Environmental Engineering Research Institute.
7B.4
Ultrafine and Fine Particulate Matter Variation in Skating
Arenas. KELLY SABALIAUSKAS, Greg Evans, University of
Toronto; Monica Campbell, Sarah Gingrich, Toronto Public
Health; Dave Stieb, Amanda Wheeler, Health Canada; Jeff
Brook, Environment Canada.
Particulate Matter (PM) is the general term used for a
mixture of solid particles and liquid droplets found in the
air and are produced by a wide variety of natural and
manmade sources. Recently, the U.S Environmental
Protection Agency announced to strengthen EPA's
previous daily fine particle standard by nearly 50 percent
from 65 to 35 micro gram per meter cube (EPA, 2006).
This standard increases protection of the public from
short-term exposure to fine particles. Such proactive
changes in standards indicate the increasing evidence of
its importance with regard to health. According to The
world health report 2002 indoor air pollution is
responsible for 2.7 percent of the global burden of disease
(http://www.who.int/indoorair/en/). Exposure to
contaminated indoor air has been identified as a
significant cause of health problems affecting the poor in
developing countries, especially women and younger
children. According to recent estimates in India, indoor
exposures to particulates appears to be responsible for
more than 7 percent of the national burden of disease
(CAI, Asia). The concentration of PM in the India is
being estimated with more rigour than before. The present
study attempts to monitor indoor and outdoor fine
particles in a control site in Mumbai city, India during
summer season for 10 days. The fine particles were
measured using AirMetrics MiniVol at the rate of 5 liters
per minute for 24 hours on a Teflon filter. The
concentration of fine particle in indoor area ranged
between 35-150 micro gram per meter cube and outdoor
area ranged between 20-106 micro gram per meter cube.
The indoor outdoors ratios were found to be in the range
of 1.2-3.1. The present study will discuss on indoor and
outdoor sources and their relationship.
Indoor concentrations of ultrafine (UFP) and fine (PM2.5)
particulate matter are becoming of increased health
concern. Skating arenas are unique indoor
microenvironments because the ice resurfacing machines
typically are powered by fossil fuels. The presence of a
combustion source in a location where individuals are
likely to be breathing more heavily than outdoors should
be of concern; however, relatively few studies have
focused on UFP and PM2.5 concentrations in arenas.
During the winter of 2006, 4 skating arenas were visited
on 3 occasions in downtown Toronto. Measurements
were collected outside the arena, in the spectator stands
and in the player's bench. Factors that contribute to UFP
and PM2.5 concentrations in the arenas were the type and
age of the ice resurfacing machines, the frequency of ice
resurfacing, the ventilation characteristics of the arena and
the spectator stands heaters.
Arena A was the only arena with a natural gas powered
ice resurfacing machine and the lowest concentrations of
UFP and PM2.5 were observed during flooding. In
contrast, Arena D had a 12 year old propane powered
machine, the most frequent ice flooding and the highest
concentrations of UFP and PM2.5. In Arena C, UFP
concentrations measured while the propane powered
edging machine was in operation were twice those
measured during flooding. With the exception of Arena C,
operators remove ice from the boards using a manual ice
breaker. With the exception of Arena D, all arenas had
large vents with fans in the ceiling. In Arena D, a manual
valve needed to be switched in order to open the vent
before flooding; however, this was not consistently done
thereby allowing UFP and PM2.5 concentrations to
accumulate. Finally, Arena A had natural gas powered
spectator stand heaters and the UFP concentration
increased by a factor of 4 while in operation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
7B.5
Size Characteristics of Airborne Particles and Bioaerosols in
Home Environments. QING CHEN, Lynn M. Hildemann,
Stanford University.
8B.1
Experimental Measurement Of Particle Resuspension From
A Tile Floor By Walking. MARK R. SIPPOLA, Richard G.
Sextro, Lawrence Berkeley National Laboratory.
Exposure to bioaerosols indoors may contribute to the
development of adverse health effects such as asthma.
Deposition of particles in the respiratory tract depends on
size. However, little information is available regarding the
size characteristics of indoor bioaerosols. Via biochemical
assay techniques, this study evaluated the size
characteristics of bioaerosols in home environments,
comparing them with outdoor air samples and house dust.
A room-scale experiment was conducted to quantify the
fraction of particles in the footprint area that are
resuspended with each step by walking people. Dry
polydisperse fluorescein powder was deposited onto the
3
pre-cleaned tile floor of a 24.7 m experimental room at
2
an initial loading of 85 mg/m . The airborne size
distribution during this deposition was measured by two
aerodynamic particle sizers (APS, TSI Inc, Model 3321)
and these measurements were used to estimate the size
distribution of particles on the floor. Two volunteers then
walked in the room at a rate of 100 steps per minute for
30 minutes while airborne particle concentrations were
measured by two APS units and filter samples. Particle
loss rates by deposition and ventilation were also
measured. The re-deposition of resuspended particles to
surfaces was quantified by deposition coupons on the
walls and floor. Fluorescent techniques were used to
quantify particle mass on air filters and deposition
coupons. Real-time APS concentration data and
measured particle loss rates were used in a mass-balance
model (which includes a decaying particle source from
resuspension) to calculate the fraction of particles
resuspended per step and the decay rate of particles
available for resuspension in the size range 0.5-10
microns.
The fluorescence measurements suggested that most of
the resuspended particle mass re-deposited to the room
floor. APS measurements suggested that 2.3 percent of
the total particle mass was resuspended in 30 minutes;
this estimate was independent of the fluorescence
measurements. The fraction of particles in the footprint
area that were resuspended with each footstep was in the
-5
-5
range 2x10 -4x10 per step and nearly independent of
particle size. The particle mass on the floor available for
resuspension was calculated to decrease 2-10 percent per
minute (depending on particle size) during the 30 minutes
of walking.
We measured airborne particle, protein, endotoxin, and (1
-3)-beta-D-glucan mass concentrations inside 10 singlefamily homes in northern California. Endotoxin was used
to track gram-negative bacterial concentrations, (1-3)beta-D-glucan to evaluate fungal levels, and protein to
reflect overall bioaerosol concentrations. Three size
ranges of particles (PM2.5, PM10, and TSP) were
collected in duplicate in each home on five occasions (9
-12 hrs/sample) over a 3-week period, including one night
and four days. Simultaneous outdoor samples were
collected for three of the sampling periods, along with
viable bacteria and fungi. We also analyzed dust from the
carpets and sofas in each home, and documented the
house characteristics and occupants' behavior.
Most of the mass concentration of both indoor and
outdoor particles and protein was in the fine fraction
(PM2.5), while the mass of endotoxin and (1-3)-beta-Dglucan was present mainly in the coarser fractions (PM10PM2.5 and TSP-PM10). A comparison of paired indoor
and outdoor concentrations revealed some significant
elevations (p<0.05, Wilcoxon signed-rank test) in the
indoor air samples, including protein in the coarser
fractions (PM10-PM2.5 and TSP-PM10), endotoxin in the
respirable fractions (PM2.5 and PM10-PM2.5), and
culturable bacteria. Indoor endotoxin levels in the TSPPM10 fraction also were somewhat elevated (p<0.10).
The coarser fractions of airborne protein and endotoxin
indoors were positively correlated with their levels in dust
samples (per gram of dust). Finally, endotoxin levels in
TSP were moderately correlated with viable bacteria
counts for both indoor samples and outdoor samples.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
8B.2
A Model for Resuspension of Particles due to Human
Walking including Electrostatic Effects. XINYU ZHANG,
Jing Qian, Goodarz Ahmadi, Andrea Ferro, Clarkson University.
A model for resuspension of particles with electrostatic
effect due to human walking was developed. The foot
stepping down and up process was treated as the motion
of an effective circular disk toward or away from a
stationary surface. The airflow generated from this
squeezing film was assumed laminar and the
corresponding gas velocity was evaluated. The squeezing
flow outside the foot range was evaluated based on radial
wall jet theory. The surface roughness was included in
the analysis. For particle detachment, the adhesion force,
the electrostatic force and drag forces were taken into
account. The particle cloud deposition, diffusion and
transport were included in the analysis. The PM
concentrations for different particle sizes due to human
walking were evaluated. The effect of various factors
affecting the resuspension process was discussed. The
model predictions were compared with the experimental
data. The results show that shoe bottom roughness, foot
size as well as foot stepping down and up velocities and
frequencies can affect the PM concentrations.
8B.3
Measurement of Ultrafine Particles Generated by Indoor
Combustion and Electric Appliances. FANG WANG, Harbin
Institute of Technology, Harbin, China; Lance Wallace, Cynthia
Howard-Reed, National Institute of Standards and Technology.
Several studies have reported the concentration of
ultrafine particles indoors due to sources such as
combustion and electric appliances. These previous
studies, however, have only measured ultrafine
concentrations for particles as small as 10 nm. The
advancement of particle measurement technology now
makes it possible to measure particles as small as 2 nm.
As a result, the National Institute of Standards and
Technology is conducting a study to measure the source
strengths of several indoor combustion and electric
appliances to include counts of particles from 2 nm to 64
nm. Experiments were conducted in an unoccupied
manufactured house equipped to semi-continuously
measure air change rates, carbon monoxide levels, gas/
electricity usage, environmental conditions (e.g., indoor/
outdoor pressure differences, temperature, relative
humidity, etc.), and local weather conditions. Ultrafine
particles were measured in multiple locations every 2.5
min to 5 min with a scanning mobility particle sizer
equipped with a nano-differential mobility analyzer.
Ultrafine particle sources included a gas stove, hair dryer,
electric toaster, and electric heater . Size distributions
were measured in two rooms, a source room (kitchen) and
a receptor room (master bedroom). Preliminary results for
the gas stove show the peak concentration occurring at a
particle size of approximately 5 nm to 8 nm and total
concentrations to be about 10 times greater than reported
in previous studies of particles greater than 10 nm. These
results suggest that ultrafine number concentrations
previously reported for combustion appliances, and
perhaps electric appliances, may be significantly
underestimated.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Indoor Aerosols
2007 AAAR Annual Conference Abstracts
8B.4
Secondary organic aerosol from ozone-initiated reactions
with terpene-rich household products. BEVERLY K.
COLEMAN, William W Nazaroff, University of California,
Berkeley; Melissa M. Lunden, Hugo Destaillats, Lawrence
Berkeley National Laboratory.
Household cleaning products can contain high levels
of terpenes that may be oxidized indoors by ozone
transported from outside, resulting in formation of
secondary organic aerosol (SOA). Characterizing the
particles formed from these reactions is necessary to
understand the possible health effects of these aerosols.
An analysis was performed on SOA data from a series of
small chamber experiments where ozone and terpene-rich
household product vapors were reacted at conditions
similar to those for typical product use indoors.
Experimental details are presented in Destaillats et al.
(Environmental Science & Technology 40, 4421, 2006).
Cleaning product vapor and later ozone were introduced
into a 198 L chamber at steady levels. Consistently, at the
time of ozone introduction, a nucleation event occurred
that exhibited behavior similar to atmospheric nucleation
events.
8B.5
SOA formation and growth from ozononlysis of terpene in
indoor environments. XI CHEN and Philip K. Hopke,
CClarkson University.
It has been suggested that secondary organic aerosol
(SOA) can form in indoor air from infiltrated ozone and
indoor reactive volatile organic compounds. In order to
provide the basis for a particle nucleation and growth
model to estimate SOA formation at various combinations
of alpha-pinene and ozone concentrations for typical
ventilated indoor environments, an experimental study of
particle formation was conducted. In addition the
production of reactive oxygen species (ROS) was also
examined. Experiments were conducted using a 2.4 m3
stainless steel chamber with steady-state conditions. This
study describes kinetics of the SOA and ROS formation
initiated from ozonolysis of terpene and contributes to the
understanding of indoor SOA formation mechanisms. In
addition to the basic study of SOA formation, the
influence of major inorganic chemical species such as
ammonia and NOx present in indoor air on particle
formation and growth was also studied.
SOA was measured with a scanning mobility particle
sizer (SMPS, 10 to 400 nm) in every experiment and with
an optical particle counter (OPC, 0.1 to 2.0 micro-meter)
in a subset of experiments. In order to fully characterize
the aerosol size distribution, we aligned the OPC and
SMPS measurements in the overlapping size range of the
two instruments. Effective bin bounds for OPC
measurements change according to the composition of the
aerosol, and this change is related to the refractive index
of the aerosol. A model was developed to determine a
representative refractive index for the cleaning product
aerosol. The results from experiments where the entire
distribution could be measured were used to infer
information for the experiments where only part of the
distribution was measured. The number and mass
distribution as a function of time for each experimental
condition was determined using the measured and
modeled data. The effects of environmental conditions,
such as ozone level, product formulation, air exchange
rate, relative humidity, and preexisting particles, on
particle distribution characteristics were explored.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2H.1
Laboratory Studies of Inhaled Simulated Downwind
Components of Coal Combustion Emissions. JAKE
MCDONALD, Matthew D. Reed, Matthew Campen, JeanClare
Seagrave, Joe L. Mauderly, Lovelace Respiratory Research
Institute.
A study of major components of coal combustion
emissions encountered in downwind exposures was
conducted. The target exposure atmosphere composition
was set by consensus from a workshop of industry,
government, and academic experts. Pulverized Powder
river Basin (PRB) sub-bituminous coal was aerosolized
and combusted in an electric \drop-tube\ furnace,
emissions were cooled, a cyclone was used to establish an
upper bound particle size, and the effluent was mixed
with sulfate aerosol generated by a vaporizationcondensation process. This stream was supplemented
with sulfur dioxide and nitrogen oxides to achieve the
target mixture. Animals were exposed 6 hr/day, 7 days/
wk to the mixture generated from PRB coal, at dilutions
containing 100, 300, or 1000 micro-gram particulate
matter/m3, to the highest concentration with particles
removed by filtration, or to clean air as controls.
Evaluations of health effects included body and organ
weights, histopathology, lung inflammation, resistance to
bacterial respiratory infection, development and
exacerbation of respiratory allergic responses,
electrocardiogram, pre-atherosclerotic changes in blood
vessels, and pre-cancer changes in DNA and
chromosomes. Atmosphere composition, and initial
biological findings, will be reported. This research is
supported by funding from 16 government and industry
sponsors, including the Environmental Protection Agency
(CR831455-01-0), the Department of Energy National
Energy Technology Laboratory (DE-FC26-05NT42304),
the Electric Power Research Institute (EP-P17972/
C8861), and Southern Company.
2H.2
Airborne Mycobacterium Tuberculosis Profile in A Hospital
After An Outbreak of Tuberculosis. Pei-Shih Chen, TAIWEI CHEN, Kaohsiung Medical University.
An outbreak of tuberculosis among eight medical
personnel was happened in a hospital in Taiwan in midAugust of 2005. Therefore, the main purpose of this
study was assessing the airborne Mycobacterium
tuberculosis profiles in this hospital to identify the high
risk area. In addition, the difference of airborne
Mycobacterium tuberculosis concentration before and
after ventilation improvement was also evaluated.
A total of 192 air samples were taken from negative
pressure isolation wards, medical wards, waiting rooms
and consulting rooms of medical department and pediatric
department in the period of December 2005 to July 2006.
The concentration of airborne M. tuberculosis was
quantitatively determined by real time qPCR. In addition,
cultivable airborne bacteria and fungus, temperature, and
relative humidity were also measured.
All positive samples were in the wards of chest and
infectious disease division in internal medicine
department. The airborne M. tuberculosis concentration
was in the range of 54 copy-cubic meter to 1109 copycubic meter. The highest concentration was found in the
nursing station of chest division. After improvement of
ventilation system, no M. tuberculosis was detected in the
air. In addition, the airborne bacteria concentrations were
also declining after the improvement. Our results showed
that the improvement of air conditioning may reduce the
risk of M. tuberculosis exposure. In addition, it was
found a good correlation between M. tuberculosis and
airborne bacteria.
In regard to the culturable bacteria and fungi
concentration in the air, 122 samples were analyzed in the
hospital. According to the Indoor Air Quality
Recommended Values of Taiwan Environmental
Protection Administration, the failure rate was 64 percent
and 8 percent for bacteria and fungi, respectively. In
addition, the airborne bacteria concentrations in the
nursing station of chest division were all higher than the
recommended values, even after the improvement of
ventilation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2H.3
Airborne Influenza and Avian Influenza Viruses from Long
Term Transportation and Its Health Effect. Pei-Shih Chen,
Qian Kun Lin, FENG-DA TSAI, Kaohsiung Medical University.
Due to desertification of northwestern China and
Mongolia, the frequency and intensity of Eastern Asia
dust storm has become higher and higher. This
phenomenon has a great effect to Taiwan, the downstream
region of Eastern Asia dust storm. Previous studies
showed that airborne fungi might be transported by dust
storm. Therefore, the main purpose of the present study is
to investigate the concentration airborne influenza and
avian influenza viruses during suspected dust storm
events and on normal days. In addition, the correlations
between influenza virus concentrations and hospitalized
admissions were also studied.
Air samples were collected at Sin-Jhuang and Shi-Men
during dust storm events and on normal days and then
analyzed by Real-time qPCR. Our results showed that the
positive rate and concentration of influenza virus at both
sampling sites were higher during dust storm events than
those on normal days. In addition, the associations
between influenza A virus concentration and hospitalized
admissions of influenza were prominent three weeks after
the event (r = 0.76, p = 0.0001). In regard to the climate
factors, the concentrations of influenza A virus have an
negative correlation (r = -0.41, p = 0.0006) with
temperature and a low correlation (r = 0.30, p = 0.014)
with rainfall. Furthermore, the peaks of virus
concentration were found earlier than the peaks of PM2.5'
PM10. In conclusion, influenza viruses might possibly
transported by airstreams for a long distance.
2H.4
Environmental Monitoring of Virus-containing aerosols
around Children with Infections. CHUN-CHIEH TSENG,
Chih-Shan Li, College of Public Health, National Taiwan
University; Luan-Yin Chang, National Taiwan University
Hospital.
