Indoor Air 2016
wileyonlinelibrary.com/journal/ina
Printed in Singapore. All rights reserved
© 2016 The Authors. Indoor Air published by John Wiley & Sons Ltd
INDOOR AIR
doi:10.1111/ina.12307
Evaluation of ozone emissions and exposures from consumer
products and home appliances
Abstract Ground-level ozone can cause serious adverse health effects and
environmental impacts. This study measured ozone emissions and impacts on
indoor ozone levels and associated exposures from 17 consumer products and
home appliances that could emit ozone either intentionally or as a by-product of
their functions. Nine products were found to emit measurable ozone, one up to
6230 ppb at a distance of 5 cm (2 inches). One use of these products increased
room ozone concentrations by levels up to 106 ppb (mean, from an ozone
laundry system) and personal exposure concentrations of the user by
12–424 ppb (mean). Multiple cycles of use of one fruit and vegetable washer
increased personal exposure concentrations by an average of 2550 ppb, over 28
times higher than the level of the 1-h California Ambient Air Quality Standard
for ozone (0.09 ppm). Ozone emission rates ranged from 1.6 mg/h for a
refrigerator air purifier to 15.4 mg/h for a fruit and vegetable washer. The use of
some products was estimated to contribute up to 87% of total daily exposures
to ozone. The results show that the use of some products may result in potential
health impacts.
Q. Zhang, P. L. Jenkins
Research Division, California Air Resources Board,
Sacramento, CA, USA
Key words: Ozone; Consumer products; Home appliances;
Indoor air quality; Emission rate; Personal exposure.
Qunfang Zhang
Research Division, California Air Resources Board
1001 I Street, 5th Floor
Sacramento, CA 95814
USA
Tel.: +1-916-323-2257
Fax: +1-916-322-4357
e-mail: [email protected]
This is an open access article under the terms of the
Creative Commons Attribution-NonCommercial-NoDerivs
License, which permits use and distribution in any medium, provided the original work is properly cited, the use
is non-commercial and no modifications or adaptations
are made.
Received for review 11 September 2015. Accepted for
publication 1 May 2016.
Practical Implications
Many consumer products and home appliances can emit ozone either intentionally or as a by-product of their functions, but, other than for air cleaners, ozone emissions have not been measured for these products. This study tested
17 products widely advertised in the market and found that over half of them emitted ozone, and some emitted high
levels. Some products increased room ozone concentrations and personal ozone exposure concentrations to levels that
greatly exceed the level of the 1-h California Ambient Air Quality Standard for ozone (0.09 ppm). The use of some
products can contribute a significant fraction of total daily exposure to ozone and pose potential health risks to their
users. However, none of these products are tested or regulated for their ozone emissions, which is a critical gap in
consumer protection.
Introduction
Ground-level ozone is widely regulated as an air pollutant because exposure to ozone can cause many
adverse health effects, including reduced lung function; increased respiratory symptoms such as cough,
wheeze, difficulty breathing, and chest tightness;
increased airway hyper-reactivity; and increased airway inflammation (California Air Resources Board
(CARB), 2005; U.S. EPA, 2013). In epidemiology
studies, ozone has been associated with premature
death, hospitalization for cardiopulmonary causes,
emergency room visits for asthma, restrictions in
activity, and increased school absences (CARB, 2005;
U.S. EPA, 2013). Research has shown that in addition
to exacerbation of asthma, ozone may play a role in
the development of asthma in children who engage in
extensive outdoor exercise in high ozone communities
(McConnell et al., 2002). In view of the significant
adverse health impacts of exposure to ozone, the
1
Zhang & Jenkins
CARB reviewed the scientific literature in 2005 relevant to California’s outdoor ozone standard and
retained the existing 1-h California Ambient Air Quality
Standard (AAQS) for ozone of 0.09 ppm and adopted
a new 8-h standard of 0.070 ppm.
Ozone also can be generated indoors. Air cleaning
devices that intentionally produce ozone, which are
inaccurately marketed as producing ‘safe’ levels of
‘activated oxygen’ that remove indoor air pollutants
such as particles, gases, allergens, viruses, odorous
compounds, mold, and bacteria, have been found to
increase indoor ozone concentrations to harmful levels
(Britigan et al., 2006; Mason et al., 2000; Poppendieck
et al., 2014; Tung et al., 2005). A study by CARB staff
tested four ozone generators marketed in California
that intentionally emit ozone and found all of the tested
models resulted in room concentrations that exceeded
state and national health-based standards (CARB,
2006). One whole-house model produced a room ozone
concentration equal to a Stage 1 Smog alert. Another
CARB study found that other types of air cleaners such
as ionizers and electrostatic precipitators also emitted
ozone as a by-product, although at much lower levels
(CARB, 2008). To protect Californians from adverse
health effects related to ozone-emitting air cleaning
devices, CARB adopted a regulation (California Code
of Regulations, Title 17, §94800–§94810) in September
2007 to limit the ozone emissions from indoor air cleaning devices to no more than 50 ppb as tested under
Underwriters Laboratories, Inc. (UL, Northbrook, IL,
USA) Standard 867. Over 1000 models of air cleaning
devices have been certified under this regulation since it
became effective (CARB, 2016).