Children are more vulnerable to viral infections than the
general population. For understanding the mode of viral
transmission, viral bioaerosols were collected by filtration
method in the emergency room and the outpatient clinics
of the pediatrics department at a medical center in Taipei.
In this study, real-time quantitative polymerase chain
reaction (real-time qPCR) was performed to detect
influenza A virus (INFAV), human adenovirus (HAdV),
and enterovirus. Among the 33 aerosol samples in the
emergency room, the positive rate was 24% (8/33) for
INFAV, 36% (12/33) for HAdV and 15% (5/33) for
enterovirus; the results suggested that these viruses could
be transmitted through droplets. In summary, this is the
first report describing the filter/real-time qPCR can detect
and quantify virus associated exposure in air. It suggests
that this technique can provide further insight into
hospital epidemiology and infection control, as well as
viral transmissibility.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2H.5
Capturing the Exhaled Protein Aerosol: Evaluation of
Rodent-Based Systems. OWEN MOSS, Earl Tewksbury,
David Nash, The Hamner Institutes for Health Sciences.
The utilization of exhaled proteins as biomarkers for
disease is a relatively recent advance in need of
validation; validation with breath condensate from animal
models of pulmonary disease. For pooled breath-samples
from 50 mice or rats we previously showed that, by taking
advantage of the small volume and short path-length of
the central column, the operation of the Cannon Nose
Only exposure system (Lab Products, Seaford, DE) can be
modified to allow efficient collection of breath
condensate. The modification consisted of utilizing zerohumidity air, and reversing the airflows in the system. By
using this same modification, we have evaluated the
application of another nose only exposure system; the
Vaccine (Nose Only Exposure) Unit -- first working
version (CH Technologies (USA) Inc., Westwood, NJ: -In this system, the shortest path from the animal port to an
exit resembles a narrow cylindrical channel.). For flow
rates similar to the minute ventilation of mice, we
measured penetration from the animal port for two cases:
(1) an aerosol of 15 nm diameter spheres; and (2) air at 90
% RH and 36 degrees centigrade. Nanospheres were
generated with a Graphite Aerosol Generator (GFG-1000,
Palas, Karlsruhe, Germany), and detected with a Scanning
Mobility Particle Sizer (SMPS 3936, TSI, St. Paul, MN).
Condensate was collected with a three-stage cold trap. For
breath collection the Vaccine Unit functioned similarly to
the Cannon System: -- greater than 90% penetration of
nanoparticles or water vapor. The impact of system
configuration appears to be minor; as long as the dew
point of the exhaled air is rapidly dropped to below the
temperature of the system walls.
2H.6
A Web-Based Interactive Aerosol Program for
Undergraduate Education-Aerosols in the Health Care
Field. YU-MEI HSU, Chang-Yu Wu, Anne Donnelly,
University of Florida; Paul Stephan, Santa Fe Community
College; Pratim Biswas, Washington University in St. Louis.
Having adequate knowledge of the penetration, deposition
and site of action of aerosols in the health care field must
be ongoing since new medications and techniques of
delivery are changing rapidly. To facilitate the teaching
of this subject matter to community college students,
University of Florida and Santa Fe Community College
have collaborated together on a project to develop a webbased program. The goal is to increase students' retention
and understanding of the physical properties of aerosols
by active participation and visual modelling of content
that is otherwise difficult to grasp.
Three types of aerosol generators commonly used in the
health care field are included: MDI (Metered-Dose
Inhaler), DPI (Dry Powder Inhaler) and SVN (SmallVolume Nebulizer). In addition to the principles,
functions, advantages/disadvantages of each type of
device, the program presents the advantages and
disadvantages of administering medications via aerosols.
The site of penetration and deposition, as well as methods
for determining clinical relevance and assessment of
patient outcomes and lung retention are included. The
program is available for public access at http://
aerosol_beta.ees.ufl.edu/.
In order to maximize the ease of use and interactivity of
the learning environment, various web and scripting
technologies are used to implement the user interfaces,
including HTML and Flash. HTML was used to deliver
normal static web pages. Macromedia Flash is a platform
for easily developing interactive multimedia animation
and audio presentations for the web and is used to create
highly engaging and active content. To assess the
effectiveness of the program, formative and summative
evaluations will be conducted using on-line surveytechniques for community college students.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2H.7
Improvement of Particle-Mediated Gene Transfer
Technology. CHIH-CHIEH CHEN, Sheng-Hsiu Huang, WeiShun Lin, College of Public Health, National Taiwan
University; Yu-Mei Kuo, Chung Hwa College of Medical
Technology.
The ways of gene transfer can be classified into three
categories: biological method (such as virus-mediated
transduction), chemical method (such as calciumphosphate mediated, DEAE - dextran mediated, liposomemediated method), and physical method (such as
microinjection, electroporation, and biolistics (gene gun)).
Among them, gene gun has many advantages: low
harmfulness to cell, small amounts of DNA required,
short reaction time, easy to operate, and applicable to both
in vitro and in vivo transformation. The objective of this
work was to enhance the uniformity and the coverage of
the gene-laden gold particles, which, as illustrated below,
resulted in better gene transfection efficiency.
In this research, the effect of gene gun configuration on
the collection efficiency and uniformity of deposited
particles was investigated. The operation parameters
include: slot width (0.05, 0.1, 0.5, 5, 20 mm), slot height
(0, 15, 25 mm), helium pressure (100, 150, 250, 400, 600
psi), transition cone (5, 15, 35, 65, 105, 155 mm),
diameter of target filter (10, 15, 25, 30, 50 mm) and
column height (25, 35, 45, 65, 95, 155 mm). The greatest
transfection efficiency occurs when the gene gun is
modified to have a slot width of 0.05 mm, slot height of
0.05 mm, transition cone of 65 mm and diameter of target
filter of 50 mm. This new gene gun performs better than
the original design. Collection efficiency, uniformity, and
gene transfection are 5.2, 2.5 and 17.3 times higher than
original gene gun.
2H.8
Use of a Non-Pathogenic Viral Model for Quantitative PCR
Analysis of Artificially Produced Airborne Viruses.
DANIEL VERREAULT, Sylvain Moineau, Caroline Duchaine,
Universit
Infectivity assessment of viral material has, for
many years, been the main method in characterizing
airborne viruses. However the outcome of this type of
study is dependent upon the preservation of the virus
infectivity, which can be influenced by many factors
including surrounding environmental conditions as well
as nebulization and sampling stresses. There is a need to
independently measure the absolute number of viral
agents to provide a basis for comparison. Hence,
analytical techniques independent of viral culture should
also be used to properly investigate samples of artificially
and naturally produced viral aerosols. Advances in
molecular technologies have led to the development of
nucleic acid-based assays for the detection of viruses.
Quantitative Polymerase Chain Reaction (qPCR), can be
used for fast and quantitative analysis of viruses in
various samples. In this study, a dual-labeled probe was
used for qPCR analysis of virus-containing air samples
collected from a nebulization chamber. The samples were
collected at a rate of two liters of air per minute with two
types of filters, polycarbonate and PTFE, mounted on 37
mm 3-piece cassettes. An aerodynamic particle sizer
(APS) measured the particle median diameter as 0.8
micrometers. The viral model was phage phiX174, a 25 to
27 nanometer non-enveloped bacteriophage of
Escherichia coli, with a nucleic acid composed of a
circular single stranded DNA molecule. The simple
structure and composition of this virus allowed the
development of a simple single step qPCR protocol,
which consisted of using a diluted aliquot of the sampled
material without the need for DNA isolation. This latter
unnecessary step was found to reduce efficiency in
quantification of this virus. Results from this study show
no statistical difference in DNA recovery from both
polycarbonate and PTFE filters. We therefore suggest
that, for equal qPCR results, filter choice should be based
on culture.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2H.9
Characteristics of Atmospheric Bioaerosols by
Fluorochrome. MIAO-CHING CHI, Chih-Shan Li,National
Taiwan University.
Total concentration and viability of bioaerosols in the
ambient atmosphere were monitored by using
epifluorescence microscopy with fluorochrome (EFM/FL)
with five fluorescent dyes (AO, DAPI, SYTO-13, PI, and
YOPRO-1). The correlation of bioaerosols with
meteorological factors and pollutants was simultaneously
assessed. Results from EFM/FL were then compared
with those using a commonly used culture method. The
total microbial cell concentration measured by the non5
3
culture-based methods averaged about 8 x 10 cells/m .
However, culture method underestimated bioaerosol
concentrations by the factor of 100 to 1000. The average
viabilities were 0.66 by EFM/FL with PI staining, 0.35
with YOPRO-1 staining, and 0.0012 by the culture
method. The viability by EFM/FL was much higher than
the culturability. In summary, the total microbial cell
concentration and viability were highly underestimated by
the culture method. Moreover, based on culture and nonculture methods results, the total bioaerosol
concentrations could be strongly correlated as a result of
the temperature, rainfall, and UV light influence.
However, there were weak correlations between
bioaerosol concentrations and air pollutants. In
conclusion, EFM/FL methods could effectively assess the
total microbial cell concentration and viability of
bioaerosols in atmospheric samples.
2I.1
Using a Human Airway Cast for Deposition Studies of
Inhaled Medicine. YUE ZHOU, Clinton M. Irvin, Steven A.
Belinsky, and Yung-Sung Cheng, Lovelace Respiratory
Research Institute.
Inhalation drug delivery is considerably more complex
than other drug delivery routes because drugs must be
delivered to the specific target tissues in the lung
appropriate to the diseases. To evaluate the deposition
pattern of an inhalation medicine, a cascade impactor is
generally used to measure the aerosol size distribution
generated from medical devices. From the size
distribution, the deposition pattern in different lung
regions can be calculated. An appropriate device for the
specific medicine can be selected utilizing the lung
deposition information. However, many factors such as
lung geometry and drug formulation may interfere in the
drug deposition patterns in human. Hollow human airway
replicas are an alternative means by which one may
evaluate the deposition patterns of an inhaled medicine. In
this study, a DNA methyltransferase inhibitor, 5azacytidine, was nebulized by three different types of
nebulizers: MicroMist, SideStream, and LC Plus. The
particle size distribution was measured using a Next
Generation Pharmaceutical Impactor (NGI) at a flow rate
of 30 L/min. Three concentrations of the drug were tested
to see the differences in the size distribution. The drug
deposition patterns were calculated with the NCRP
(Nation Council on Radiation Protection and
Measurements) model. The LC Plus nebulizer was
selected as the optimized device for further study,
delivering the drug in a human lung replica. The replica
includes the oral cavity, oropharynx, larynx, trachea, and
four generations of bronchi. Three flow rates (15, 30, and
60 L/min) were tested to simulate human breathing
patterns. The deposition efficiency at the oral region was
compared with the deposition calculated by the NCRP
model and obtained a good agreement. The NCRP model
also can predict the deposition efficiencies in the
tracheobronchial and pulmonary regions.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2I.2
Turbulence Modeling in the Human Nasal Cavity. KEVIN
T. SHANLEY, Goodarz Ahmadi, Clarkson University.
The extreme complexity of the human nasal cavity makes
finding an accurate Reynolds number difficult. Numerous
experimental studies have identified the flow as being
largely laminar with some regions of recirculation. The
recirculation regions may not be properly represented
with the standard laminar flow model. Deposition of
nano-particles may be affected significantly by
recirculation. This work makes a comparison of the
different turbulence models for predicting the airflow in
the nasal passage. MRI scans of an anonymous, adult
male, human subject were used to construct a threedimensional volume. Airflow was simulated for breathing
rates corresponding with low to moderate activity (5 to 15
L/min). Comparisons are made between the standard kepsilon model, enhanced k-epsilon model, and the
Reynolds Stress model. A uniform concentration of
particles ranging from 10 nm to 100 nm were injected at
the nostril and tracked by a Lagrangian Discrete Phase
Model. Comparisons are made between the deposition
across models, as well as, with previously published
results for laminar flow.
2I.3
Numerical Simulations of Inertial Particle Deposition in a
Realistic Nasal Cavity. KEVIN SHANLEY, Parsa
Zamankhan, Goodarz Ahmadi, Philip K. Hopke, Clarkson
University; Young-Sung Cheng, Lovelace Respiratory Research
Institute.
The nasal valves, anterior to the main airway, are believed
to be an efficient collector of aerosols with aerodynamic
diameter larger than 1-micron. This work focuses on
developing a numerical model for predicting aerosol
deposition in the human nasal cavity under low to
moderate breathing conditions. MRI scans of a healthy
adult male human were used to construct the threedimensional volume of the nasal passage. The
computational volume was discretized into 965,000
tetrahedral elements and 250,000 computational nodes
and used in the computational model. The commercially
available software FLUENT was used to perform
computational fluid dynamics analysis and particle
tracking. The Lagrangian particle tracking approach was
used for analyzing the particle trajectories in the nasal
passage. The predicted capture efficiencies are shown to
correlate with the impaction parameter and have good
agreement with published experimental results. An
empirical expression for the capture efficiency is also
proposed. The affects of the direction of gravitational
acceleration is also studied.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2I.4
Deposition of Fiber and Spherical Aerosols in the Human
Tracheobronchial Airway. YUNG SUNG CHENG, Yue
Zhou, Wei-chung Su, Lovelace Respiratory Research Institute.
2I.5
Improved Conversion Scheme for Eulerian Deposition
Probability Rates. Mohammad I. Rahman, CARLOS F.
LANGE, University of Alberta.
Inhalation exposure of fiber aerosol may have serious
health consequences including lung cancers. The
deposition pattern in the respiratory tract as a function of
fiber dimensions is the information critical to
understanding respiratory dosimetry and etiology.
Controlled studies of fiber deposition in human volunteers
are not available because of ethical concerns. The
purpose of this study is to investigate the effects of fiber
dimension and breathing rate on the deposition pattern in
an oral/tracheobronchial airway cast with a defined
geometry. Two types of fibers including a carbon fiber
and a glass fiber were used for the deposition study. The
fiber was generated with a small-scale powder disperser
(Model 3433, TSI Inc., St Paul, MN). Regional fiber
deposition pattern was measured at a constant inspiratory
flow rate of 7.5,15, 30 and 43.5 L min-1. As a
comparison we also did deposition experiments in the
same cast using polystyrene latex (PSL) test particles
tagged with fluorescent. Fiber depositions in different
sections of the airway cast and the backup filter were
extracted and prepared for optical microscopy and image
analysis. From the counting data, deposition efficiency as
a function of fiber length /diameter was calculated. Our
experimental data of fiber deposition in the
tracheobronchial region show large variability but
generally agree with the numerical simulation results
published by Zhang et al. (1996). The deposition
efficiency can be expressed as a function of Stokes
diameter, Reynolds number and branching angle. We
also show that deposition efficiencies of spherical
particles are higher than those of fibers at the same
impaction parameter. This can be explained by the
orientation of fibers, which tend to align with the flow
direction. This information is useful in predicting the
deposited dose of inhaled fiber particles. (This research
was supported by the NIOSH under the Grant 1R01
OH03900).
Eulerian models of aerosol lung deposition typically
employ deposition probability functions derived from
Lagrangian analysis. The Lagrangian deposition functions
are either empirical or derived under steady state
conditions. In these models, the deposition rate is
calculated by first estimating the particles deposition per
unit time for a whole lung generation, simply dividing the
total deposited amount over a certain period by this time
period. Then the total deposition rate is distributed,
dividing it by the length of the generation.
The main advantage of Eulerian deposition models is
treatment of the unsteady deposition. To capture the true
effect of the breathing pattern and to predict the transient
local deposition, deposition functions based on
instantaneous deposition rate are required. But, at present,
no such deposition function is available. Development of
such analytical expressions involves complex
mathematics and requires sophisticated experimental
facilities for the development of empirical correlations,
both of which are time consuming.
An attractive alternative is to develop a numerical
technique to convert the generational deposition
probabilities into an Eulerian frame of reference valid for
each control volume within the domain. In this study, two
such conversion schemes, one based on Generation
Residence Time (GRT) and the on other on Control
Volume Residence Time (CVRT), were investigated, and
then used to estimate total and regional lung deposition
under several clinically important cases. Both schemes
showed excellent agreement with experimental aerosol
deposition data in the human respiratory tract, with CVRT
giving slightly better results.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2I.6
Prediction Of Particle Deposition In An Expanding Alveolar
Model. JESSICA M. OAKES, Risa J. Robinson, Rochester
Institute of Technology.
2I.7
Inhalability of particles and fibers in the human lung.
BAHMAN ASGHARIAN, CIIT at the Hamner Institutes for
Health Sciences.
In order to fully understand particle deposition in the most distal
airways of the lung, several parameters need to be considered;
specifically the mixing between the tidal and residual air that
occurs due to expansion and contraction of the alveolar walls.
The mixing induced by the moving walls allow for the diffusive
length scales to change with time (Tsuda et al., 2002)). When
the lateral length scale and diffusive length scales are equal there
is a sudden increase in entropy and therefore a decrease in the
time it takes for a particle to diffuse (Butler and Tsuda, 1997)).
Molecules that are breathed in, such as oxygen, have a high
diffusivity, and therefore the time that is required to diffusive is
small and insignificant. Particles such as aerosols have a much
lower diffusivity, therefore the time required for the particles to
diffusive from the tidal to the residual air is much higher. In
order for the aerosol particles to deposit on alveolar walls,
mixing must occur, which allows for the particles to travel to the
alveolar surface.
Particle inhalability is the fraction of particles in the
inhaled air that enter the extrathoracic airways.
Knowledge of airborne material inhalability is needed for
accurate assessment of lung internal dose. Models of
inhalability of particles and fibers for indoor
environments were studied based on variables controlling
movement of particles in the air. Non-dimensionalization
of the transport equations under calm conditions (no
wind) showed that inhalability depends on particle
aerodynamic diameter and inhalation flow rate. A model
of inhalability as a function of particle diameter and flow
rate was constructed. Parameters in the model were
estimated by fitting the model to available inhalability
measurements for oral and nasal breathing. The
inhalability model for spherical particles was extended to
fibers by including the effects of fiber orientation and
viscous drag. A diameter for fibers was obtained using the
equation where is related to fiber Stokes diameters, and
and are fiber minor axis and aspect ratio respectively.
Fiber diameter can be replaced in the inhalability
expression for spherical particle diameter to predict fiber
inhalability. Fibers were predicted to have a lower
inhalability than spherical particles of the same mass. The
influence of breathing rate on inhalability for both fibers
and spherical particles was more pronounced for oral
breathing than for nasal breathing. While fibers have a
lower tendency during inhalation to enter the
extrathoracic airways and reach lung airways, the
elongated shape of fibers leads to slower lung clearance.
Therefore, fiber inhalability must be considered when
studying their carcinogenic effects.
Previous alveolar models have been created in order to predict
particle deposition. Darquenne (2001) conducted research on a
two dimensional static model, in which the only forces acting on
the particle were gravitational and drag. The model did not take
into consideration the mixing that occurs due to expansion and
contraction. An expanding numerical model was created by
Tsuda et al. (1995) which proved that expansion and contraction
is significant in predicting particle deposition.
In the current study a computational fluid dynamic (CFD) model
was created based off of Weibel’s 23rd lung generation. The
CFD model coupled the Navier – Stokes equation, a moving
mesh application, and particle dynamics in order to visualize the
fluid flow and predict particle deposition. The model was
simulated with varying frequencies in order to represent a range
of breathing cycles. Particle deposition was determined using
two techniques; particle deposition due to diffusion acting alone
in a static model and particle deposition due to mixing and
diffusion. The data collected from the model was compared to
analytical, numerical, and in vivo experimental data found in
literature. Based on data collected from the models it was
concluded that mixing in the acinus of the lungs is critical in
determining particle deposition.
Butler JP and Tsuda A. Effect of convective stretching and
folding on aerosol mixing deep in the lung, assessed by
approximate entropy. J.Appl.Physiol. 83: 3: 800-809, 1997.
Darquenne C. A realistic two-dimensional model of aerosol
transport and deposition in the alveolar zone of the human lung.
32: 10: 1161-1174, 2001.
Tsuda, A Henry, F S Butler,J P. Chaotic mixing of alveolated
duct flow in rhythmically expanding pulmonary acinus. 79:
1055, 1995.
Copyright
В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
2I.8
3D Reconstruction of a Female Upper Respiratory using the
Visible Human Data Set to Predict Cigarette Smoke Particle
Deposition JACKIE RUSSO, Risa Robinson, Dept. of
Mechanical Engineering, Rochester Institute of Technolgy.
2I.9
Puff Profile Simulator for Tobacco Smoke Particle Diameter
and Mass Measurement. JOHN McAUGHEY, British
American Tobacco; Barrie Frost, Consultant; Kingsley Reavell,
Colin Dailly, Cambustion.
The goal of this research was to create a 3D lung model
from 2D medical images to generate a more realistic
geometry for CFD simulations. The lung geometry was
created in 3D Doctor (Able Software Corp, 2006) based
on the cryosectioned images from the Visible Female
Dataset as part of the NIH Visible Human Project. The
lung model consists of 4-5 generations of airways. The
model morphometry was compared to several accepted
lung morphometries from the literature and it was found
that the Visible Female closely correlates with dimensions
given by Horsfield and Cumming (1971). The lung
model was attached to a larynx based on medical
illustrations and to two separate scanned impression of an
oral cavity. The first oral cavity represents normal
breathing while the other represents the oral cavity during
the inhalation of a cigarette. Commercial CFD software
was used to simulate breathing during a realistic smoking
cycle– puff, inhale bolus followed by clean air, which was
compared to normal breathing.
This paper describes a system for cigarette testing
measuring fresh TPM (tar particulate material) mass,
median particle diameter, and particle number
concentration, with any desired flow profile. This allows
measurements at conditions representative of human
smoking or for regulatory pre-defined machine smoking
profiles. The data are collected at 10 Hz time resolution
with cumulative number and TPM mass measurement in
real time on a puff by puff basis.
The system is designed to be used with real-time aerosol
instruments such as DMS-type fast electrical mobility
spectrometers to provide continuous measurement of the
aerosol inhaled from the cigarette during smoking. The
flow through the cigarette is metered with an orificepressure-drop type flow sensor and controlled to follow a
specified profile at 12.5 Hz. To follow highly dynamic
puff profiles a feed-forward type controller is used. The
complete smoking of a cigarette with a different profile
for each puff can be reproduced.
The flow drawn through the cigarette is diluted with
filtered air close to the filter holder to halt agglomeration
processes. The system operates with a constant total
diluted flow to minimise errors in the measurement of
total mass emissions from the cigarette. A dilution ratio
signal is provided to allow calculation of the undiluted
concentrations if desired.
The system has been tested with standard machine
profiles and those measured from human smokers.
Control of cigarette flows down to approximately 1 ml/s
is possible, with a dynamic range of at least 30:1. The
typical error in the integrated volume of a puff is around
1%.
Puff by puff were measurements carried out on a series of
1- and 4-mg yielding products using a 35 ml puff of 2 s
duration every 60 s, using an ISO puff profile, and
normalised to 7 puffs. These data were compared on a
puff by puff basis with gravimetric measurements.
Count median diameters (CMD) were measured puff by
puff and ranged from 163 - 247 nm, depending on the
puff number and yield of each puff. Individual puffs were
measured from 0.18 to 1.05 mg TPM and correlated well
with the equivalent gravimetric data (r2 = 0.88).
Horsfield, K. and G. Cumming. Models of the Human
Bronchial Tree. Journal of Applied Phsyiology. 31:207
-217, 1971.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
3B.1
Molecular Source Tracking of Bioaerosols in the
Quarantined Katrina Flood Zone. MARI RODRIGUEZHERNANDEZ, Jeffrey Walker, Norm Pace, Mark Hernandez,
University of Colorado Boulder.
The damage Hurricane Katrina caused to the city of New
Orleans resulted in massive clean up efforts not seen since
the September 2001 attacks. Remediation efforts included
large scale pumping and aeration operations to reduce
floodwater contaminant loads prior to their diversion into
the Lake Pontchartrain canals. These remediation efforts
resulted in enhanced aerosolization of microorganisms,
which presented a potential inhalation hazard to
emergency response personnel working in the vicinity.
The objectives of this study were to identify and
enumerate airborne microorganisms associated with
remediation efforts, determine whether the observed
aerosol ecology was associated with proximal
floodwaters, and to ascertain if microbes present in the air
posed a potential health risk to emergency response
personnel working in the immediate vicinity of water
aerosolizing operations. To achieve these objectives,
direct microscopy, broad spectrum PCR, and DNA
sequencing analysis were conducted on air samples and
adjacent water samples throughout flooded New Orleans.
Widely-accepted phylogenetic analysis was used to
quantitatively assess the relatedness between airborne and
waterborne microbial communities. We report here that
total bioaerosol loads near floodwater remediation
operations were approximately 20 times greater than
those observed in comparable outdoor environments.
Phylogenetic observations suggest that there was no
obvious relationship between the microbial ecology found
in (local) composite New Orleans Parish air samples, and
the ecology present in adjacent floodwaters. These results
also suggest that the aerosol sequences observed in this
study may be part of a large scale, inter-regional
bioaerosol community. Potential pathogens identified
were not associated with any respiratory illnesses, and
would most likely only affect immunocompromised
individuals. Regardless of airborne pathogen exposure
assessment, these methods and observations contribute to
a small, but growing genetic catalogue of airborne
microorganisms in the outdoor environment.
3B.2
Airborne Aspergillus Particles in a Hospital: Effects of
Construction and other Potential Factors. MARIAN D.
GOEBES, Lynn Hildemann, Stanford University.
Aspergillus is a common genus of mold that can grow
indoors, and that includes several species capable of
causing fatal pulmonary infections in
immunocompromised individuals. A longitudinal study
investigating concentration fluctuations of airborne
Aspergillus particles was conducted in a hospital
undergoing renovations in one portion of the building.
The remainder of the hospital continued to function as
usual, including several specialty clinics serving
immunocompromised patients.
Size-segregated samples of airborne particles were
collected before, during, and after construction, using
filters downstream of cyclones. Aspergillus particles were
quantified with quantitative Polymerase Chain Reaction
(qPCR), a DNA-based method. Results of the year-long
study suggest that construction did not cause elevated
concentrations of Aspergillus particles in the parts of the
hospital that remained open, both because large
concentrations of Aspergillus were generally not
generated, and because the containment system used for
the construction area appeared to be effective.
However, in one lounge of the hospital, concentration
fluctuations greater than an order of magnitude that
appeared unrelated to construction were observed.
Intensive air sampling campaigns were conducted in late
fall and late spring, coupled with observations of
activities and conditions in this lounge, to investigate
potential sources of Aspergillus particles, including
diurnal variations in temperature and relative humidity of
the lounge, the outdoor Aspergillus particulate
concentrations, and foot traffic through the lounge.
Correlations indicate that Aspergillus particulate
concentrations are strongly influenced by foot traffic in
this carpeted area. The behavior of Aspergillus particles
relative to levels of particulate matter <= 5 um (PM5) was
also examined.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
3B.3
Effect of Protein Loading on Particle Size, Density and
Shape. PATRICIA FRITZ, Lupita Montoya, Rensselaer
Poytechnic Institute; Daniel Hershey, New York State
Department of Environmental Conservation.
The allergenicity of common indoor allergens is often
linked to source specific proteins, such as those
originating from dust mite (e.g., Der p 1) and cat (e.g., Fel
d 1). Allergenic proteins can be present in particles
derived directly from the allergenic source, or can be
carried on available non-specific particles such as house
dust. In most cases, there is no commonly recognized
concentration of these proteins, or particle size that can be
interpreted as necessary for producing sensitization, or
capable of eliciting an allergic response. Previously, we
demonstrated that altering the volume and ratios of
aqueous to solvent phases can influence shape, size,
surface texture and protein loading of custom made
polymeric microspheres. We characterized the
morphological and aerodynamic size of those particles
using scanning electron microsopy (SEM) in combination
with an Aerodynamic Particle Sizer (APS) and a Dekati
Electrical Low Pressure Impactor (ELPI). Our current
efforts focus on the optimization and reproducibility of
protein loadings of 15% or more when encapsulating
ovalbumin in polymeric microspheres in the 0.5 to 5
micron size range. Recently we have expanded our work
to try to achieve similar protein loadings through
adsorption of ovalbumin on blank microspheres.
Additionally, we have initiated studies to see if allergenic
proteins from cat hair can be efficiently encapsulated or
adsorbed using similar procedures. These particles with
improved (higher) loadings of encapsulated or adsorbed
ovalbumin, or cat hair protein will be analyzed using
these three methods to monitor any alterations in
morphology or aerodynamic size due to changes in
protein content. Use of these microspheres in in-vivo
models of allergic disease may be useful for identifying
particle characteristics that are important for eliciting an
immune response. This knowledge can lead to better
control methods for indoor allergenic aerosols,
particularly asthma triggers.
3B.4
Indoor air quality of four Southern High Plains dairy
milking parlors in summer and winter. CHARLES W.
PURDY, R. Nolan Clark, USDA-ARS; David C. Straus, Texas
Tech University Health Sciences Center.
Milking parlor indoor air quality of 4 large dairies was
sampled to investigate: 1) bacterial and fungal
3
concentration/m of air, 2) bioaerosol microbial types, and
3) respirable and non-respirable bioaerosol
3
concentrations/m of air. Equipment used were cascade
biological samplers, a laser strategic aerosol monitor
(SAM), and a weather station. Design & Methods: two
milking parlor sampling sites were established for the
equipment, one site on each end (front and back) of the
milking parlor center alley with cows on both sides facing
the alley. Cascade impactors were loaded in duplicate
with each of 10 different media. Vacuum pumps
displaced 28.3 L of air/min, and media were exposed
from 30 seconds to 15 minutes depending on the medium
used. Statistically the overall bioaerosol ANOVA model
statement included the following parameters: one of 10
bioaerosol types, 2-stage or 6-stage impactors, winter and
summer, parlor indoor aerosols compared to outdoor
aerosol, AM and PM aerosols, and aerosols of 4 dairies.
Conclusions: milking parlor indoor aerosols colony
3
forming unit (CFU)/m for all bacterial types were
significantly increased compared to outdoor aerosols.
Mesophilic fungi were significantly increased in outdoor
aerosols compared to indoor aerosols; however,
thermophilic fungi were significantly increased in indoor
aerosols compared to outdoor aerosols. Six-stage total
mean microbial aerosols among dairy parlors ranged from
2,124
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
3B.5
TBA
3B.6
Design and Development of an Electrostatic Sampler for
Biological Aerosols with High Concentrating Rate.
GEDIMINAS MAINELIS, Tae Won Han, Rutgers University.
Integration of bioaerosol sampling with modern analysis
techniques, such as PCR, requires samplers that can not
only efficiently collect particles, but also to concentrate
them in small amounts of fluids. In this research, we
began development of a novel bioaerosol sampler, where
a combination of electrostatic collection mechanism with
superhydrophobic (\Lotus leaf\ type) collection surface
allows for efficient particle collection, removal and
concentration in small water droplets: 10 to 50 micro-L.
This new sampling concept allows achieving very high
sample concentration rates (up to 1 million) and could be
applied to detect low concentrations of bioaerosols in
various environments.
The prototype Electrostatic Precipitator with
Superhydrophobic Surface had a shape of a closed halfpipe, where top surface served as a ground electrode,
while 3 mm wide collecting electrode covered by a
superhydrophobic substance was positioned in a groove
of the flat bottom surface. Airborne particles drawn into
the sampler were positively charged and then by the
action of electrostatic field deposited onto the negatively
charged electrode. The sampler was positioned at a ~20
degree angle, and the injected water droplets rolled-off of
electrode's surface removing deposited particles. Our tests
have shown that at a sampling flow rate of 10 L/min we
achieved retention efficiency of about 90% for 3 micrometer PSL particles. By using 20 and 40 micro-L water
droplets, we achieved concentration rates as high as
100,000 as indicated by counting of removed particles by
microscopy. Majority of the particles are removed by the
first applied droplet, and few particles are removed by
subsequent droplets. Tests with other particles also
yielded high concentration rates, which points to the
suitability of this new method for measuring low
concentrations of bioaerosols. The sampler's performance
is being improved further by adjusting the sampling
flowrates, strength of ion source and collection voltage.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
4B.1
Generation of Hydroxyl Radicals from Ambient Particulate
Matter in a Surrogate Lung Fluid. EDGAR VIDRIO, Chin
Phuah, Ann M. Dillner, Cort Anastasio, University of California
- Davis.
While epidemiological research links exposure to
particulate matter (PM) to several adverse health effects,
including cardiovascular and pulmonary disease, the
mechanisms for these effects are still poorly understood.
The generation of reactive oxygen species, such as
hydroxyl radical (OH), from inhaled particles is one of the
many hypotheses for PM toxicity. Although there are
several studies that have measured the generation of OH
from PM, very few have done so in a quantitative manner
in actual or surrogate biological fluids. Furthermore, there
is little data illustrating how OH generation from PM
samples varies as a function of time of year. Therefore,
the goal of our work here was to quantitatively measure
OH formation in a surrogate lung fluid (SLF) from PM2.5
collected over the course of a year. To do this we
collected three consecutive 24 hour PM 2.5 samples in
Davis once each month. Samples were extracted in our
SLF solution and the amount of OH generated was
quantified using a benzoate chemical probe.
Deferoxamine (DSF), a chelating agent, was added to a
portion of our samples to remove transition metal
reactivity in order to assess the fraction of OH produced
via metal mediated pathways. Overall, if we express the
amount of OH produced in each sample normalized by
3
volume of air sampled (e.g., nmol OH per m air), the
amounts vary greatly day by day and show no consistent
seasonal variation. However, when the amount of OH
produced in our samples is normalized by particulate
mass (e.g., nmol OH per micro-gram PM), we see a clear
seasonal variation, with a maximum in the summer and
minimum in the winter. In addition, metals play a key
role in OH formation from our particles: on average, the
addition of DSF reduced the OH reactivity of our PM2.5
samples by 93%.
4B.2
Removal Efficiency and Disinfection Capacity of IodineTreated Filter for Virus Aerosols. JIN-HWA LEE, Chang-Yu
Wu, Katherine M. Wysocki, Christiana N. Lee, University of
Florida; Joseph Wander, Brian Heimbuch, Air Force Research
Laboratory, Tyndall Air Force Base.
The iodine-treated filter, which combines mechanical
filtration and the disinfection capacity of iodine, was
tested for protection against airborne pathogens. The
removal efficiency of the test filters was assessed by
challenging it with MS2 bacteriophage. The experiments
were conducted at three environmental conditions: room
temperature (23 +/- 2 Celsius) & low RH (35 +/- 5 %),
high temperature (40 +/- 2 Celsius) & low RH, and room
temperature & medium RH (55 +/- 5 %). After removal
efficiency experiment, the filter was vortexed to extract
the collected MS2 from the filter.
The pressure drag of the tested filters was 352 Pascal/
(meter/second) with a negligible variation during the
entire experiment. In comparison, the pressure drag of a
glass fiber filter was 38,625 Pascal/(meter/second). Both
iodine-treated and untreated filter exhibited a similar
removal efficiency at room temperature and low RH, 94
+/- 3 % and 92+/-2 %, respectively. At high temperature
and low RH, the removal efficiency of the iodine-treated
filter presented a higher value (99.98+/-0.04 %), while
that of untreated filter (93+/-4 %) was similar to the
results of room temperature and low RH. The iodine
vapor released from the iodine-treated filter at high
temperature may affect the infectivity of MS2, though the
filter has similar removal efficiency to the untreated filter.