The current CARB regulation does not include other
consumer products and home appliances that may emit
ozone either intentionally or as a by-product of their
functions. Due to its strong oxidative ability, ozone is
widely advertised for disinfection or odor removal by
manufacturers of facial steamers, fruit and vegetable
washers, home drinking water treatment appliances,
deodorizers for refrigerators and shoes, and residential
laundry water treatment systems. Some consumer
products such as hair dryers are designed to emit negative ions which may react with oxygen to produce
ozone. Additionally, there are a few types of air cleaning devices (e.g., personal air purifiers) on the market
that warranted testing because of their ozone-producing claims and their illegal marketing in California.
However, the effects of ozone are not as positive as
claimed by some manufacturers. In terms of odor
removal, ozone only reacts with some gases of concern,
such as some compounds with unsaturated carbon
double bonds (e.g., limonene, pinene, and styrene; Boeniger, 1995; Weschler, 2000). Such reactions produce
other air pollutants such as formaldehyde and ultrafine
particles (Coleman et al., 2008; Nazaroff and Weschler,
2004; Singer et al., 2006), which may also have adverse
2
health impacts (Ibald-Mulli et al., 2002; NTP, 2011;
Oberd€
orster, 2000; Oberd€
orster et al., 2005; Peters
et al., 1997). In addition, there are no recognized
antimicrobial effects of gas-phase ozone at low concentrations on either airborne or surface microorganisms
(Cole, 2003). Only at high concentrations – in the range
of 6000 to almost 10 000 ppb – does ozone significantly
kill fungi and bacteria (Foarde et al., 1997). The benefits of such devices to remove odor or kill viruses and
bacteria as claimed by their manufacturers are not well
supported, and the health problems associated with
ozone exposures may be a greater concern.
Although their market share is unknown, these
devices can easily be obtained due to their low unit price
and widespread advertising via the Internet and television. Thus, the exposures associated with the use of such
products may be widespread. However, no studies were
found that have collected actual ozone emission data
for the products discussed above. The goal of this study
was to investigate the impacts of various consumer
products and home appliances on indoor ozone levels,
and to assess the associated exposures.
Methods
Products obtained for ozone testing
Seventeen products were obtained for ozone testing
(Table 1). These devices were selected because they
either are advertised as producing ozone, or have the
potential to generate ozone as a by-product of their
functions. In addition, they can easily be obtained by
California residents. For each product category, one to
three products from different manufacturers, if available, were obtained to assess the range of ozone emissions from products with similar functions. Duplicate
product units were obtained for three products to
assess the interunit variability of ozone emissions. One
product, PAP2, an air purifier, was removed from the
study after a preliminary test, because it caused electric
shock and shut down the ozone-monitoring instruments during the preliminary test.
Equipment
Ozone concentrations were measured at a rate of once
every 10 s by three 2B Technologies Model 202 Ozone
Monitors. These ozone monitors can detect ozone
ranging from a limit of detection of 1.5 ppb to an
upper limit of 100 ppm. The precision and accuracy
are the higher of 1.5 ppb or 2%. These ozone monitors were calibrated using a gas dilution calibrator
(Model 322; Tanabyte Engineering, Inc., Riverview,
FL, USA) and a Scott-Marrin UltraPure air bottle
(Scott-Marrin Inc., Riverside, CA, USA) prior to this
study and received daily zero checks during the study.
The ozone monitors were collocated to check their
Ozone from consumer products and home appliances
Table 1 Description of products and tests
Category
ID
Description
Refrigerator
air purifier
RAP1a
RAP2
RAP3
A continuously operating product with an ionizer which is advertised as emitting nature’s sanitizer ‘ozone’
A continuously operating product with an ionizer which is advertised as emitting activated oxygen (ozone)
A continuously operating product with an ionizer which is advertised as utilizing ozone purification
technology
A handheld product which is submerged under water and releases ozone and ultrasound waves; two
cleaning settings (low and high); each cleaning cycle lasts for 1.5 min; 2–4 cleaning cycles are
recommended for every 1 lb. of fruit and vegetables. This study used 10 cleaning cycles (15 min total)
for a typical cycle of use
A product with an ozone diffuser plate on the bottom which releases tiny ozone bubbles into water in
a bowl on the top; a built-in timer which can be set to 1–20 min of operation
A product with an on-board ozone generator which generates and diffuses ozone into the water supply
for washing machines. The unit generates ozone only when water runs through it
Fruit and
vegetable
washer
FVW1a
FVW2
Laundry water
treatment
device
Drinking water
treatment
device
OLS
Shoe sanitizer
SS1a
TWS
OWC
SS2
Facial steamer
FS1
FS2
Personal air
purifier
PAP1
PAP2
Ionic hair
device
Sanitizing
wand
IHD1
IHD2
SW1
A product which is connected to a tap water faucet and powered by an internal hydroelectric generator
that produces ozone and diffuses it into the water as it flows through the unit
A tabletop water cooler with an on-board ozone generator which generates ozone and diffuses it into
the water tank
A product with two small UV light bulbs which can be inserted into shoes; automatically powers off
after one cleaning cycle, which lasts for 15 min
A product with a similar size & shape to a small microwave oven which emits ozone into the sanitizing
chamber; automatically powers off after one cleaning cycle, which lasts for 8 min
A table top product with a UV light bulb; automatically turns off when the water level is lower than
the sensor
A table top product with a UV light bulb; automatically turns off when the water level is lower than
the sensor
A continuously operating product that is plugged into a wall outlet and is advertised to generate
negative ions
A table top, continuously operating product that is advertised to generate high-density negative ions,
includes a warning sign on the device about ‘a harmless but uncomfortable static discharge’
if touched or held when it is operating. Caused electric shock when touched, which shut down
the ozone monitor
An ionic hair dryer with multiple levels of output: 3 heat and 2 speed settings
An ionic hair straightener with multiple levels of output: 5 heat settings
A handheld, continuously operating product with a long UV light bulb
ESM
Face
Room
Expo.