No significant difference between the survival fraction
(CE/CC, CE: Extracted MS2, CC: Collected MS2) of
collected MS2 on the iodine-treated and untreated filter at
the same environmental condition was observed: 2.2*10 2
+/-8.0*10 3 vs. 4.0*10 2+/-3.0*10 2 at room temp. & low
RH, and 7.8*10 1+/-7.7*10 1 vs. 8.7*10 1+/-4.4*10 1 at
high temp. & low RH. According to the t-test, the
difference between the average of survival fraction
between room and high temperature was not significant
(p-value was 0.08). The insignificant effect of iodine on
the infectivity of MS2 can be explained by the shielding
effect of aggregated MS2 particles collected on the filter.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
4B.3
Collection of influenza virus aerosols: comparison of
sampler efficiencies with molecular and infectivity assays.
PATRICIA FABIAN, James McDevitt, Harvard School of
Public Health; Donald Milton, University of Massachusetts
Lowell.
Apprehension regarding an influenza pandemic is on the
rise due to concern regarding high rates of morbidity and
mortality that could occur if a new highly virulent strain
of influenza spreads throughout the world. Methods to
evaluate transmission of influenza via aerosols are limited
and largely un-validated. Using a benchtop aerosol
generation chamber we collected air samples using 2.0
um Teflon filters, gelatin filters, the SKC Biosampler and
the Polyurethane Foam (PUF) Compact Cascade Impactor
(CCI) fitted with a single, 0.16 um 50% cut diameter
stage. Samples were analyzed for total influenza A virus
nucleic acid using reverse transcriptase-quantitative
polymerase chain reaction (RT-qPCR) and infectivity was
quantified using a cell culture based fluorescent focus
reduction assay. When compared to the SKC Biosampler,
total virus recovery, measured by PCR, from the gelatin
filter, CCI and Teflon filter was 69%, 38%, and 66%,
respectively (all differences were significant at the 0.05
level). Differences between the Teflon and gelatin filters
were not statistically significant and both were
significantly greater than the CCI results. When compared
to the SKC Biosampler, infectious virus particles assayed
from the gelatin filter, CCI impactor, and Teflon filter
were 10%, 7% and 24%, respectively (all differences
were statistically significant at the 0.05 level). Analysis of
the ratio of the PCR results to the infectivity assay results
for the SKC Biosampler, gelatin filter, CCI and Teflon
were 0.34%, 0.06%, 0.09%, and 0.10%, respectively.
These results suggest that recovery of viruses from filters
and other dry impaction substrates is problematic in terms
of virus removal from the surfaces and virus survival.
Collection of influenza aerosols directly into liquid media
favors virus recovery and continued infectivity. Despite
these advantages, the SKC Biosampler is not an ideal
sampler due to relatively low flow rates and dilution of
samples into large impinger fluid volumes.
4B.4
Detection of Airborne Influenza And Avian Influenza Virus.
Pei-Shih Chen, Qian Kun Lin, FENG-DA TSAI, Kaohsiung
Medical University.
To our knowledge, there was no study to quantify
airborne influenza and avian influenza virus. Therefore,
the purpose of our study is to establish a method to quick
quantifying the airborne influenza and avian influenza
virus in the air. Then, field validation was also held at an
air quality monitoring station and a live-bird market.
Air samples were sampling with Teflon filter within a 37
milli-meter cassette and analyzed by real-time qPCR
(ABI PRISM 7500 Sequence Detection System). In
addition, sampling stress and storage effect were also
evaluated. Furthermore, 48 samples were collected at a
live-bird market and 12 samples were collected at SinJhuang air quality monitoring station during February
2006 to September 2006.
2
Our results showed that the R value of standard curves
were 0.988, 0.995 and 0.998 for influenza A, influenza B
and influenza H5 virus, respectively. In regard to the
3
3
detection limits, they were 0.78 copy/m ,0.58 copy/m
3
and 1.1 copy/m for influenza A, influenza B and
influenza H5 virus, respectively. For sampling stress,
close phase of three piece cassette was batter than open
phase. In regard to storage effect, it was 94% within three
days at 4 degrees C of sampled virus. For the isolated
RNA, it can be conserved at least three months without
any degradation at -80 degrees C.
For field validation, the positive rate of air samples at air
monitoring station was 67%. The virus concentration was
3
in the range of 4~439 copy/m . In live-bird market, the
positive rate for chicken stall was 42% and the virus
2
4
concentration was in the range of 4.36x10 ~1.97x10
3
copy/m . For duck stall, the positive rate was 33% and
2
4
the virus concentration ranged from 8.10x10 to 2.83x10
3
copy/m . These results showed that this quantifying
method was successfully developed and validated in the
field.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
4B.5
Acute Injury to Rat Airway Epithelium by Exposure to
Flame-Generated Soot Particles Doped with 1Nitronaphthalene. BENJAMIN KUMFER, Lindsay Davison,
Evan Wallis, Michelle Fanucchi, Ian Kennedy, University of
California - Davis.
4B.6
Comparative Composition and Inhalation Toxicity of Urban
versus Rural Samples of Resuspended Paved Roadway
Material. JAKE MCDONALD, JeanClare Seagrave, Matthew
Campen, Joe Mauderly, Lovelace Respiratory Research
Institute.
Particles emitted from anthropogenic sources, such as
diesel engines, are often found to contain on the surface
organic species, including PAHs, oxy-PAHs, and nitroPAHs, that may contribute to the adverse health effects
associated with ambient PM. The large number of organic
species found in ambient PM makes it difficult to assess
the toxicity of these components individually.
Alternatively, using pure organic materials without carrier
particles is not desirable for in vivo exposure studies,
since this does not mimic the natural exposure route. To
overcome these problems, a method was developed for
the synthesis of flame-generated soot particles doped with
specific organics for toxicity studies. Soot particles,
which are initially clean of PAHs, are generated from an
acetylene diffusion flame. The post-flame aerosol is
subsequently mixed with a heated gas stream containing
saturated PAH vapor and then cooled to promote
condensation of the PAH onto soot particles. This system
was evaluated using 1-nitronaphthalene (1-NN), a
substance found in ambient PM and a known toxin. Good
control over the amount of condensed 1-NN was achieved
by varying the temperature of the saturated PAH stream.
Particles of variable 1-NN loading were instilled into rat
airways using a dry powder insufflator. The injury to
epithelium was determined by high-resolution
histopathology and by measurement of the release of
cytokines associated with oxidative stress. Results show
an increase in injury with 1-NN loading. The injury
induced by flame-generated soot was also compared to
that by commercial carbon black. Carbon black particles
were found to be more efficiently cleared from the
airways than were soot particles, suggesting that the
clearance mechanism is dependent upon particle size,
morphology and possibly surface composition.
On June 30, 2005 the U.S. EPA released a Staff Paper
with a recommendation to regulate PM10- PM2.5, and to
focus the regulation on urban areas because of a
perception that urban dust will be more of a health
concern than rural dust. Unfortunately the data on
composition (and toxicity) of dust in these specific size
fractions is extremely sparse, and much of the suggestions
that dust in more urban areas is more toxic than non-urban
areas is based on speculation. We have conducted a study
to characterize the composition of resuspended dust in
several areas throughout the U.S., to define the chemical
and physical characteristics of dust in the coarse and fine
fractions by region, including contrasts in cities that are
defined as rural and urban dominated. Samples were
collected from street surfaces in NY, NJ, GA, AL, NM,
TX, KS, and CA. Initial aerosol experiments showed
clear differences in the composition of urban versus rural
samples. A strategy to composite samples according to
urban versus rural was developed, and two separate
exposure atmospheres were employed after aerosolization
with a Wright Dust Feeder. Dust traversed through a
PM2.5 cyclone and into a rodent nose-only inhalation
exposure system where pulmonary toxicity in rodents was
assessed. Samples were collected for analysis of particle
size, metal content, organic species and total organic
carbon, endotoxin, protein, and carbohydrate.
Composition and differential health response after
inhalation will be reported. Funding from the National
Environmental Respiratory Center, with contributions
from multiple federal and non-federal sponsors.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11E.1
Development of Sampling and Analysis Methods to Monitor
Nanoparticles in the Workplace Environment. GARY
CASUCCIO, Traci Lersch, Keith Rickabaugh, RJ Lee Group,
Inc.; Randall Ogle, John Jankovic, Oak Ridge National
Laboratory.
11E.2
Increases of Iron Concentrations of Human Airway
Epithelial Cells in Vitro by Exposure to Magnetic
Nanoparticles Coated with Organic Aerosol and Inorganic
Acid. MYOSEON JANG, The University of North Carolina at
Chapel Hill; Andrew J. Ghio, Environmental Protection Agency.
The recent thrust in research related to nanotechnology
has created opportunities to improved materials, devices,
and systems that can exploit the physical, chemical, and
biological properties at the nanoscale level (1 to 100
nanometers). While nanotechnology is still in the
developing stages, nanoparticles are already being used in
a number of industries including electronic, magnetic and
optoelectronic, biomedical, pharmaceutical, cosmetic,
energy, catalytic and materials applications.
BACKGROUND. Applications of iron oxide magnetic
nanoparticle (MNP) are diverse including magnetic
recording media, magnetic filters for the removal of
selected impurities from various types of fluids, and
numerous biomedical uses for hyperthermia, drug
delivery, and nuclear magnetic resonance contrast agents.
However, little is known about the health effects of iron
oxide MNPs. The toxicity of airborn MNPs mainly
depends upon coating materials and dose amounts. We
hypothesize that atmospheric inorganic acids and organics
can interact wtih airborn MNP and accelerate the
dissolution of MNP and potentially increases the adverse
biological effects.
APPROACH. The study began with an appropriate
device design and the dose model to deliver MNP onto
the target area of human airway epithelial BEAS-2B cells
in vitro. MNPs were nebulized into the gas-phase of
indoor Teflon film chambers and directly coated with
seconadry organic aerosol (SOA) created from ozonolysis
of alpha-pinene. The resulting airborn matter was
targeted on in vitro human airway epithelial cells through
the exposure device. Each experimental set for the
exposure study comprised of background air, MNP, SOA,
SOA/MNP, inorganic acid/MNP, and inorganic acid/
MNP/SOA and was compared to the base control cells
that remained in the incubator for the duration of
exposure experiments. Cells were grown to confluence on
inserts, an air-liquid interface introduced, and exposed.
Twenty-four hours following the exposure, supernatant
was collected to determine cell iron concentrations using
colorimetric analysis and various biological effects.
RESULTS. The solubility of MNP is accelerated in the
presence of sulfuric acid, increasing ferric and ferrous
ions in aerosols. The SOA products also accelerate the
dissolution of iron oxide by chelating metal ions. The
human airway epithelial BEAS-2B cells in vitro exposing
to sulfuric acid or SOA products increase all biological
effects (IL-8, ferritin, acetaldehyde, and DMT1 RNA)
being tested, except ferritin RNA. This study indicates
that the interaction of MNP with sulfuric acid or SOA can
produce unpleasant health effects through increases ionic
iron species in airborne MNP matter.
Although the field of nanotechnology is developing
rapidly, there is concern regarding the potential toxicity of
nanoparticles, and little is known regarding worker
exposure to particles in this size range in the workplace
environment. This issue is compounded by the fact that
nanoparticles, from both natural occurring and
anthropogenic sources, are ubiquitous in the ambient
environment and engineered nanoparticles will vary in
physical and chemical properties from industry to
industry. Furthermore, methods to monitor nanoparticles
in the workplace and references to acceptable dose levels
have not established. The combination of these factors
has created a unique challenge for health and safety
professionals with respect to monitoring the workplace
environment.
This presentation will discuss the development of
sampling and analysis methods to monitor worker
exposure.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11E.3
A Study on Magnetic Passive Aerosol Sampler for
Measuring Aerosol Particle Penetration through Protective
Ensembles. Zhong-Min Wang
Current testing for particle penetration through protective
ensembles is mainly based on active filtration principles,
which may overestimate the particle penetration due to
additional driving force. In contrast, passive sampling has
the potential to be a simple, small size, light weight, and
inexpensive device for aerosol sampling. However, most
passive devices are not suitable for the testing because of
low collection efficiency and the requirement for long
sampling times.
A novel prototype magnetic passive aerosol sampler
(MPAS) has been developed for measuring particle
penetration through protective ensembles. The MPAS
consists of a 25 mm diameter sampler body, a piece of
Teflon (or PVC) film, and magnet(s). The magnet could
be a single magnet or a number of mini disc magnets
(multi domain). For the latter, the magnets were arranged
with an alternative N and S pole pattern. Iron oxide
nanoparticles were generated using an atomizer as the
challenge aerosol. Passing through a diffusion dryer, the
aerosol was then sent to a testing chamber where the
MPAS and an active sampler were located. The active
sampler was a 25 mm PVC filter cassette operated at 200
ml/min. After each test, the active sampler and MPAS
were analyzed using a microbalance. Compared to a
conventional passive sampler, the MPAS was able to
collect more particles due to the magnetic force.
Although the collection efficiency of the MPAS was
approximately 100 times higher than that of a
conventional passive sampler, it obtained only about 0.8
percent of the collection efficiency of the active sampler.
With sampling times of 2, 4, and 6 hours, particles
collected by the MPAS were approximately 5, 10, and 15
micrograms, respectively. The multi domain pattern
greatly improved the uniformity of particle deposition and
magnetic force was able to drop off within a few mm
from the magnets' surface.
11E.4
Measurement of Airborne Nanoparticle Exposures
Associated with the Use of Fume Hoods. SU-JUNG TSAI,
Earl Ada, Michael J. Ellenbecker, University of Massachusetts
Lowell.
Manual handling of nanoparticles is a fundamental task of
most nanomaterial research; such handling may expose
workers to ultrafine or nanoparticles. The chemical
laboratory hood, also called the fume hood, is commonly
used in university and research laboratories as the primary
local exhaust ventilation (LEV) system. Such hoods rely
on the proper face velocity for optimum performance. As
air flows around the worker toward the hood,
counterrotating eddies occur on the downstream side of
the worker and the reverse flow can pull the airborne
nanoparticles back into the worker's breathing zone. The
end of the reverse flow zone reaches at least two body
widths downstream of the worker and implies that a handheld contaminant source cannot escape the influence of
the recirculating flow.
Experiments were performed to measure airborne particle
concentration while handling nanoparticles in two fume
hoods located in different buildings under a range of hood
operating conditions. A TSI Fast Mobility Particle Sizer
(FMPS) was used to measure airborne particle
concentration from 5 nm to 560 nm in 32 size channels.
Nanoalumina was selected as the primary material. Air
samples were also collected on titanium dioxide-filmed
TEM grids placed on polycarbonate membrane filters and
particles were characterized by transmission electron
microscopy (TEM) and scanning electron microscopy
(SEM). Handling tasks included transferring particles
from beaker to beaker by spatula and by pouring.
Measurement locations were the room background, the
researcher's breathing zone and upstream and downstream
from the handling location. Airborne particle
concentrations measured at breathing zone locations were
analyzed to characterize exposure level. Statistics were
used to test the significance of differences between data.
Measurements at the researcher's breathing zone using
Hood 2 found elevated airborne particle number
concentrations during both particle handling and post
handling. The complete results will be fully discussed in
our presentation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11E.5
Generation of Agglomerates of Nanoparticles for use in
Biological Studies. DAVID G. NASH, Owen R. Moss, Brian
A. Wong, The Hamner Institutes for Health Sciences.
11E.6
Occupational Monitoring of Carbonaceous Nanomaterials.
M. EILEEN BIRCH, Douglas E. Evans, Bon-Ki Ku, National
Institute for Occupational Safety and Health.
Because of their high diffusivity, nanoparticles released
into the atmosphere will likely begin to agglomerate. The
state of agglomeration upon inhalation and the potential to
deagglomerate back into nanoparticles may affect the
toxicity of the inhaled material. In order to investigate
this, a system was set up to generate aggregates from
agglomerates. Primary particles, composed of zinc, were
generated using zinc rods in a spark generator (Palas GFG
-1000, Karlsrhue, Germany). These agglomerates from
the spark generator were passed through a room
temperature aging chamber or through a tube furnace
(Carbolite HST, Derbyshire, UK). Agglomerate size was
measured with a scanning mobility particle sizer (SMPS
model 3936, TSI Inc., Shoreview, MN). When furnace
temperature was set near the zinc coalescence
temperature, instead of decreasing in size, agglomerate
size increased up to 30%; a percentage increase
duplicated with the room temperature aging chamber.
Starting with an aerosol of primary zinc particles, equal
concentrations of agglomerate and aggregrate aerosol
were produced.
Carbon nanotubes and nanofibers (CNTs/CNFs) are one
of the most mass-produced engineered nanomaterials. The
annual global production of CNTs is over 100,000 tons;
CNF production and use are increasing at a similar pace.
High volume production presents an exposure concern for
workers who handle these materials, which may be
especially toxic if inhaled because of their composition
(metal catalysts) and fibrous structure. Field
investigations to evaluate potential exposures to CNTs/
CNFs are important because the toxicological properties
of these materials are not yet understood, and the
manufacturing processes are rapidly changing. Studies
were conducted at two facilities that produce CNFs or
composite materials thereof. Among other measurements,
air and surfaces samples collected in different areas were
monitored for total carbon (TC) by NIOSH Method 5040
[1]. Nearby office areas also were monitored to check for
possible contamination outside the process areas. In
addition, air samples for analysis by transmission electron
microscopy (TEM) were collected, and multiple, directreading instruments were used for air monitoring. At one
facility, TC concentrations in the processing areas were 2
to 64 times higher than those in an office area. Surface TC
loadings were about 3 to 30 times higher. Several TEM
samples evidenced fibers or entangled fiber bundles.
Particle number concentrations in 11 process areas were
well below outdoor background. Slight increases relative
to laboratory background occurred during weighing/
mixing CNFs and cutting a composite material with a wet
saw. Particle mass concentrations for the 11 processes
exhibited the same general trend as number
concentrations. The major finding was with the wet saw,
where the mass concentration increased to about three
times laboratory background. Preliminary TC results for a
second facility also indicate air contamination. TC
concentrations in six processing areas were 3 to 155 times
higher than that found in an office area.
Disclaimer: The findings and conclusions in this abstract
have not been formally disseminated by the National
Institute for Occupational Safety and Health and should
not be construed to represent any agency determination or
policy.
[1] Birch, M.E., Monitoring diesel particulate exhaust in
the workplace, Chapter Q, NIOSH Manual of Analytical
Methods, DHHS(NIOSH) Publication No. 2003-154
(2003).
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11E.7
Modeling of Workplace Nanoparticle Exposure. CHRISTOF
ASBACH, Heinz Kaminski, U. Rating, Heinz Fissan, Thomas
A.J. Kuhlbusch, Institute of Energy and Environmental
Technology (IUTA)
11F.1
Physical and Chemical Characteristics of Aerosol Mists in
Fertilizer Manufacturing Facilities. YU-MEI HSU, ChangYu Wu, Dale A. Lundgren, University of Florida; Brian Birky,
Florida Institute of Phosphate Research.
Inhaled nanoparticles are currently being discussed to
induce adverse health effects. Particles in this size range
can either originate from unintended anthropogenic or
natural formation processes or can be intentionally
produced engineered nanoparticles. Exposure to
engineered nanoparticles in e.g. workplaces has recently
raised increased interest as these particles may have
health relevant properties beyond their mobility due to
particle size. Monitoring of nanoparticle exposure at
workplaces in the nanotechnology industry is thus an
important issue for assessing the safety of workers.
For an expedient investigation of possible workplace
exposure as well as save workplace design it is essential
to understand the behaviour and dispersion of particles
within a given workplace. Precise modeling of particle
dispersion and physical reactions can therefore help to
improve work place design, identify hot spots within the
room, and give guidance for the definition of suited
measurement locations. Furthermore, such simulations
can provide a three dimensionally resolved picture of the
exposure scenario. Therefore exposure can also be
inferred from the model based on the posture of the
worker, e.g. whether she/he is sitting or standing.
In this work, three exemplary, realistic workplaces have
been simulated using the commercial computational fluid
dynamics (CFD) code FLUENT, along with the Fine
Particle Model (FPM). The modeled scenarios comprise a
welding workplace, a large hall with several pipes and a
reactor with (nano-) particles trickling from a conveyor
belt, and a hall with a leak in a slightly pressurized
transport pipe of freshly formed nanoparticles. Besides
convective transport of the particles and dilution with
background air (assumed to be particle free), the
considered physical processes also comprised
sedimentation and coagulation.
Results from the different simulated scenarios will be
presented along with recommendations for representative
and conclusive modelling of workplace exposure to
nanoparticles.
Strong inorganic acid mists containing sulfuric acid have
been reported to correlate well with lung and laryngeal
cancers in humans. Phosphate fertilizer manufacturing
facilities are listed as one of many occupational exposures
to strong acid. To better protect workers from potential
exposure, a field campaign was carried out to determine
the physical and chemical characteristics of mist aerosols
in fertilizer manufacturing facilities.
The sampling was carried out at 8 phosphate fertilizer
plants using the UW-cascade impactor to obtain size
fractionated information. The sampling time was 24
hours and 3 samples were obtained at each location. Ion
chromatography was used to analyze water soluble
species, including sulfate, phosphate, fluoride, chloride,
nitrate, sodium, potassium, magnesium, calcium and
ammonium. The sampling results indicate that the
highest sulfuric acid mist concentration was obtained at
the sulfuric acid pump tank area and the highest
phosphoric acid mist concentration was obtained at the
belt/rotating table filter floor. Acid mists at these areas
were dominant in the coarse mode when high
concentrations were identified. The major species found
at the plants were phosphate, sulfate, fluoride, ammonium
and calcium. An aerosol thermodynamic model was used
to estimate the acidity of aerosols with sulfuric acid
concentration higher than 200 micro-gram/m3. The
calculation indicates the mode size of hydrogen
concentration in the ambient condition was 1.8-3.8 micrometer for 5 samples and 3.8-10 micro-meter for 2
samples. In the high relative humidity environment, i.e.
the human respiratory system ( < 95%), the aerosol can
reach its equilibrium size within 0.014 seconds, which is
longer than the traveling time of an aerosol in the upper
respiratory system. Particles with these sizes mainly
deposit in the upper respiratory tract and the results agree
with the relation between strong inorganic acid mists
containing sulfuric acid and laryngeal cancer.
This work has been supported by the government of
North-Rhine Westphalia and the European Union under
grant number 005-0406-0004.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11F.2
Time-Dependent Release of Iron from Soot Particles by Acid
Extraction and the Reduction of Fe3+ by Elemental Carbon.
STEPHEN DRAKE, Bing Guo, Texas A&M University.
Elemental carbon reduces Fe3+ to Fe2+ in aqueous
solutions. This process has potential implications in the
adverse health effects of fine particle air pollution,
because both elemental carbon and iron are major
components in atmospheric particulate matter. In this
study we measured the time-dependent release of iron
from laboratory flame soot particles that contained low
concentrations of iron, and the reduction of Fe3+ to Fe2+
in an acid extraction process. The concentration of Fe3+
and Fe2+ ions in the extraction solutions was measured
by a spectrophotometric method. The original valence
state of iron in the soot particles was measured by
Mossbauer spectroscopy. The results showed that while
Fe3+ was the dominant valence state in the dry soot
particles, significant fraction of iron was reduced to Fe2+
in the aqueous solution. Further investigation is needed to
assess the significance of this phenomenon in the
biological effects of Fe-containing soot particles.
11F.3
Tracking personal exposure to diesel exhaust at a trucking
industry freight terminal using organic tracer analysis by
thermal desorption GCMS. REBECCA J SHEESLEY, James
J Schauer, University of Wisconsin, Madison; Thomas J Smith,
Francine Laden, Drew Blicharz, Harvard School of Public
Health; Eric Garshick, VA Boston Healthcare System, Channing
Laboratory, Brigham and Women's Hospital and Harvard
Medical School; Jeff DeMinter, Mark Meiritz, University of
Wisconsin-Madison, Wisconsin State Lab of Hygiene.
Personal exposure samples were collected in St. Louis,
MO as part of a larger epidemiologic project aimed at
assessing carbonaceous fine particulate matter (PM)
exposure at trucking terminals. The mixture of personal
exposure, ambient worksite and ambient urban
background samples provides an unique opportunity to
track the work-related exposure to carbonaceous fine PM
in a trucking terminal. A comparison of the samples
provides information on the origin of the personal
exposure to elemental and organic carbon: urban
background, work site or personal activity. To accurately
assess the impact of PM emission sources, particularly
motor vehicle exhaust, elemental and organic carbon by
thermal optical transmittance and nonpolar organic tracer
analysis by thermal desorption gas chromatography/mass
spectrometry (TD GCMS) were conducted on all of the
PM samples.
This data set provides an excellent base for the discussion
of important questions: How well does an area sample
(local urban or indoor) represent personal exposure and
does this vary by source? Do elemental carbon and
hopanes track the same motor vehicle source within a
diesel-impacted environment? For primarily on-site
workers, this data suggests that the worksite sample can
be used to estimate personal exposure to motor vehicle
exhaust. This does not appear to be true for the truck
drivers, as the urban background and yard (terminal
background) samples do not accurately depict their
exposure to motor vehicle exhaust; the drivers tend to
have a much higher exposure than the ambient samples
and the correlation for specific compounds is poor. Also,
an overall assessment of the personal exposure, work site
area and local background samples indicates that hopanes
and elemental carbon do not necessarily depict the same
source. Thus, elemental carbon measurement would not
be sufficient to depict the lube oil/organic carbon
component of the diesel source in freight terminals or
other high impact areas.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
11F.4
Characterization of welding fume particles generated from a
robotic welding system. BEAN T. CHEN, Sam Stone, Diane
Schwegler-Berry, Amy Frazer, Michelle Donlin, Jared
Cumpston, Aliakbar A. Afshari, David G. Frazer, Vincent
Castranova, James M. Antonini, National Institute for
Occupational Safety and Health.
Epidemiological studies suggest that the long-term
inhalation of welding fumes may lead to lung disease,
neurotoxicity, and cancer. To study health effects of
welding fume exposure, a computer controlled, robotic
welding system was developed at NIOSH to allow for
continuous welding for animal exposure. Gas metal arc
welding was performed using a stainless steel electrode.
A flexible trunk was attached to the robotic arm of the
welder and was used to collect and transport fume from
the vicinity of the arc to the animal exposure chamber.
During welding, fume mass concentrations were
continuously monitored with a real-time aerosol monitor
and gravimetrically measured with Teflon filters. Fresh
fume particles attained concentrations as high as 150 mg/
3
m which could be diluted with air to maintain a desired
exposure concentration in the chamber. At a mean
3
concentration of 40 mg/m , the aerosol samples were
taken using polycarbonate filters for scanning electron
microscopy and grids for transmission electron
microscopy to assess particle morphology and elemental
composition. In order to estimate the deposited dose in the
pulmonary region of exposed animals, the particle size
distribution was measured with both a Micro-Orifice
Uniform Deposit Impactor (MOUDI) and a Scanning
Mobility Particle Sizer (SMPS). Results from MOUDI
indicate the mass median aerodynamic diameter was
approximately 0.24 um with a geometric standard
deviation of 1.39. Although this median diameter has a
value similar to that obtained from SMPS, there were
discrepancies between the two distributions in both mass
and number. Results show that welding fume particles
having the same electrical mobility appear to exhibit very
different aerodynamic behaviors because of their highlyaggregated morphologies and electric charging capacities.
These findings may apply to other ultrafine particles such
as diesel exhaust and carbon nanotube particles and
should be considered when interpreting their size
distributions.
11F.5
Stimulation of Rat Alveolar Macrophages by Water-Soluble
Components of PM2.5 Aerosols. Amy Prasch, MARTIN
SHAFER, Jocelyn Hemming, James Schauer, University of
Wisconsin-Madison; Michael Hannigan, University of
Colorado.
Daily PM2.5 samples were collected for one year in the
Denver metro area on pre-cleaned Teflon filters. Filters
were sectioned for chemical characterization, including
high-resolution ICP-MS analysis of both water soluble
and total forms of major and trace inorganic species (over
40 elements). In addition major ions (sulfate, nitrate,
chloride, ammonia) and carbon and nitrogen were also
analyzed. To assess the potential toxicity of water-soluble
components of the PM2.5 samples, bioassays using a rat
alveolar macrophage cell line were developed and applied
to a subset of the samples. Macrophages are exposed to
aqueous filter leaches and subsequently assessed for
viability and production of reactive oxygen species (ROS)
as an indicator of macrophage stimulation. The novel
macrophage bioassay exhibited high sensitivity (100
microgram PM samples are workable) and excellent short
and long-term precision. Importantly, because the
macrophages are maintained in a well-defined media,
devoid of complex organic ligands, the bioassay is
compatible with further trace element speciation
protocols. The method is rapid (2-hour exposures),
automated (96-well plate reader), and suitable for
implementation in the context of large scale air
monitoring and health effects studies. Results from the
Denver study show that variations in the magnitude of the
ROS response observed between samples were only
partially explained by differences in the total mass
loading. This suggests that the chemical composition of
the PM2.5 is also important in mediating this response
and may be an important factor in explaining how PM2.5
exposure leads to adverse health effects. We are currently
exploring statistical relationships between ROS response
and the comprehensive elemental dataset.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
12B.1
Effects on manufactured nanoparticles on lung and vascular
cells. JOHN VERANTH, N. Shane Cutler, Cassandra Deering,
Agnes Ostafin, Garold Yost, University of Utah.
Recent publications by our laboratory related to the
effects of manufactured nanoparticles on lung epithelial
cells and on vascular endothelial cells will be integrated
and summarized. Fluorescent labled nanoparticles are
shown to be taken up by various cell types leading to
concern about the potential biological effects of other
nanomateials as well. Nanoparticles from metal oxides
appear to have low potency for the induction of
proinflammatory signaling compared to soil-derived
dusts. In addition the metal oxide particles have limited
ability to induce formation of reactive oxygen species in
the tested cell cultures. Use of various pathway-specific
inhibitors provides insights into the cell signaling
pathways mediating biological responses. The in vitro
lung and vascular models provide an easily manipulated
biological system for studying particle-induced toxicities,
but details of the cell culture conditions can affect
observed results.
12B.2
Size Distribution and Characteristics of Airborne Unrefined
Carbon Nanotube Particles. JUDY Q. XIONG, Maire S.A.
Heikkinen, Beverly S. Cohen, New York University School of
Medicine.
Carbon nanotubes (CNTs) are among the most dynamic
and fast-growing nanomaterials due to their novel
properties. With a compound annual global production
growth rate of well above 60%, the potential of human
exposure to this new type material in the workplace as
well as in the general environment are rising, and their
impacts on human health are of largely concern.
A method has been developed in our laboratory for
sampling, quantifying and characterizing airborne CNT
particles utilizing a 13-stage Electrical Low Pressure
Impactor (ELPI) combined with image analysis by
Atomic Force Microscopy (AFM). The method is capable
of identifying agglomerated nanoparticles in the presence
of other airborne particles, and measuring size-resolved
number concentrations.
The technology has been applied for sampling and
characterizing airborne unrefined CNT samples (raw
material) of various types including single-walled
(SWNT), double-walled (DWNT) and multi-walled
(MWNT) nanotubes. The experimental data showed that
the particle sizes generated from all types of CNT raw
materials were widely distributed across all 13 stages of
the ELPI including the filter stage ranging from 7 nm to
10 um in diameter. The particle size distributions varied
with the type of CNTs and with the methods by which
they were manufactured. AFM results also showed that
the CNTs tend to agglomerate rather than exist as single
particles, physically. As deposition efficiency and sites of
inhaled particles within the respiratory system largely
depends on particle size, the deposition pattern of
agglomerated nanoparticles should be similar to those
larger equivalent sized non-agglomerated particles.
Nevertheless, entrained particles depositing on/in the deep
lung surfaces of the bronchioles or alveoli will contact
pulmonary surfactants in the surface hypophase and the
agglomerated CNT are likely to (ultimately) be deagglomerated. Therefore, to investigate human exposure
to airborne CNTs, the full size range of inhalable particles
must be taken into account.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
12B.3
Measured Airborne Nanoparticle Exposures at an NSF
Nanoscale Science and Engineering Center. SU-JUNG TSAI,
Kwangseog Ahn, Earl Ada, Michael J. Ellenbecker, University
of Massachusetts Lowell.
The NSF Center for High-Rate Nanomanufacturing
(CHN) is a collaboration of the University of
Massachusetts Lowell, Northeastern University, and the
University of New Hampshire which has a unique
purpose to integrate occupational and environmental
health and safety into its mission. As a part of this effort
we have monitored nanoparticle air concentrations and
size distributions in various research laboratories at CHN
and Massachusetts Institute of Technology.
A TSI Fast Mobility Particle Spectrometer (FMPS) was
used to measure airborne particle concentration from 5
nm to 560 nm diameter in 32 size channels. We have
monitored particle exposure at workplaces for 7 process
types and 20 operations to date. Measurements were
taken at background locations, source locations, and
researchers' breathing zones. Monitored processes
include electrospinning, compounding, carbon nanotube
furnace, fullerene reaction, twin screw extruding, silica
handling and carbon black handling. Monitored
nanoparticles include nanoclay, nanoalumina, carbon
black, fullerenes and carbon nanotubes. Some processes
were characterized by collecting aerosol particles for
further analysis by transmission electron microscopy
(TEM) or scanning electron microscope (SEM).
The change in breathing zone concentration before, within
and after an operation varies with process type and
operating conditions. Significant increases in particle
number concentration measured 8 centimeters (3 inches)
from a particle release source were detected on most
processes. One of the processes, twin screw extruder
(TSE), is the standard industrial equipment for
compounding nanocomposites. We have monitored the
compounding process using polymer beads with
nanoalumina particles. For the synthesis of carbon
nanotubes (CNTs), the measurement on the CNT furnace
found high particle number concentrations in the exhaust
air during synthesis of single-walled CNT by chemical
vapor deposition (CVD).
In this presentation, monitoring data from most
processes evaluated will be presented.
12B.4
The fate of airborne nanoparticles from a leak in a
manufacturing process into a working environment.
NICHOLAS STANLEY, David Y.H. Pui, Thomas Kuehn,
University of Minnesota; Christof Asbach, Thomas Kuhlbusch,
Heinz Fissan, Institute of Energy and Environmental
Technology.
Nanoparticle toxicology has become a major issue in
recent years as the potential for human exposure has risen.
A leak in nanoparticle production equipment can cause
large quantities of nanoparticles to be emitted into a work
environment. These nanoparticles can cause adverse
health affects, and toxicologists have proposed using
nanoparticle surface area as a health relevant measure to
assess worker exposure. However particle properties can
change as these particles traverse from the leak and
undergo physical and chemical reactions. The particle
properties need to be examined by looking at changes in
chemical composition, particle morphology, and number
and surface area concentrations.
Nanoparticles of different materials were injected through
an experimentally simulated leak into an ASHRAE 52.2 1999 classified wind tunnel. Measurements were taken
with various instruments to determine the fate of the
injected nanoparticles. Lung deposited surface area was
measured using an NSAM; an SMPS measured the size
distribution. Particle morphology and chemical
composition were determined using TEM or SEM/EDX
analysis. Particle size, material, and concentration were
altered, as well as other parameters. By adjusting the
experimental parameters (such as particle material and
pressure drop across the leak) the effect of various
conditions on the fate of nanoparticles could be studied.
Measurements however could only be taken at discrete
locations throughout the wind tunnel, so a numerical
simulation were used to show a more in depth look at the
fate of these nanoparticles.
The background of this project and the experimental set
up and conditions will be presented along with initial
results from experiments and modeling.