Emis.
CE
U
U
U
U
U
U
U
U
U
U
U
U
CE
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
U
CE
U
U
CE
U
U
U
ETL
ETL
U
U
U
ESM: Electrical safety mark. CE or ‘Conformite Europeenne’ mark shows that the product complies with the ‘essential requirements’ of European laws or directives. It also indicates the product’s conformance to legal requirements with respect to safety, health, the environment, and consumer protection in the European Union. The ETL Mark is proof of product compliance (electrical, gas, and other safety standards) with North American safety standards by Intertek.
Face: face test; Room: room concentration test; Expo.: personal exposure test; Emis.: emission rate test.
a
Duplicate product units were obtained to test the interunit variability of ozone emissions.
comparability before experiments started every day.
Figure S1 in the Supporting Information shows good
agreement of the readings from these ozone monitors
(R2 = 1.00 and R2 = 0.99, respectively). The data collected by units B and C were corrected to those of unit
A using the formulas shown in Figure S1 to correct
for minor drift. The ozone monitors were turned on
20 min before recording as required for the lamp,
photodiode, and internal temperature of the absorption cell to stabilize. After the warm-up, the background ozone concentrations were measured for
15 min.
Test methods
All 17 products were first tested for face ozone concentrations in a small enclosed air monitoring shelter
(Room 1). Room 1 is approximately 13.8 m3
(2.4 9 2.4 9 2.4 m) and furnished with a wooden
table and a wooden chair. The flooring is vinyl tile, and
the interior is aluminum with an inactive coating.
Room 1 has a door but no windows. The air exchange
rate (AER) in Room 1 was near zero (0.04 0.03/h).
The product to be tested and an ozone monitor were
placed on top of the table in the center of the room.
The ozone sampling inlet pointed to the air stream outlet of the tested product and measured ozone at
increasing distances along the air stream, for example,
at 5, 10, 20, 41, and 81 cm (2, 4, 8, 16, and 32 inches).
Each measurement lasted for a minimum of 2 min, or
longer if necessary for equipment-specific measurement
stabilization, as required in the UL Standard for Safety
for Electrostatic Air Cleaners, UL Standard 867. Products with more than one operational setting were tested
first at the highest emission setting and then at 1 to 2
lower emission settings.
The products that emitted more than 5 ppb ozone at
5 cm (2 inches) were further tested for their
3
Zhang & Jenkins
contributions to room ozone concentrations and personal exposure concentrations for one cycle of use. The
ozone fruit and vegetable washers, shoe sanitizers, and
facial steamers do not require special installation or
deployment; thus, they were placed on top of a table in
the center of a small meeting room (Room 2). Room 2
is approximately 36.0 m3 (3.0 9 4.0 9 3.0 m). It was
furnished with a wooden table and four upholstered
chairs, and has carpet flooring and painted wallboard
throughout. The tests were conducted with the door
closed, during weekends when the central air ventilation
system was shut down completely. The average AER
was 0.43 0.03/h, within the range of the AERs
observed in typical California homes. Two ozone monitors were employed, one for room ozone concentrations
and one for personal exposure concentrations. The sampling inlet for room ozone was placed at 91 cm (3 ft)
from the product, toward the center of the room, and
fixed at 91 cm (3 ft) above the floor to approximate the
‘breathing zone height’ of children either sitting or
standing. To simulate a user’s exposures to ozone when
operating the product, the ozone sampling inlet of the
second ozone monitor was placed at close proximity to
the product, for example, at 5 cm (2 inches) from the
facial steamers, and 30 cm (1 ft) above the shoe sanitizers and the ozone fruit and vegetable washers. The
research staff left the test room once the measurements
started, except when testing the devices that required
manual operation, such as the handheld fruit and vegetable washer.
The ozone laundry system (OLS) was tested by connecting to a 42.5 l (1.5 ft3) residential top-loading
washing machine in a small bathroom of a volunteer’s
house (Room 3). Room 3 is approximately 10.9 m3
(1.4 9 2.6 9 3.0 m). The room surfaces were comprised of ceramic tiles and painted wallboard. The window and door of the bathroom were closed, and the
central air ventilation system was turned off. The AER
was 0.40 0.01/h. One ozone monitor measured room
ozone concentrations in the center of the bathroom,
and a second monitor measured ozone 30 cm (1 ft)
above the washing machine to measure personal exposure levels of machine users.