This project is sponsored by NSF G2006-Star-F2 (Fate/
Transport). The financial support is gratefully
acknowledged.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
12B.5
Evaluating the potential for release of carbon nanotubes and
subsequent occupational exposure during processing of a
nanocomposite. AMIT GUPTA, Mark L. Clark, Battelle
Toxicology Northwest; Daniel J. Gaspar, Pacific Northwest
National Laboratory; Michael G. Yost, University of
Washington; Gwen M. Gross, Paul E. Rempes, The Boeing
CompanyL; John C. Martin, Jr., Washington Technology
Center, Seattle, WA.
Evaluating the potential impacts of nanotechnology on human
health and the environment requires an understanding of the
potential for nanoparticle exposure during manufacturing
operations. Routine manufacturing processing steps for
nanocomposite materials (materials consisting of a nanoparticles
or nanotubes in a bulk, typically polymer, matrix) such as
sanding, grinding, water-jet cutting etc., have the potential to
liberate harmful components of a composite material. Little is
known about the potential for the release of nanoparticles when
common operations are performed on nanocomposites and there
is a chance the nanomaterial will be liberated, thereby creating
the potential for occupational exposure or environmental release
of waste containing free (unbound) nanoparticles or nanotubes.
This study is focused on the potential to generate respirable
nanoparticles when sanding a structural nanocomposite material.
In this initial study we determined the number, concentration,
size distribution, and morphology (including agglomeration)
generated by a sanding process on carbon fiber:MWNTmodified epoxy resin composite panels. The MWNT's (50-70
nm diameter) were dispersed into the epoxy resin binder after
surface modification to ensure good encapsulation by the epoxy.
The sanding experiments were carried out in a glove box
customized to support simultaneous real-time monitoring of
particle size distribution, number and concentration (using both
an optical particle counter and a condensation particle counter)
and collection of filter samples for ex situ characterization.
Particle size, morphology and degree of agglomeration were
evaluated using both scanning and transmission electron
microscopy. During sanding of a MWNT-containing material,
real-time monitoring instrument (Scanning Mobility Particle
Sizer) detected the presence of loose and unbound nanoparticles
whereas, they were not seen during sanding of a standard
carbon-carbon composite material (without MWNT). The TEM
images showed the presence of carbon nanotubes projecting
from the edges of larger particles. Preliminary attempts to
determine the chemistry of the nanoparticles were unsuccessful
using Raman and NIR spectroscopy. Future studies are planned
to identify the chemistry of the generated nanoparticles.
12B.6
Murine Pulmonary Pathology and Systemic Immune
Function Following Inhalation of Multiwalled Carbon
Nanotubes (MWCNTs). LEAH A. MITCHELL, Andrew
Gigliotti, Jacob D. McDonald, Lovelace Respiratory Research
Institute; Jun Gao, Scott W. Burchiel, University of New
Mexico.
The purpose of the following studies was to create a novel
inhalation exposure system for Multiwalled Carbon
Nanotubes (MWCNTs) and to determine if acute doses of
inhaled MWCNT cause significant pulmonary damage
and/or systemic immune function alterations. C57Bl/6
adult male mice were exposed to control air or 300 microg/m3 and higher particle concentrations caused systemic
immunosuppression. None of the doses administered for
7 days were adequate, and did not cause immune function
alterations compared to control animals.
Immunosuppressed animals were determined to be
suppressed in their T-dependent antibody response to
sheep erythrocytes as well as T cell proliferative ability in
presence of mitogen, Concanavalin A (Con A).
Furthermore, assessment of nonspecific Natural Killer
(NK) cell activity showed that animals exposed to 1000
micro-g/m3 had decreased NK cell function. Bronchial
Alveolar Lavagate and Histopathological analysis of
lungs from exposed animals showed little more than
macrophage engulfing black particulate and did not
indicate pulmonary inflammation, fibrosis, or granuloma
formation as has been suggested by others. Real time RT
PCR was conducted on RNA collected from spleen and
lung. Splenic mRNA expression of interleukin-10 (IL-10)
and NAD(P)H quinone oxidoreductase 1(NQO1) was
increased in 1000 micro-g/m3 exposed splenocytes but
was unaltered in RNA isolated from exposed lung. These
results taken together support the hypothesis that
MWCNT, when delivered in vivo by inhalation, have a
systemic suppressive effect on the immune system. This
suppression may be a result of systemic oxidative stress
and the production of IL-10, a cytokine that regulates cell
mediated immunity.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
13B.1
Relationship between redox activity and chemical speciation
of size-fractionated particulate matter. CONSTANTINOS
SIOUTAS, Leonidas Ntziachristos, University of Southern
California,; John R Froines, Arthur K Cho, UCLA.
13B.2
Correlation of atmospheric ultrafine particle ferrous iron
and mitochondrial toxicity. ANNE M. JOHANSEN,
Stephanie L. Bryner, Eric L. Bullock, Justin M. Johnston, Carin
Thomas, Josie K. Wells, Central Washington University.
Background
Atmospheric ultrafine particles (UFPs, <0.1 micro-meter
diameter) have been shown to induce oxidative stress in
murine macrophages and bronchial epithelial cells and to
disrupt mitochondrial membrane ultrastructure. To
further understand the mechanisms that control UFP
toxicity, ambient UFPs were collected in rural
Washington State and exposed to bovine heart
mitochondria. Reactive oxygen species production, lipid
peroxidation and electron transport chain function were
monitored throughout the experiments and results were
compared with UFP ferrous iron concentrations as
determined spectrophotometrically. Other analyses of
UFP surface chemical composition include time of flight
secondary ion mass spectroscopy (TOFSIMS) and x-ray
photoelectron spectroscopy (XPS). Results indicate that
mitochondrial electron transport chain inhibition
correlates with ferrous iron concentrations in UFPs.
Although the mechanisms of airborne particulate matter
(PM) related health effects remain incompletely
understood, one emerging hypothesis is that these adverse
effects derive from oxidative stress, initiated by the
formation of reactive oxygen species (ROS) within
affected cells. Typically, ROS are formed in cells
through the reduction of oxygen by biological reducing
agents, with the catalytic assistance of electron transfer
enzymes and redox active chemical species such as redox
active organic chemicals and metals. This study aims at
relating the chemical composition of ambient sizefractionated fine particles to their redox activity,
determined by the dithiothreitol (DTT) assay.
Results
Size-fractionated (i.e. <0.15; <2.5 and 2.5 - 10 micro
meters in diameter) ambient PM samples were collected
from four different locations in the period from June 2003
to July 2005, and were chemically analyzed for elemental
and organic carbon, ions, elements and trace metals and
polycyclic aromatic hydrocarbons. The redox activity of
the samples was evaluated by means of the dithiothreitol
activity assay and was related to their chemical speciation
by means of correlation analysis. Our analysis indicated a
higher redox activity on a per PM mass basis for ultrafine
(<0.15 micro meters) particles compared to those of larger
sizes. The PM redox activity was highly correlated with
the organic carbon (OC) content of PM as well as the
mass fractions of species such as polycyclic aromatic
hydrocarbons (PAH), and selected metals.
The results of this work demonstrate the utility of the
DTT assay for quantitatively assessing the redox potential
of airborne particulate matter from a wide range of
sources. Studies to characterize the redox activity of PM
from various sources throughout the Los Angeles basin
are currently underway.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
13B.3
Personal Exposures and Cardiopulmonary Responses of
Children Riding Diesel Powered School Buses, A Pilot Study
(Phase II). Xing Sheng, Sheela V Surisetty, Xiaodong Zhou,
Bozhao Tan, Emily MacWilliams, Ryan LeBouf, Stephanie
Schuckers, Alan Rossner, Andrea R. Ferro, PETER A.
JAQUES, Clarkson University.
Children that ride the bus to school are continuously
exposed to exhaust particles within the cab. Exposures
can be highest while on the bus, and can persist during
periods of idling. Time on buses in the morning and
afternoon may range from several minutes to more than
an hour. Acute exposures to diesel exhaust particles
(DEP) may result in cardiovascular and respiratory
symptoms. Children residing in close proximity to
roadways have shown corresponding acute responses,
such as pulmonary inflammation, exacerbations of
asthma, and changes in pulmonary function.
Inflammatory biomarkers, such as oxides of nitrogen can
be elevated in asthmatic children exposed to DEP, and
normal variability in heart rate may be compromised as a
function of respiratory induced sinus arrhythmia. For
example, asthmatic children have been shown to have a
seasonal variation in HRV (Kazuma et al., 200, 2001). In
this study, 20 healthy third and forth graders carried a
hand held condensation particle counter (CPC) and
Dustrak on the bus, or while walking to school, to the
classroom and back home as a measure of continuous
personal exposures. A LifeShirt, which measures
continuous electrocardiogram (ECG), respiration, and
accelerometry, was worn throughout the day. Five times,
exhaled breath condensate (EBC) and exhaled nitric oxide
(eNO) was collected as biomarkers of pulmonary
inflammation: immediately before and after both
commutes between school and home, and immediately
before lunch. The bus route was tracked with a global
positioning system. Pick-up and drop-off times reflect incab spikes of ultrafine particle concentrations of 100,000
particles/cc with slow decays to between 10,000 and
20,000, reflecting influence of DEP infiltration and its
persistence corresponding to the frequency of pickups.
Exposures to DEP and cardiopulmonary outcomes on a
continuous and integrated basis will be compared and
presented.
13B.4
Applying the thermal optical transmittance (TOT) method
for estimating elemental carbon particle concentrations in
biological samples. Rajiv Saxena, Jawaharlal Nehru
University; Ian Gilmour, MICHAEL HAYS, U. S.
Environmental Protection Agency.
Inhalation of submicrometer soot particles - also referred
to as elemental [EC] or black carbon [BC] - poses serious
human health risks. Yet, there is scant quantitative
information about soot deposition and retention in lung
tissue and its subsequent impact on health. To address this
challenge, a novel bio-analytical technique for
quantifying soot carbon deposits in biological samples
was developed. This study investigates the technique's
ability to isolate exogenous diesel engine particle EC
from lung epithelial cell, alveolar macrophage, and tissue
cultures using a series of chemical and physical
pretreatment steps, and to subsequently measure the
isolated particle EC concentration using thermal optical
transmittance (TOT). We demonstrate how the sample
pre-treatment steps developed permitted us to disregard
the artifacts normally associated with the TOT laser-based
char correction. We also focus on how the TOT response
to different biological and particle matter sample matrices
influenced method development. Results from applying
the new technique showed that (i) diesel engine and
control particle EC mass uptake by cultured lung
epithelial cell lines and alveolar macrophages can be
reproducibly estimated; (ii) biological uptake of particle
EC is dose and time dependent; (iii) macrophages
consume diesel engine and control particle EC with equal
efficiency; (iv) and LA4 cells ingest substantially more
diesel engine particle EC than control particles. The
method's potential to evolve as a valuable research and
diagnostic tool in health studies of fine particulate matter
air pollution will be illustrated.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
13B.5
Reduction of Fe3+ by Elemental Carbon and Its Implication
in the Health Effects of Aerosols. BING GUO, Stephen
Drake, Texas A&M University, College Station; Airat
Khasanov, John Stevens, University of North Carolina,
Asheville.
Fe3+ can be reduced to Fe2+ by elemental carbon in an
aqueous solution. Atmospheric aerosols contain both Fe3
+ and elemental carbon. When the particles are uptaken
by a cell, the Fe3+ may become bioavailable and be
reduced to Fe2+ by the elemental carbon. The Fe2+ can
then inflict oxidative stress through the Fenton reaction.
To assess the significance of this reaction in the health
effects of aerosols, we measured the oxidation state of
iron in laboratory-generated flame aerosols and NIST
standard reference materials, using Mossbauer
spectroscopy and a spectrophotometry method combined
with acid extraction. In Fe-containing soot particles
generated from a laboratory flame, the iron was mostly
Fe3+, with some metallic iron, as shown by the
Mossbauer results. However, the iron extracted from the
soot and the NIST SRMs all had a significant Fe2+
fraction. These results suggest that the reduction of Fe3+
to Fe2+ by elemental carbon is potentially important in
the health effects of aerosols that contain both elements,
and further research is needed to investigate this effect.
13B.6
The relationship between particle active surface area,
number and respirable mass concentration in an automotive
foundry and engine machining facility. WILLIAM A.
HEITBRINK, University of Iowa; Douglas E. Evans, ;Bon Ki
Ku, National Institute for Occupational Safety and Health;
Andrew D. Maynard, Woodrow Wilson International Center for
Scholars; Thomas M. Peters, University of Iowa; Thomas J.
Slavin, International Truck and Engine.
Concentration mapping performed at a co-located
automotive engine plant and foundry resulted in 891
simultaneous measurements of particle number, active
surface area, and respirable mass concentrations. Aerosol
concentrations were measured with a 15-channel optical
particle counter (OPC) used to estimate respirable mass
concentration, a condensation particle counter (CPC) used
to obtain number concentration, and an instrument which
used diffusion charging to measure the active surface area
concentration. At selected locations, particle size
distributions were characterized with the optical particle
counter and an electrical low pressure impactor. Resulting
data were analyzed to evaluate whether particle number
concentration and respirable mass concentrations were
predictive of surface area concentration. Statistical
analyses demonstrated that active surface area
concentration was essentially independent of respirable
mass concentration. However, active surface area
concentration was correlated with ultrafine particle
number concentration. Correlation was stronger during
the winter (R2 = 0.6 for both plants) than in the summer
(R2= 0.38 and 0.36 for the foundry and engine plant
respectively). The larger value of R2 was attributed to
aerosol generated by direct-fire gas fired heaters used
during winter. Generally, particles from the gas heaters
were much smaller (number mode between 0.007 and
0.023 micrometers) than the aerosol generated by routine
foundry and engine plant operations (number mode
between 0.023 and 0.05 micrometers). Optically derived
surface areas, for particles larger than 0.3 micrometers,
were estimated from optical particle counter number
measurements and converted into corresponding surface
areas. During summer measurements, the ratio of optical
surface area to active surface area had a geometric mean
of 2.37 indicating that active surface area is not
predicative of surface area derived with OPCs. Overall,
active surface area concentration is a concentration metric
that is distinct from particle number concentration and
respirable mass concentration.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
14B.1
Recent Advances in Mathematical Modeling of Lung
Deposition of Inhaled Particles. CHONG KIM, USEPA
National Health and Environmental Effects Research
Laboratory; Jung-Il Choi, North Carolina State University.
Dose of inhaled particles is an important factor for
determining toxic effects of inhaled pollutant particles on
one hand and efficacy of drug aerosols on the other hand.
Over the last decades many mathematical models have
been proposed with varying designs of lung morphology
and computational scheme. Most models use Weibel's
symmetric lung morphology for its simplicity, but
asymmetric lung morphology and multi-lobe models are
also used for more realistic and versatile modeling. Under
steady state transport conditions deposition is calculated
by sequential filtration method. Dynamic transport
methods, however, is used to solve a set of time-variant
transport equations allowing for investigating effects of
varying inhalation patterns. Both deterministic and
stochastic approaches have been used in model design.
Stochastic approach is particularly useful for investigating
effects of random variation of lung morphology and
airflow pathways. Although mathematical models provide
a convenient means of estimating the dose at varying
inhalation conditions, all models use simplified lung
morphology and idealistic flow conditions and as such, no
mathematical models are considered complete. Models
need continuing validation and adjustments and
modifications as new experimental data become available.
Recently, we have developed a versatile model capable of
handling a variety of inhalation situations; different lung
morphology (single path vs. multi-path), inhalation wave
patterns, oro-nasal simultaneous breathing, multi-modal
polydisperse aerosols, single vs. multiple breaths
inhalation and whole vs. partial volume aerosol. The
results show excellent agreement with experimental data
available. The present model study will be discussed in
comparison with other models that have been widely
used. This is an abstract of a proposed presentation and
does not necessarily reflect EPA policy.
14B.2
Airflow and Particle Deposition in the Central Airways of
the Human Lung. KAMBIZ NAZRIDOUST, Bahman
Asgharian, CIIT at the Hamner Institutes for Health Sciences
Detailed solutions of airflow and particle transport are
needed for accurate assessment of the deposition of
airborne pollutants in the lung. While a number of studies
have recently been conducted on this topic, the
significance of lung physiology that presents itself
through airflow boundary conditions has often been
overlooked. In this work, airflow and particle deposition
in the central airways of the human lung were studied.
Nine common airways of the human lung were included,
consisting of the trachea, main, lobar, and segmental
bronchi connected as a branching network of cylindrical
tubes with dimensions based on morphometric
measurements. Airflow fields in this geometry were
solved numerically for a 2-second inhalation under three
different boundary conditions: (a) prescribed flow
entering the trachea (inlet) with atmospheric pressure at
the exit to distal airways (outlets), (b) atmospheric inlet
pressure with prescribed outlet flows, and (c) variable
pressure at the outlets induced by attaching expanding
lobes to the distal end of each segmental bronchus. To
evaluate particle deposition patterns, spherical particles of
sizes from 1nm to 10Вµm were injected at the inlet to the
trachea. A Lagrangian particle tracking method was used
that included particle inertia, gravitational settling, and
Brownian motion. The resulting airflow and deposition
patterns were different among all cases, which indicated
the importance of selecting the proper boundary
conditions. These results could assist in identifying
preferred deposition sites in the respiratory tract and
thereby helping to set standards for minimizing exposure
to environmental pollutants and also aiding in improved
efficiency of drug delivery for inhalation therapy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
14B.3
The Comparison of Fiber Deposition in the Human Nasal
Airway. WEI-CHUNG SU, Yung Sung Cheng, Lovelace
Respiratory Research Institute.
Many occupational lung diseases are associated with
exposure to aerosolized fibers in the workplace. The nasal
airway is a critical route for fiber aerosol to enter the
human respiratory tract. The efficiency of fiber deposition
in the nasal airway could directly indicate the fraction of
the inhaled fiber transported to the lower airway. In this
research, a large number of fiber deposition experiments
were conducted to compare the deposition characteristics
of different fiber materials in the human nasal airway.