The refrigerator air purifiers were placed on the middle shelf of a 600-l (21.2 ft3) refrigerator filled with
food (most in sealed containers), in a small room
(6.0 9 4.0 9 3.0 m; Room 4). The AER was
0.45 0.04/h. The refrigerator door was kept closed
for 8 h while one refrigerator air purifier was operating
inside the refrigerator to simulate overnight use. The
refrigerator door was opened for 5 min to simulate an
episode of loading food. The ozone sampling inlet of
one ozone monitor for personal exposure measurements was placed about 5 cm (2 inches) away from the
opening and at a height of about 122 cm (4 ft) above
the floor to approximate the ‘breathing zone height’ of
a standing adult. Another ozone monitor was placed in
4
the center of the room to determine the contributions
to room ozone concentrations. Preliminary data
showed a large amount of ozone was emitted during
the first half-hour of operation of the refrigerator air
purifiers, so the tests described above were repeated
after each refrigerator air purifier had been turned on
inside the refrigerator for 0.5 h.
Air exchange rates
The AER were estimated using the single zone tracer
gas decay method with carbon dioxide (CO2) as the tracer gas. As described above, four rooms (Rooms 1–4)
were used for different tests. Room 1 is a stand-alone,
tightly sealed air monitoring shelter, so it is truly a single
zone space. Rooms 2 and 4 are located in the internal
area of a high-rise office building; these rooms do not
have direct connection (e.g., windows and doors) to the
outdoors. The tests in these two rooms were carried out
during weekends when the central air ventilation system
was shut down completely; therefore, air exchange
through the central ventilation system was minimal.
Room 3 is a small bathroom inside a two-story residential house with one window that can be opened to the
outdoors and a door that can be opened to a bedroom.
During the tests in this room, the window and door
were closed completely, and the central air heating and
air conditioning system was turned off to minimize air
exchange between the test space and other rooms.
The AERs in each test space were determined before
experiments started each day. CO2 concentrations in
the test space were increased to levels above
1500–2000 ppm by dry ice sublimation or occupant
exhalation. A small table fan was used to mix the air
inside the test space. CO2 concentrations were measured
in the center of the test space using a QTrak (Model
8554, TSI) for a decay period of 30 min. Meanwhile,
CO2 concentrations outside of the test space, which were
the ambient concentrations for Rooms 1 and 3, and the
concentrations in the hallway outside of the rooms for
Rooms 2 and 4, were measured by a second QTrak.
Then, the AERs were calculated using the dynamic
method by Baptista et al. (1999). The accuracy of the
QTrak is (3% of reading + 50 ppm) at 25°C.
Ozone emission rates
For the products with more than 5 ppb ozone at 5 cm
(2 inches), ozone emission rates were tested three times
each in Room 1 described above. One exception is the
OLS, as it needs to be connected to a washing machine
which could not fit in Room 1. A small table fan was
used to mix the air inside the room. Each product to be
tested was placed on top of the table in the center of
the room. Each refrigerator air purifier was turned on
for 4 to 5 h. For the other products, each product was
operated for one cycle of use, ranging from 15 to
Ozone from consumer products and home appliances
RT
20 min. An ozone monitor was placed at about 91 cm
(3 ft) from the product, toward the center of the room,
and at 122 cm (4 ft) above the floor to measure indoor
ozone concentrations. Another ozone monitor measured the outdoor ozone concentrations simultaneously via tubing that passed through a small hole in the
wall of the room to the outdoors.
A mass-balance model for a single zone was used to
calculate the ozone emission rate:
dCin E
¼ þ Cout pAER Cin ðAER þ DÞ
ð1Þ
V
dt
where E is the ozone emission rate (lg/h), Cout is the outdoor ozone concentration (lg/m3), p is the ozone penetration factor (dimensionless), Cin is the indoor ozone
concentration (lg/m3), D is the ozone deposition rate in
Room 1 (/h), and V is the air volume inside Room 1 (m3).
By multiplying all terms by dt and integrating over
the period of one cycle of use, Equation 1 is transformed to:
ZT
ZT
ZT
E
dt þ Cout pAERdt
dCin ¼
V
0
0
0
ð2Þ
ZT
Cin ðAER þ DÞdt
0
0
Fruit and vegetable washers
(a)
þ
ZT
dt
0
VðAERþDÞ
T
ZT
Cin dt ¼
VðCinðTÞ Cinð0Þ Þ
T
ZT
Cin dt
ð3Þ
0
Since high-resolution data (10 s intervals) were
RT
obtained
for Cin ; 0 Cin dt
is
estimated
as
Pn
1 ½ðCinði1ÞþCinðiÞ Þ=2Dt, where Cinði1Þ and CinðiÞ are the
(i 1)th and ith data points of Cin, Dt is the interval
between two data points (10 s, or 0.0028 h), and n is
the number of data points.
The ozone deposition rates inside Room 1 were
measured on a daily basis. A refrigerator air purifier
was operated for about 30 min to increase the
indoor ozone level inside the room to above
300 ppb. The decay of indoor ozone concentrations
was monitored, and the ozone deposition rate (/h)
was calculated as follows:
D¼
lnðCt1 Cb Þ lnðCt2 Cb Þ
AER
t2 t1
ð4Þ
where Ct1 and Ct2 are the concentrations at the beginning and the end of the measurement period,
Refrigerator air purifiers
(b)
2500
FVW1 (high)
FVW1 (low)
FVW2
4000
3000
2000
Ozone (ppb)
2000
5000
RAP1
RAP2
RAP3
1500
1000
500
1000
0
20
40
60
80
0
100
0
20
Distance (cm)
800
300
700
Ozone (ppb)
250
FS1
FS2
150
60
80
100
Other products
(d)
350
200
40
Distance (cm)
Facial steamers
(c)
Ozone (ppb)
VCout pAER
dCin T
VðAERþDÞ
VCout pAERþ
T
6000
Ozone (ppb)
ZT
0
7000
0
V
¼
T
T
0
0
where T is the duration of one cycle of use (h).
Equation 2 is rearranged to obtain the average emisRT
sion rate over one cycle of use, Edt=T:
Edt
100
50
600
500
OLS
SS1
400
300
200
100
0
0
0
20
40
60
Distance (cm)
80
100
0
20
40
60
80
100
Distance (cm)
Fig. 1 Average ozone concentrations at increasing distances from tested products that emit some ozone
5
Zhang & Jenkins
Results
Face ozone concentrations
The measured face ozone concentrations showed great
variability. Nine products were found to emit ozone
higher than the background levels, including three
refrigerator air purifiers, two fruit and vegetable washers, two facial steamers, one shoe sanitizer, and one
laundry water treatment system. These ozone emitters
either have ionizers or built-in ozone generators, or use
ultraviolet (UV) bulbs to intentionally produce ozone.
The remaining products emitted no or little ozone, generally <1 ppb with the background ozone levels subtracted,
even though they also have either ionizers or UV bulbs.
The face ozone concentrations of the nine ozoneemitting products are shown in Figure 1. For all of
these products, the ozone concentrations were highest
at 5 cm (2 inches), and then decreased rapidly at
increasing distances. At 5 cm (2 inches), the average
ozone concentrations all greatly exceeded the California ozone limit for indoor air cleaning devices
(50 ppb). At 81 cm (32 inches), the ozone concentrations decreased to the background levels for all of the
products, except for two products – the FVW1, a fruit
and vegetable washer, and the OLS – that still had
ozone concentrations higher than 50 ppb.
Contributions to room ozone concentrations
The room ozone concentrations during one cycle of
use, with the background ozone levels subtracted, are
summarized in Table 2. During one cycle of use, the
ID
Max
Mean
s.d.
Duration (min)
Laundry water treatment device
Fruit and vegetable washer
OLS
FVW1 (high)
FVW1 (low)
FVW2
SS1
FS1
FS2
RAP1
RAP2
RAP3
246
88
76
3
8
5
3
ND
ND
ND
106
28
8
1
5
2
1
ND
ND
ND
36
17
11
1
1
1
1
ND
ND
ND
60
15
15
20
15
10
10
5
5
5
OLS, ozone laundry system; s.d., Standard deviation; ND, non-detectable.
a
OLS was tested in a room with AER = 0.40 0.01/h and V = 10.9 m3. Refrigerator air purifiers were tested in a refrigerator in a small room with AER = 0.45 0.04/h and V = 72.0 m3.
Other devices were tested in a small room with AER = 0.43 0.03/h and V = 36.0 m3.
6
200
150
100
50
0
10
20
30
40
50
60
Personal exposure
(b)
3500
Category
Refrigerator air purifier
Room concentration
250
0
a
Ozone concentration
(ppb)
Shoe sanitizer
Facial steamer
(a)
Ozone concentration (ppb)
Table 2 Contributions of one cycle of use to room ozone concentrations
peak concentrations in the room ranged from nondetectable to 246 ppb, while the average concentrations ranged from non-detectable to 106 ppb.
The greatest increase in room ozone concentration
over one cycle of use was from the OLS. Figure 2a
shows the time course of the room ozone concentration
during one wash cycle. During the water fill period,
water runs through the OLS and activates the built-in
ozone generator. Ozone is injected into the water,
which then flows into the washing machine. Room
ozone concentrations were fairly stable or even
decreased during three water fill periods while the OLS
was producing ozone. However, marked increases of
room ozone concentrations were observed during the
second and third drain periods, and the incremental
increases were about 140 and 100 ppb, respectively,
within 5 min. Smaller incremental increases, ranging
from 20 to 60 ppb, were observed during wash and
rinse periods. These results suggest that ozone dissolved in water was released into the air when the water
was agitated and especially when drained.
The second greatest increase in room ozone concentrations was observed during the operation of one fruit
and vegetable washer, the FVW1. The FVW1 has two
Ozone concentration (ppb)
respectively (ppb), Cb is the background ozone concentration in Room 1 (ppb), and t1 and t2 are the times at
the beginning and the end of the measurement period,
respectively.
4%
3000
14%
2500
4%
35%
2000
43%
1500
1000
500
0
0
10
20
30
40
50
Elapsed time after the washing machine starts (min)
Fill
Wash
Drain
Rinse
60
Spin
Fig. 2 (a) Room ozone concentrations and (b) personal ozone
exposure concentrations during one wash cycle using the ozone
laundry system. Colors indicate different operation status of the
washing machine during one wash cycle. Pie chart shows the
percentage of total ozone exposures of each operation status
Ozone from consumer products and home appliances
power outputs, and the high-power output resulted in
slightly higher room ozone concentrations than the
low-power output. Both power outputs resulted in peak
room ozone concentrations higher than 50 ppb, but the
average room concentrations over 15 min of operation
were only 28 and 8 ppb for high- and low-power outputs, respectively.