Carbon, glass, and TiO2 fibers were used as test
materials. Deposition studies were carried out by
delivering aerosolized fibers into a human nasal airway
replica at constant human inspiratory flow rates ranging
from rest state (15 l/min) to moderate exercise (43.5 l/
min). The results showed that the deposition efficiency of
the carbon fiber increases as the fiber impaction
parameter increases. Many carbon fibers were found
deposited in the anterior region of the nasal airway. In
contrast, very few glass or TiO2 fibers were found
deposited in the nasal airway, and most the fibers were
able to pass through the entire nasal airway. These results
imply that the inhaled glass and TiO2 fibers could enter
the human lower airway relatively easily compared to
carbon fiber.
14B.4
Transport and Deposition of Ellipsoidal Fiber in Human
Tracheobronchial Tree. LIN TIAN, Goodarz Ahmadi, Philip
K. Hopke, Clarkson University; Yung-Sung Cheng, Lovelace
Respiratory Research Institute.
Elongated fibers are hazardous to human health due to the
vulnerability of removing it from the respiratory system
once inhaled. The high occurrence of bronchial carcinoma
and lung cancer in certain occupational environment is
linked to human exposure to these substances. Due to the
anisotropy of fiber geometry, very limited work has been
conducted to study its dynamical behavior in human
airway passages. In this study, Lagrangian simulation of
ellipsoidal fiber transport and deposition in human
tracheobronchial tree is presented. The computational
model accounts for the hydrodynamic drag and torque,
shear induced lift, gravitational sedimentation and
turbulence diffusion. The coupled translational and
rotational motions of the fiber are resolved in the model
formulation. The computational model provides a detailed
description of the fiber's rotational characteristics during
its motion. The influences of the fiber's diameter, aspect
ratio, fiber inertia, and the intensity of shear field on fiber
motion are analyzed. The airflow is assumed quasi-steady
during inhalation process under light, moderate and
intensive breathing conditions. The transport and
deposition mechanisms of elongated particles in human
lung are studied. The deposition patterns of ellipsoidal
fibers in the human tracheobronchial tree are compared
with the spherical particles', and their differences are
discussed.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Health Related Aerosols
2007 AAAR Annual Conference Abstracts
14B.5
Development of a two-phase drift flux model for the
deposition of fine respiratory aerosols with comparisons to
experimental results. P. WORTH LONGEST, Virginia
Commonwealth University; Michael J. Oldham, University of
California, Irvine (currently Philip Morris USA).
The transport and deposition of fine aerosols in the upper
respiratory tract is governed by convective, diffusive, and
inertial transport mechanisms. However, continuous twophase models of submicron respiratory aerosols typically
neglect particle inertia effects. The objective of this study
is to develop and test a continuous two-phase model for
simulating the regional and local deposition of dilute fine
aerosols in an idealized bifurcation geometry. To
evaluate the developed transport model, novel in vitro
deposition results for 400 nm particles have been
determined in a double bifurcation geometry of
respiratory generations G3-G5. In addition, previously
reported local deposition characteristics for 1 micrometer
aerosols have also been considered. Computational twophase models that have been evaluated include a standard
chemical species (CS) mass fraction approximation, the
drift flux (DF) approach to account for finite particle
inertia, and a novel extension of the drift flux model to
correct for near-wall particle velocity. The velocity
correction model (DF-VC) applies a sub-grid near-wall
Lagrangian solution to determine particle velocity at
initial contact with the wall. Localized experimental
results for the deposition of 400 nm particles indicated
elevated deposition contours ranging from 1-5% of total
deposition at the first bifurcation and 0.1-1% at the
second. Of the computational models tested, the DF-VC
method provided the best match to experimental
deposition values on a regional and highly localized basis.
Specifically, the DF-VC model matched regional
experimental deposition results to within 10% for both
400 nm and 1 micrometer particles. Considering the local
deposition of fine aerosols, the DF-VC model matched
the experimentally determined elevated contours at the
first and second bifurcations for both 400 nm and 1
micrometer particles. In conclusion, a drift flux particle
transport model with near-wall velocity corrections
appears to provide a highly effective solution for the
deposition of fine respiratory aerosols.
14B.6
Micro- and Nano- Particle Deposition in Human
Tracheobronchial Airways. ZHE ZHANG, Clement
Kleinstreuer, North Carolina State University.
In total and regional lung deposition models, the particle
deposition in each airway generation is typically computed with
analytical equations which were developed for simple
geometries, e.g., straight circular tubes or bent tubes. Clearly,
the use of analytical formulas based on simple tube models for
predicting deposition in local bronchial airway segments has to
be carefully examined considering in light of the actual
complex geometric features, realistic inlet conditions, and airparticle flow characteristics. In this study, inhalation and
deposition of both micro- and nano-sized particles are
numerically simulated for a human tracheobronchial airway
model, starting from the trachea to generation G15 employing 3D bifurcating airway geometries. Specifically, the conducting
zone, in terms of G0-G15 is subdivided into five blocks, or
levels, which are approximated by \triple-bifurcation units\
(TBUs). Thus, air-particle outflow conditions of the oral/nasal
airways are adjusted as inlet conditions for G0-G3, which at
their outlets are again adjusted to become inlet conditions for
G3-G6, and so on.
Using a commercial finite-volume software with user-supplied
programs and an in-house, parallelized particle trajectory code
as solvers, validated solution approaches, i.e., Euler-Euler (for
nano-particles) and Euler-Lagrange (for micro-particles), are
employed with a low-Reynolds-number k-omega model for
laminarto-turbulent airflow transitions. Validated computational
results are obtained in terms of particle distributions and
deposition patterns, deposition fractions, efficiencies as well as
deposition enhancement factors. Both the essential (averaged)
and variable (local) features of each indicator are analyzed \in
series\ and \in parallel\ under different inspiratory flow
conditions and compared to results obtained with analytical
deposition formulas. Effects of branch orientation are discussed
as well, and the deposition parameters are correlated with airway
geometric features, particle characteristics and local flow rates.
Finally, the computer model predictions of total deposition in
the human tracheobronchial airways are compared with: (i) the
new correlation equations; (ii) available experimental
measurements; and (iii) other mathematical modeling results.
This study may provide useful information for both health
assessments of inhaled toxic particulate matter as well as
optimal drug aerosol delivery via inhalation.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2P.1
Design and Development of a Passive Large Particle
Impactor. SANG-RIN LEE, Suresh Dhaniyala, Thomas M
Holsen, Clarkson University.
2P.2
Wind Tunnel Evaluation of a Novel Large Particle Inlet
(LPI). SANG-RIN LEE, Suresh Dhaniyala, Thomas M Holsen,
Clarkson University.
In this study a new passive, large particle impactor (PLPI)
(PM<10) was developed and evaluated in a wind tunnel.
The PLPI is designed to be isoaxial and isokinetic to
enable large particles sampling without bias. The PLPI
can provide representative sampling without requiring a
mass flowmeter controller or pump. The PLPI is
especially useful for aerosol wind tunnel studies to
evaluate particle generation techniques and characterize
particle concentrations and spatial uniformity at the test
location.
Previously a novel Large Particle Inlet (LPI) was
designed and developed using CFD simulation. Design
parameters which were found to affect the performance of
the LPI were slot entrance width, inner wall shapes and
sampling flowrate. Inlet efficiency curves as a function of
Stokes number and wind speeds were derived empirically
based on CFD simulation results. CFD simulations
showed that the LPI can sample particles up to 100 micrometer at low ambient wind speeds. The objective of this
study is to evaluate the performance of the LPI with wind
tunnel experiments. In order to accomplish this,
techniques were developed to generate large particles of
known size uniformly in the wind tunnel. These
techniques had to work for particles up to 100 micrometer with which is especially difficult because of their
large inertia, large settling velocity, and long relaxation
time. Several injection methods including the use of a
Dust feeder loaded with Arizona road dust (coarse grade)
or glass beads (5 micro-meter <dp<100 micro-meter) and
Vibrating Orifice (Oleic acid) were investigated. Several
injection configurations were tested, including: placing
the dust feeder inside the wind tunnel, aerosol injected
from the side wall, and particles injected against the flow
direction. More details about the generation method and
LPI evaluation will be addressed in the presentation.
The original passive impactor, developed in 1941 by May,
was designed to sample particles larger than 30 micrometer at wind speeds greater than 3.6m/s. The PLPI
improves on this original design. Its cut-off size is
reduced to 13.5 micro-meter at a wind speed of 3.5m/s.
The PLPI is an open-design, flow-through impactor that
enables near isokinetic sampling conditions, when
directed towards the wind. Numerical simulations
suggest that the PLPI has a sharp collection efficiency
curve. In this study, the design of PLPI will be presented.
Design parameters such as S/W, W/T, and W/P, where S
is jet to plate distance, T is the length of throat, W is
throat diameter, and P is plate diameter, were investigated
using impaction theory and computation fluid dynamics
(CFD) modeling. Results of the PLPI calibration in a
wind tunnel using the Aerosol Particle Sizer (APS, TSI)
will be presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2P.3
Tribulations in the Development of an Aerosol
Concentrator. DAVID ALBURTY, Zachary Packingham,
Alburtylab; Andrew Page, Page Applied Research.
2P.4
Improved Versatile Aerosol Concentration Enrichment
System (iVACES). YONGJING ZHAO, Boris Grits, Anthony
S. Wexler, University of California-Davis.
Development of an air-to-air concentrator is described,
that operates by charging aerosols drawn into the device
and removing a portion of air through a charged screen
into a \major flow\. In theory, this results in the
concentration of the particles into a much smaller \minor
flow.\ The inlet must be designed to provide sufficient
charge to the target particles to provide adequate mobility
to resist the air flow through the major flow. This process
potentially can greatly multiply the concentration of
aerosols with diameters from 0.5 microns up to 10
microns in the minor flow. In practice, many designs
failed to produce the anticipated results.
Numerous epidemiological studies demonstrate that
elevations in PM10 and PM2.5 are correlated to increases
in acute morbidity and mortality in the population and
that children growing up in more polluted environments
experience reduced lung function, which may predispose
these populations to acute effects. Yet, the vast majority
of the human population and typical animal models do not
elicit measurable physiological changes to normal levels
of air pollutants. One approach to understand these health
effects is to expose animal models to particle
concentrations greatly in excess of ambient by means of
particle concentrators. The Versatile Aerosol
Concentration Enrichment System (VACES) designed by
USC is portable and relatively inexpensive so is very
popular with toxicologists and others who are
investigating the health effects of ambient PM.
Under CARB funding, we have tested a number of
operating parameters for VACES finding that although it
operates as claimed under the meteorological conditions
prevailing in southern California, it fails to concentrate
well under a range of conditions include those that often
prevail in northern California. Also, the VACES design
and construction can be improved to make it more robust.
In this work, we made a number of improvements to the
VACES, called iVACES, to increase the range of
meteorological conditions under which it can concentrate
ambient particles.
The iVACES consists of a saturator which increases the
humidity of the particle-laden inlet air, a cooler in which
the particles grow by water condensation, a low-pressuredrop virtual impactor, and a humidity/heat exchanger
which balances the humidity and temperature with
ambient air and subsequently return particles to their
original size. The iVACES was designed based on
theoretical evaluation of different air temperature and
airflow Reynolds number regimes, as well as on the
analysis of the current VACES. Results of theoretical
calculations and laboratory tests on iVACES are
presented.
The parameters controlling the enrichment of particles in
the gas stream by this method are complex; and include
the initial charge on the particles, charges taken up by the
particles during active charging, the repelling force
developed at the screen, and the electrical/physical design
of the system to optimize the collection of particles into
the minor flow. Many aerosol collection/detection
systems could use the potential advantage of preconcentration of aerosols to improve their performance.
Scaling has been demonstrated from 5-400 LPM (major
flow) and 0.25-190 LPM (minor flow).
Each design begins with determination of the required
total system flow rate and major to minor flow ratio.
Then, a CFD model is constructed; the device is built to
specifications and demonstrated at the desired flow rates
and particle sizes to verify its performance. The model
must then be adjusted to explain the sometimes
unexpected experimental results. Several prototypes are
discussed. An electrostatically driven air-to-air
concentrator may hold significant advantages over
traditional concentrators, such as virtual impactors;
namely, reduced power requirements, elimination of small
clog-prone paths, and reduced fabrication cost.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2P.5
A Model for Designing Sampling Cyclones with Specific
Cutpoint and Slope. THOMAS PETERS, The University of
Iowa; Lee Kenny, Health and Safety Laboratory; Robert
Gussman, BGI Inc.
Experiments were carried out on a new addition to the
sharp-cut-cyclone family of cyclones to determine how
collection characteristics change with dimension and flow
rate. The collection efficiency curve of the cyclone was
determined with respect to aerodynamic particle size at
three flow rates. Measurements were made with an
aerodynamic particle sizer. An empirical model was fitted
to these new data and that from previously tests of
cyclones within this family. The new model enables a
cyclone to be designed with known cutoff characteristics
(50% cutoff diameter and slope of the collection
efficiency curve) for a given flow rate. It extends the
capability of the original model, presented by Kenny and
Gussman (2000), in both flow rate and particle cutoff
diameter to include greater flow rates and particle
cutpoint diameter of up to 15 micro-meter.
L.C. Kenny and R.A. Gussman (2000) A direct approach
to the design of cyclones for aerosol-monitoring
applications. J. Aerosol Sci. 31(12)1407-1420.
2P.6
A New Instrument for Large Particle (10-100 micron) Sizesegregated Analysis. KRISHANU BANERJEE,Sang-Rin Lee,
Suresh Dhaniyala,Thomas Holsen,Clarkson University.
Large Particles (10-100) microns) are major constituents
of dry deposition. The objective of this research is to
obtain size distribution of large particles so that their net
deposition rates can be accurately determined. The
existing instruments are largely designed to sample and
analyze particles smaller than 10 micron. Here, we
present a new instrument design for size-segregated
analysis of particles larger than 10 microns. In the
proposed instrument a curved pipe (90o bend) geometry is
considered. The sampled aerosol particles are injected
into clean sheath air in the instrument. Particles are
gravitationally sedimented into the clean flow as they the
instrument curves from a vertical to horizontal
orientation. The location of sampling channels along the
curve edge results in size-segregated collection of
sampled particles. Numerical modeling was used to
determine the effect of pipe geometries (circular and
rectangular cross-sections) on the collection
characteristics. The rectangular cross sectional geometry
is seen to separate particles with higher resolution than
the circular geometry. This is because, in the circular
cross sectional pipe counter rotating vortexes are formed
for typical flowrates of interest. To minimize secondary
flow contribution with circular geometries, small Dean
numbers must be maintained, but this is not practical for
ambient applications. The resolution of the instrument is
seen to be a function of aerosol to sheath flow ratio, pipe
curvature ratio, classifier aspect ratio, particle settling
velocity and the sampling flow ratio. Details of the
instrument geometry, sizing performance, and windtunnel test results will be presented.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2P.7
Use of CFD for Design of Circumferential Slot Virtual
Impactors. SHISHAN HU, Daniel LaCroix, Clinton Adams,
John S. Haglund, Andrew R. McFarland, Texas A&M
University.
Numerical simulation techniques were used to in design if
circumferential-slot virtual impactors (CSVIs), which are
employed to concentrate bioaerosol particles in the size
range of 1-15 micrometers. The CSVI's have nominal
flow rates of 10 and 100 L/min and operate at 10% minor
flow ratios. Fluent Version 6.2 was used to analyze
problems encountered with the 100 L/min unit, including
unstable flow and wake-flow deposition. For an earlier
version of the 100 L/min, CFD showed an asymmetric
flow structure in the axisymmetric geometry. Unsteady
simulations showed the jet in the receiver nozzle
fluctuates with time, whereas the flow in the 10 L/min
unit was stable. In experimental tests, the symptoms of
the instability in the 100 L/min unit were a pulsing noise
and heavy internal deposition of particulate matter. That
unit was found to have an adverse pressure gradient in a
large receiver region, while the 10 L/min has a much
more favorable pressure gradient and a smaller receiver
section. The 100 L/min unit was re-designed to have a
higher jet velocity and a smaller receiver section.
Physical experiments with the re-designed system showed
there to be no noise and a minor flow transmission
efficiency for 10 micrometer particles of 80%, which is
considerably higher than the value of 30% determined for
the earlier unit at the same particle size. However it is
lower than the 95% predicted from numerical methods. It
was noted during the experiments that there was particle
deposition on surfaces in the wake region of the posts that
are used to connect and align the two halves of the CSVI.
CFD found a strong wake flow downstream of the posts,
which can propagate into the receiver nozzle and could be
the cause of the particle deposition. The posts were
moved outward, and physical tests on a prototype showed
95% transmission efficiency for 10 micrometer particles.
The prototype has a wide dynamic range, i.e., it transmits
particles in the Stokes range of 1-100 with efficiencies
larger than 50%. Pressure drop is only 2-inches of so the
unit can be battery operated in the field.
2P.8
Transmission Efficiency of a PM2.5 Aerodynamic Lens:
Comparison of Model Calculations and Laboratory
Measurements. DAGMAR TRIMBORN, Leah R. Williams,
Achim M. Trimborn, Timothy B. Onasch, John T. Jayne,
Douglas R. Worsnop, Aerodyne Research, Inc.; Jennifer P.
McInnis, Cornell University; Dahai Tang, Kenneth A. Smith,
Massachusetts Institute of Technology.
We present new results for the transmission of the
aerodynamic lens system for larger particles into the
Aerodyne Aerosol Mass Spectrometer (AMS). This lens
system is designed for the transmission of relatively large
particles (but less than 3 micrometers) by operating in the
10-15 Torr range and was investigated with
computational fluid dynamics (CFD) calculations and
experimental measurements. The CFD calculations
include the critical orifice, a valve assembly and the
aerodynamic lens itself and predict near unit transmission
in the particle size range from 70 nanometer to 2.5
micrometers. The beam quality was qualitatively
investigated by visually observing collected polydisperse
ammonium nitrate particles on a substrate. Quantitative
experiments of the particle collection efficiency as a
function of particle size were conducted with
monodisperse aerosols of different compounds using an
AMS equipped with a light scattering particle detector.