Although measurable levels of ozone were detected
for the remaining products at their faces, their contributions to room ozone concentrations were minimal,
all lower than 5 ppb. It should be noted that as refrigerator air purifiers are not typically used in a room,
their contributions to room ozone concentrations were
measured as the changes of room ozone concentrations
during 5 min when the refrigerator door was open
while these products were operating inside the refrigerator. No changes of room ozone concentrations were
observed for any of the three refrigerator air purifiers.
Table 2 indicates that one cycle of use of the FVW2
did not increase room ozone concentrations markedly.
However, multiple cycles of use of this product with
reused water can substantially increase room ozone
concentrations, as shown in Figure 3a. For the first
(a) 50
Room ozone concentration (ppb)
1
40
Cycle 3
Cycle 2
Cycle 1
2
1
22
1
9
30
20
10
0
0
10
20
30
40
50
60
Personal ozone exposure concentrations
Personal ozone exposure concentrations during one
cycle of use are summarized in Table 3. The peak personal ozone exposure concentrations over one cycle of
use ranged from 60 to 3330 ppb, while the average personal ozone exposure concentrations ranged from 12
to 424 ppb.
The highest personal ozone exposure concentrations
were from the use of the OLS. Figure 2b shows the
time course of personal ozone exposure concentration
during one wash cycle using the OLS. The pie chart
indicates the percentage of total ozone exposure for
each operation. High exposures occurred during water
fill, rinse and drain periods, which together accounted
for over 90% of total ozone exposure during one wash
cycle. The second highest average personal ozone exposure concentrations during one cycle of use resulted
from the use of the FVW1 at high-power output. The
average personal exposure concentrations for the
remaining products were relatively low, but the peak
exposure concentrations were very high for some products. For example, the peak exposure concentrations to
ozone were 1050 ppb for one facial steamer, the FS2,
70
Table 3 Personal ozone exposure concentrations during one cycle of usea
(b) 30 000
Cycle 2
Cycle 1
Ozone exposure concentration (ppb)
two wash cycles, room ozone concentrations were
1 1 and 2 1 ppb, respectively. During the third
wash cycle, the room ozone concentration increased up
to 43 ppb, with an average of 22 9 ppb. A possible
reason for this increase is that, because the water was
reused, chemicals in the water that can react with
ozone had been consumed completely during the first
two wash cycles, thus allowing the water to become
saturated with ozone at the end of the second cycle.
This would result in the release of ozone into the air
during the third cycle.
25 000
12
1890
16
2070
Cycle 3
Ozone concentration
(ppb)
5740 ± 4750
20 000
15 000
10 000
Category
ID
Max
Meanb
s.d.
Duration
(min)
Measurement
distance (cm)
Laundry water
treatment device
Fruit and
vegetable washer
OLS
3330
424
640
60
30
FVW1 (high)
FVW1 (low)
FVW2
SS1
FS1
FS2
RAP1
RAP2
RAP3
1750
449
74
80
396
1050
440
85
60
356
42
12
26
46
77
47
23
14
330
78
16
25
89
198
97
19
16
15
15
20
15
10
10
5
5
5
30
30
30
30
5
5
5
5
5
Shoe sanitizer
Facial steamer
5000
0
0
10
20
30
40
Elapsed time (min)
50
60
70
Fig. 3 (a) Room ozone concentrations, and (b) personal ozone
exposure concentrations during three wash cycles using the fruit
and vegetable washer 2. Water was reused for the whole duration of three wash cycles. Means (in ppb) and standard deviations are denoted for each wash cycle
Refrigerator air
purifier
OLS, ozone laundry system; s.d., Standard deviation.
Personal ozone exposure concentrations were measured at realistic distances that users
would be from each device during operation. Background levels have been subtracted.
b
Average personal ozone exposure concentration over the duration of one cycle of use
shown in the column of ‘Duration’.
a
7
Zhang & Jenkins
and 396 ppb for the other facial steamer, the FS1. The
time courses of personal exposure concentrations during the use of these two products are shown in Figure S2 in Supporting Information.
Personal ozone exposure concentrations from refrigerator air purifiers were measured at the moment when
the refrigerator door was opened. These products do
not emit ozone continuously. Instead, they emit a large
amount of ozone during the first half-hour, and then
release a blast of ozone for a few minutes every 2 h (see
Supporting Information Figure S3). Thus, personal
ozone exposure concentrations related to the use of
these products depends on when the refrigerator door
is opened. When the door was opened 30 min after
these products had been operating inside the refrigerator, as shown in Supporting Information (Figure S4),
the user was exposed to as much as 50–450 ppb ozone,
although it was diluted very quickly. For example,
after 2 min, exposure concentrations decreased 80–
99%, and the average ozone exposure concentrations
over 5 min when the refrigerator door was opened
were 14–47 ppb. When the door was opened after these
products were operated overnight inside the refrigerator, the user’s personal exposure concentrations were
not different from the background levels.
For some products, multiple cycles of use may result
in high personal exposure concentrations. The room tests
revealed that substantial amounts of ozone can be
released after using one fruit and vegetable washer, the
FVW2, for two wash cycles with reused water, so personal ozone exposure concentrations for this product
were also measured for multiple wash cycles. Figure 3b
shows the time course of personal ozone exposure concentrations over three wash cycles with reused water. For
the first wash cycle, the average personal exposure concentration was only 12 ppb, but it increased to 1890 and
5740 ppb for the second and third wash cycles, respectively. Over the three continuous wash cycles, the average
personal exposure concentration was 2550 ppb.
To assess the contributions of these products to total
daily exposure to ozone, four hypothetical exposure
scenarios were assessed, considering the following two
factors: (i) a person living in a polluted area versus one
living in a clean area and (ii) a person who stays in very
close proximity to an ozone-emitting product versus
one who stays in the same room but about 3 ft away
from the product. The assumptions and scenarios are
described in the Supporting Information, and complete
results are presented in Table S3. For a person living
in a polluted area, the exposure during one cycle of use
of the OLS would account for 23–38% of their total
daily exposure to ozone. For a person living in a clean
area, the percentage is estimated to be as high as 35–
52%. The contributions from the FVW1 at high-power
output are also substantial. At high-power output, it
contributes to 20% and 31% of total daily exposure to
ozone for those living in a polluted area and in a clean
8
area, respectively. The second fruit and vegetable
washer, FVW2, contributed minimally to total exposure when used for just one cycle of use, but when used
for three wash cycles with reused water, it is estimated
that up to 78% (in the polluted area) and 87% (in the
clean area) of total daily exposure to ozone would
result from the use of this product if one spends
30 min in very close proximity to the product while it
is operating. Contributions from the other products,
given their lower ozone emissions and shorter durations of use, generally appear to be inconsequential.
Ozone emission rates
Ozone deposition rates inside Room 1 ranged from 0.7
to 1.4/h, with an average of 1.0 0.2/h. Ozone emission rates were calculated for most of the products with
significant ozone emissions (Table 4). The emission rate
of the OLS was not determined, because for proper
operation this product needs to be connected to a washing machine, which did not properly fit in Room 1. The
FS1 and FS2 were tested for ozone emission rates; however, they did not increase the ozone concentrations
inside Room 1 to a high enough level to accurately estimate their emission rates. Table 4 shows that average
ozone emission rates ranged from 1.6 to 15.4 mg/h. The
FVW1 had the greatest average ozone emission rate for
both power outputs. The average ozone emission rate
of the SS1 was lower than those of the FVW1. The
average ozone emission rate of the FVW2 was less than
one-half the emission rate of the FVW1. Because the
emissions from refrigerator air purifiers were intermittent, as shown in Figure S3, their emission rates were
reported for the first 30–50 min, when the majority of
ozone was emitted. The RAP1 had the highest average
ozone emission rate among the tested refrigerator air
purifiers; the other two were notably lower, with rates
just 16% and 40% of the RAP1 rate.
Table 4 Calculated ozone emission rates from this study
Emission ratea (mg/h)
Category
ID
Test 1
Laundry water treatment device
Fruit and vegetable washer
OLS
FVW1 (high)
FVW1 (low)
FVW2
SS1
FS1
FS2
RAP1
RAP2
RAP3
N/A
14.2
15.0
6.1
8.0
N/A
N/A
10.5
3.1
1.8
Shoe sanitizer
Facial steamer
Refrigerator air purifier
a
Test 2
Test 3
Mean
18.1
11.4
8.9
15.5
13.9
16.7
3.6
13.5
15.4
14.4
6.2
12.3
8.4
5.5
1.6
12.1
3.8
1.3
10.3
4.1
1.6
The emission rate of the ozone laundry system (OLS) was not determined because for
proper operation this product has to be connected to a washing machine which did not fit
in Room 1. The FS1 and FS2 did not increase the ozone concentrations inside Room 1 sufficiently to allow emission rates to be estimated accurately.
Ozone from consumer products and home appliances
Variability of ozone emissions
To study the intraunit variability of ozone emissions,
room concentration and personal exposure tests under
identical operating conditions were repeated three
times for the OLS, for the FVW1 at high-power output
and for the SS1. For the contributions to room ozone
concentrations, the differences between repeated measurements ranged from 15% to 47%. The contributions to personal ozone exposure concentrations
showed much greater variability. The highest personal
exposure concentration was about 4 times higher than
the lowest personal exposure concentration measured
for the OLS, 7 times higher for the FVW1 at highpower output, and 2 times higher for the SS1. Ozone
emission rates, as shown in Table 4, were measured
three times for most of the products, and the results
indicate the variability ranged from 2% to 44%.
To determine the interunit variability of ozone emissions, duplicate product units of the RAP1, the FVW1
and the SS1 were tested. The results shown in Table S1
in the Supporting Information suggest substantial variability. For the contributions to room ozone concentrations, the mean high/low ratios were <2 for two
products. The highest mean high/low ratio, 5.0, was
measured for the FVW1 at low power. For the contributions to personal exposure concentrations, the mean
high/low ratios ranged from 1.5 to 18.4. The highest
mean high/low ratio was also observed for the FVW1
at low power. Ozone emission rates for duplicate product units were not measured because Room 1 (the air
monitoring shelter) used to measure emission rates was
not available when duplicate product units were
received.
Discussion
Comparison to other ozone-generating products
The results indicate that ozone emissions are fairly
common for some products tested in this study, and
their emissions are not negligible. Of 17 consumer
products and home appliances used in this study, nine
were found to emit measureable levels of ozone. At
5 cm (2 inches) from the air stream outlets, these nine
products produced ozone exceeding CARB’s ozone
limit for indoor air cleaning devices of 50 ppb by a factor of 3–125.