The calculations and the measurements agree
qualitatively. The visualization experiments show that
extremely high accuracy is required in the machining of
the device. Reducing the lens pressure shifts the
transmission to smaller particles and increasing the lens
pressure shifts the transmission to larger particles as
predicted by the calculations.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2Q.1
The Influence of Ice Crystal Bounce and Fragmentation on
Aircraft-based Optical Particle Probe Measurements.
DEREK J. STRAUB, Susquehanna University; Darrel
Baumgardner, Centro de Ciencias de la Atm
Instruments that use optical scattering to measure particle
concentrations and size distributions are routinely used
for aircraft-based microphysical research. These
instruments provide a valuable, automated, in-situ data
collection technique. However, aircraft-based particle
sampling is challenging and measurement errors can
sometimes be difficult to recognize and diagnose.
Inconsistent measurements in mixed phase and ice clouds
have lead to speculation that ice crystals that impact
surfaces upstream of the sensing optics can potentially
fragment and pass through the sensing volume leading to
an overestimation of particle number concentration.
To investigate potential measurement errors resulting
from ice crystal impact, fragmentation, and rebound, air
flow and particle trajectory modeling has been performed
as part of this study. The specific probes of interest
include the Forward Scattering Spectrometer Probe, the
Cloud and Aerosol Spectrometer, and the Cloud Droplet
Probe. The computational fluid dynamics (CFD) software
FLUENT was used to generate air flow fields around each
of these probes. Operating conditions were selected to
represent cirrus conditions and a domain inlet velocity of
180 m/s was specified. The flow modeling also featured a
Reynolds Stress Model for turbulence parameterization
and wall functions to define near-wall properties.
2Q.2
Aerosol Penetration Through Electoformed Wire Screens.
TAEWON HAN, Sridhar Hari, John S. Haglund, Andrew R.
McFarland, Texas A&M University.
Experimental and numerical investigations undertaken to
characterize aerosol deposition on commercially available
electroformed wire screens. Deposition of particles was
characterized over a size range of 3 to 20 micrometers, a
porosity range of 0.56 to 0.9, a range of wire widths from
35 to 160 micrometers, and superficial velocities from
0.04 to 1.99 m/s. The obtained data encompass a range
of particle Stokes numbers 0.49 to 20 and wire Reynolds
numbers from about 0.2 to 30. Three-dimensional
Computational Fluid Dynamics (CFD) calculations were
undertaken using Fluent (version 6.1.22). Results of
numerical predictions are in good agreement with
experimental data, supporting the utility of the numerical
technique. Equations correlating the actual efficiency with
Stokes number and areal solidity (0.1 to 0.44) were
obtained. For each screen, results showed that beginning
at a critical value of Stokes number, efficiency increased
gradually to its maximum value, which is approximately
the asymptote to the areal solidity. It is shown that data
obtained from experimental and numerical studies
collapse to a single curve if the collection efficiency is
expressed in terms of an empirical non-dimensional
parameter (standardized screen efficiency). Correlations
expressing the pressure loss coefficient, Cp, as a function
of the Reynolds number and areal solidarity were also
generated.
Air velocity fields exported from the CFD analyses were
used to initialize a particle trajectory model that
specifically resolved particle impact and re-entrainment
into the flow field. Trajectory simulations were
performed for particles that were released into the flow
field upstream of the probe inlets. Particles that passed
through the sensing volume after impacting an upstream
probe surface were counted as artifacts. To
comprehensively assess the outcomes of particle/inlet
contact, individual particles were assigned a range of
rebound angles after impact. In this way, upper and lower
limits were placed on the potential for counting and sizing
errors in optical particle probe measurements.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2Q.3
Development and Validation of the Releasable Asbestos
Field Sampler. JONATHAN THORNBURG, Jeremy
Seagraves, RTI International; John Kominsky, Environmental
Quality Management Inc.; John Tish, Tisch Environmental
Asbestos aerosolization (or releasability) is the potential
for fibrous asbestos structures that are present in a
material or on a solid surface to become airborne when
the source is disturbed by human activities or natural
forces. In turn, the magnitude of the airborne
concentration that can be generated from the release of
asbestos is a function of the concentration of asbestos at
the source, source matrix properties, the nature of the
processes disturbing the source, and local environmental
conditions. The primary utility of assessing asbestos
aerosolization relates to the ability to predict airborne
exposure (and associated health risk) from asbestos fibers
on surfaces.
2Q.4
Digital Microfluidic Impactor for Measurements of the
Aerosol Chemical Composition. ANDREY KHLYSTOV,
Ming-Yeng Lin, Randy Evans, Richard Fair, Duke University.
We describe the development and characterization of a
prototype system that integrates a single stage impactor
with a digital micro-fluidic substrate for determination of
sulfate and other ions in aerosol. The aerosol is impacted
directly onto the surface of a Teflon-coated micro-fluidic
chip. After a brief collection phase, the collection surface
is extracted with a micro-liter droplet of ultra pure water
that is digitally directed across the surface dissolving
collected aerosol constituents. The extraction droplet is
then analyzed ether colorimetrically or using on-chip
capillary electrophoresis. In this work we characterize the
impactor cut-off, the extraction efficiency for laboratory
and ambient aerosols and the performance of on-chip
analysis.
Currently available methods for repeatable, representative
measurement of asbestos aerosolization from bulk
materials are not suitable for field use. Current methods
require removal and transport of the source matrix,
thereby potentially altering the matrix physical
characteristics and subsequent aerosolization. The
Releasable Asbestos Field Sampler (RAFS) was designed
to provide repeatable and representative asbestos
aerosolization data from soil in situ. The RAFS simulates
a raking motion to disturb the soil. A gentle airflow
transports the generated aerosol laterally inside a tunnel to
one end where filter samples or real-time instruments are
located.
The RAFS was tested in laboratory and field experiments.
Laboratory data showed the RAFS generated repeatable,
representative aerosol concentrations. Concentrations
varied between 0.3 to 30 particles per cubic centimeter.
The aerosol concentration and size distribution was
dependent on RAFS operating conditions and soil
moisture content. The variability in aerosol generation
for specified test conditions were statistically insignificant
(p-value < 0.05). Field tests showed the RAFS
aerosolized asbestos at similar rates as human activities
and aerosolized concentrations were correlated with total
particle concentrations.
Although this work was reviewed by U.S. EPA and
approved for publication, it may not necessarily reflect
official Agency policy.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2Q.5
Collection Efficiency and Diffusion Broadening in an
Electrostatic Classification Aerosol Inlet for Thermal
Desorption. ANGELA I. SHIBATA, Sonya C. Collier, Denis J.
Phares, University of Southern California.
2Q.6
Development and Experimental Evaluation of Aerodynamic
Lens as an Inlet of Single Particle Mass Spectrometry.
KWANG-SEUNG LEE, Sung-Woo Cho, Donggeun Lee, Pusan
National University.
This study presents theoretical and experimental analyses
of the collection efficiency in a low pressure cylindrical
classification inlet. The device resembles a cylindrical
Differential Mobility Analyzer (DMA) in that a sample
flow is introduced around the periphery of the annulus
between two concentric cylinders, and charged particles
migrate inward towards the inner cylinder in the presence
of a radial electric field. The operating pressure ranges
from 1 to 40 Torr. Instead of being transmitted to an
outlet flow, the sample is collected onto a Nichrome
filament located on the inner cylinder. The primary
benefit of this mode of size-resolved sampling, as
opposed to aerodynamic separation into a vacuum, is that
chemical ionization of the vapor molecules is feasible.
Collection efficiency curves are computed with and
without Brownian diffusion, which becomes more
significant at lower pressures. Since there is no outlet
aerosol flow, experimental characterization of the inlet
requires chemical analysis of the vapor produced from
desorption of particle standards. In this study, polystyrene
latex spheres ranging in diameter from 20 nm to 200 nm
are introduced into the inlet and the resulting vapor is
analyzed using low pressure chemical ionization mass
spectrometry.
In this study, we performed a numerical simulation on
aerodynamic focusing of nanoaerosol particles. The
focusing and transmission efficiencies of single
aerodynamic lens were estimated as a function of particle
size. As a result, we found that particle behavior was
characterized by two dimensionless parameters such as
particle Stokes number and flow Reynolds number when
outer diameter and length of the lens are kept constant.
An optimal design achieved from the single lens analysis
was evaluated by simulating an assembly of those single
lenses. The resultant aerosol beam diameters from
numerical simulation were compared and agreed
reasonably well with those from light scattering imaging
of as well as deposition and observation of aerosol beam.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2Q.7
An Overview of NASA-Sponsored Research to Characterize
and Improve Methods for Measuring Aircraft Particle
Emissions. BRUCE E. ANDERSON, NASA Langley Research
Center; Chowen C. Wey, NASA Glenn Research Center; David
S. Liscinsky, United Technologies Research Center; Anuj
Bhargava, Pratt and Whitney; Phillip Whitefield, University of
Missouri at Rolla; Richard C. Miake-Lye, Aerodyne Research
Inc.; Robert Howard, AEDC/ATA.
Tests conducted to measure the size and concentration of
particles in hot, high-velocity gas turbine engine exhaust plumes
have produced highly variable results, primarily because of a
lack of uniform and well-characterized sampling and
measurement approaches. To quantify the sources of this
variability and to help establish standard measurement
methodology, NASA recently sponsored laboratory and field
experiments to evaluate aircraft exhaust sampling and
measurement systems. The work focused in three primary
areas: assessing size and mass-dependent particle losses through
sample transmission and distribution lines; determining inlet
probe collection and transmission efficiencies; and comparing
the relative and absolute precision of instruments designed to
measure particle concentration, composition, and microphysical
properties. Participants in these tests included researchers from
NASA, the United Technologies Research Center, Pratt and
Whitney, the University of Missouri at Rolla, Aerodyne
Research Inc., and the Air Force's Arnold Engineering
Development Center.
Three separate experiments were conducted during 2006: two at
the NASA Langley and the third at NASA Glenn. The Langley
experiments focused upon characterizing the diagnostic
instruments, sampling probes and transmission lines used during
the NASA-sponsored Aircraft Particle Emission Experiment
(APEX) test series using particles generated by a variety of
sources including a tube furnace, torch, spark source, and a
turbo-jet powered ground support vehicle. The NASA Glenn
tests evaluated a broader assortment of sampling probes
(including uncooled and water-cooled gas and aerosol probes)
and transmission lines (heated and unheated, flexible, small and
large diameter) using a tube furnace, ambient aerosols, the
center's Learjet as particle sources. Component performance
was evaluated (where applicable) as functions of particle size,
concentration, and composition; exhaust gas velocity and
temperature; sample flow rate, temperature and pressure; and
test article heating or cooling. Test results along with a
discussion of recommended sampling and measurements
strategies will be presented.
2Q.8
Sample Line Efficiency Measured with a Real Time
Particulate Size Spectrometer. JONATHAN P.R.
SYMONDS, Jason S. Olfert, Kingsley St.J. Reavell,
Cambustion Ltd, U.K.
Penetration efficiencies of particle lines as a function of
particle size were measured with a DMS500 Fast
Particulate Size Spectrometer. A 25 m length of
conductive silicone tubing was compared with a 1 m
length by passing nebulised, dried and bipolar neutralised
sodium chloride aerosols (in the ca. 10-100 nm size
range) through each sample line into the DMS500. Use of
an up-stream orifice plate allowed the experiment to be
repeated at 0.25 atm as well as at 1 atm pressure. Various
flow rates (<8 slpm) were used, with Reynolds numbers
within the laminar regime. A TSI 3081 SMPS was also
used for a 1.5 slpm flow rate at 1 atm.
Particle penetration vs. size was compared with various
diffusive deposition models. Even for the laminar flows
considered, the laminar flow penetration model given by
Hinds (1999) gives poor correlation with the experimental
data. However, the semi-empirical model for diffusive
deposition velocity in the turbulent regime given by Wells
and Chamberlain (1967), combined with an expression for
penetration efficiency derived from the principle of
conservation of mass, does give good correlation with the
data.
The penetration efficiency of particles from a light-duty
common-rail Diesel engine at fast idle was also measured
with the DMS (the 200 ms time response giving better
measurement of the fluctuating aerosol than the SMPS).
Again, even for low Re, correlation was better with the
turbulent model than with the laminar model. Also
combinations of hot (150 degrees C) and cold conductive
PTFE sample lines and hot and cold dilution air at the
point of sampling were considered, and the results
compared with likely thermophoretic losses.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2Q.9
The Effectiveness of Bubble Aerosol Generators for
Sensitive Bacteria. GEDIMINAS MAINELIS, Heyreoun An,
Rutgers, The State University of New Jersey; Jana Kesavan, US
ARMY ECBC.
Various studies often need bioaerosol generators that
can not only efficiently produce a stable bioaerosol
output, but also minimize injury and inactivation of
sensitive microorganisms. The widely-used nebulization
technique has been shown to cause damage to some
sensitive organisms. As a less injurious alternative, the
bubbling mechanism has been explored for aerosolization
of sensitive microorganisms.
In this research, we used a non-circulating bubble
aerosol generator (NCBG) designed in an earlier study
and evaluated its performance when aerosolizing sensitive
bacteria, Erwinia herbicola. This was achieved by
measuring the percentage of viable Erwinia herbicola in
the generated aerosol and in the generator's reservoir at
different aerosolization times. The same testing protocol
was applied for Collison nebulizer (CN) and a circulating
bubble aerosol generator (CBG), which was built for this
study.
Among the three investigated bioaerosol generators,
the Collison nebulizer (CN) exhibited the highest average
viability at the aerosolization times of 0 and 90 min.
However, the viability of the bacterial aerosol produced
by CN decreased by 43% after 90 minutes of
aerosolization. In addition, the aerosol concentration
produced by the CN has steadily increased during the
aerosolization process and after the 90 min of
aerosolization it was higher by 60% compared to t = 0
min. The non-circulating bubble generator (NCBG), on
the other hand, had produced very stable bioaerosol
concentration during 90 minutes of aerosolization and the
viability of bacterial aerosol has decreased only slightly
(13%) after 90 min of continuous aerosolization. Thus,
the NCBG may provide a more suitable bioaerosol
generator choice when stable bioaerosol concentration
and viability are required over extended periods of time.
The circulating bubble generator (CBG) also provided
aerosol with stable concentration and viability; however,
the aerosol concentration and viability were substantially
lower than that of the other two generators.
2Q.10
Experimental evaluation of electrodynamically focused
nanoparticle behavior in the quadrupole electric field.
JINYOUNG CHOI, Sangsoo Kim, Korea Advanced Institute of
Science and Technology; Seokjoo Park, Korea Institute of
Energy Research.
The aerosol focusing technology is used in many areas
such as particle monitoring apparatus, atmospheric
chemistry and semiconductor processes. An aerodynamic
lens system, which consists of a single or a series of
concentric orifice, is the method that has been most
widely used to produce the focused particle beams. This
aerodynamic focusing method is conceptually simple and
effective for various applications, but has some
limitations. In order to overcome these limits and propose
the alternative way to aerodynamic focusing, we have
already reported the concept of the electrodynamic
focusing using the quadrupole electric field by numerical
simulations in our previous researches. In this study, we
have constructed the actual system based on the results in
previous works and investigated the focusing
characteristics of charged nanoparticles in the quadrupole
electric field experimentally.
A nanoparticle focusing apparatus using the quadrupole
electric field was designed and constructed based on the
numerical simulations. The singly charged monodisperse
particles were prepared through a silver nanoparticle
generator using a small ceramic heater and DMA, and
then entered into the vacuum chamber in which the
quadrupole electrode system was installed. The
surrounding gas pressure at the vacuum chamber and
particle size were each kept below 10torr and 100nm. The
Faraday cup and micro electrometer were used for
detecting the focused particles.
Experiments were performed for different frequencies and
amplitudes of AC voltage applied to the quadrupole
electrodes with the change of the system pressure and
particle size. The experimental results are in good
agreement with the tendency predicted by the numerical
simulation. In conclusion, the possibility of
electrodynamic focusing of nanoparticles using the
quadrupole electric field was validated through the
fundamental experiments.
Copyright В© 2007 by the American Association for Aerosol Research (AAAR).
AAAR hereby grants contributing authors full rights to use of their own abstracts.
Instrumentation
2007 AAAR Annual Conference Abstracts
2R.1
The Use of Gold-Coated Filters to Measure Mercury
Deposition. Ying Liu, JIAOYAN HUANG, Thomas M.
Holsen, Clarkson University.
Mercury (Hg) is a toxic pollutant due to its
bioaccumulation in the food chain and its adverse effects
on human health. In this study, a direct Hg dry deposition
measurement technique using gold-coated filters was
developed. The analysis apparatus was made up of a two
parts quartz furnace. Water was circulated at 50 degrees
Celsius, and contained quartz chips and quartz wool to
enhance the conversion of RGM to Hg0 which was then
analyzed with a Tekran 2537A. An injection port was
added to the apparatus upstream of the quartz furnace.
The zero air flushing flowrate was an important
determinant in mercury recovery. Recoveries of 25, 50
and 75 uL of Hg0 were 88%, 90%, and 92%, respectively
at 1.5 L/min flushing rate, but decreased to 76%, 81% and
86%, respectively at flushing rate of 1.0 L/min. At a
flushing rate of 1.5 L/min, the recovery of mercury in
NIST coal fly ash was approx. 90%. There was no
significant difference between the filters coated with gold
for 1 and 2 min and exposed in a particle-free clean lab
for up to 24 h. Sampling rates (0.56 - 0.83 m3/d), and
RGM and Hg0 transfer velocity (0.32 to 0.56 cm/s)
decreased with increasing sampling time. HgCl2 solution
and mercury in NIST urban particulate mat