Ozone emissions from the tested products were
comparable to or higher than other ozone-generating products reported in the literature, as summarized in Table 5. For example, the ozone levels at
5 cm (2 inches) were all higher than those of electrostatic precipitators and ionizers, and the 5-cm
ozone levels of three products were even higher
than the highest levels emitted by ozone generators
(air cleaners that intentionally emit ozone). Because
Table 5 Comparison of ozone emissions from products tested in this study and from other
ozone-generating products
Product
Consumer
products
and home
appliances
Air cleaners
intentionally emit
ozone
Electrostatic
precipitators and
ionizers
In-duct air cleaning
devices
Ionic and
ozonolysis
air purifiers
Portable air
cleaners
Photocopiers
Printers
Face conc.
(ppb)
Room conc.
(ppb)
147–6230
ND–106
329–1287
2–448
0.079–94
CARB (2006)
2.1–15.1
0.50–2.9
CARB (2008)
ND–194
<MQL–349
9–650
0.16–220
Britigan et al. (2006)
0.07–6
Destaillats et al. (2014)
1.3–7.9
0.58–1.75
Leovic et al. (1996)
Maddalena et al. (2011)
3.3–44
Emission rate
(mg/h)
1.6–15.4
Reference
This study
Morrison et al. (2014)
MQL, Method quantification limit; ND, Non-detectable.
one cycle of use only lasts for <20 min for most
products tested in this study, their contributions to
room ozone levels were minimal, generally <30 ppb,
except for the OLS. Over a 1-h wash cycle, the
OLS increased room ozone concentrations by
106 ppb, above the level of the 1-h California
AAQS of 0.09 ppm. Such incremental increase was
comparable to some air cleaners that intentionally
emit ozone and some in-duct air cleaners.
Quantification of ozone emission rates was feasible
for six of the products tested. As shown in Table 5,
their emission rates were within the range of ozone
emission rates determined for other indoor ozone-generating products, and generally comparable to those of
portable air cleaners and photocopiers.
However, it should be noted that most of the products tested in this study are used intermittently; thus,
their impacts on indoor air quality are limited to
shorter time periods compared to the continuously
operating products such as portable air cleaners and
in-duct air cleaners.
Exposures, health impacts, and other impacts
Average ozone personal exposure concentrations from
these products ranged from 12 to 424 ppb over one
cycle of use. This is much higher than the corresponding increases in room ozone concentrations, due to
users’ close proximity to these products. It should be
noted that personal exposure concentrations were measured at very close proximity to the products over one
cycle of use. However, users may not stay at such close
proximity over the full period of one cycle of use. Thus,
9
Zhang & Jenkins
the expected actual exposure concentrations would fall
somewhere between the room concentrations shown in
Table 2 and the personal exposure concentrations
shown in Table 3. The personal exposure concentration observed from the use of the OLS was almost 5
times higher than the level of the 1-h California AAQS
of 0.09 ppm. In addition, the average room ozone concentration during one wash cycle when using the OLS
was also higher than the level of the 1-h California
AAQS. Therefore, users of this product and anyone in
the same laundry room can be exposed to levels above
the health-based standard. Further tests of the OLS in
a chamber with more controlled environment is warranted.
Although some products produce relatively low personal ozone exposure concentrations for one cycle of
use, multiple cycles of use may result in personal exposure concentrations substantially higher than the
health-based standard. The FVW2 was found to
increase personal ozone exposure concentration by
only 12 ppb for the first wash cycle. However, over a
1-h period of three wash cycles in a row (with a 3- to 5min break between two wash cycles) using reused
water, the average personal ozone exposure concentration was 2550 ppb, over 28 times higher than the level
of the 1-h California AAQS of 0.09 ppm. Although it
is unlikely that users will keep their heads at 1 ft above
this product for 1 h, they may be exposed to ozone as
high as 20 000 ppb for short durations as shown in
Figure 3b. A breath or two of such high levels of ozone
might not cause immediate health effects, but repeated
exposures to ozone of this high level may be a health
concern. The maximum capacity of this product is only
6 l. Users from some households may need to use this
product multiple times in succession, especially if they
want to wash produce types separately. Hence, prolonged use of the FVW2 and associated high exposure
concentrations may occur more frequently than
expected.
The results presented in this study suggest that there
may be a health concern for users of OLSs and certain
fruit and vegetable washers. Their exposure levels to
ozone can be similar to or greater than the concentrations at which health impacts have been observed for
people at rest or performing light-to-moderate activities (Bates et al., 1972; Folinsbee et al., 1978; Horvath
et al., 1979). Those health effects included reduced pulmonary function, cough, and other symptoms. In addition, although the average personal exposure
concentrations were low for some products, surprisingly high spikes were commonly observed. For example, one can be exposed to ozone as high as 440 ppb
when the refrigerator door is opened while the RAP1 is
emitting ozone. Users of the FS1 and FS2 can be
exposed to ozone levels as high as 396 and 1050 ppb,
respectively. Although the health effects of exposure to
high levels of ozone over very short time periods have
10
not been recognized in the literature, such exposures
may raise a concern for sensitive populations, for
example, those with respiratory diseases, who use these
products repeatedly.
As discussed above, despite the short durations of
use of these products, some of them can contribute a
significant fraction (up to 87%) of total daily exposure
to ozone. And there are other concerns about these
ozone-generating products beyond their health risks
from exposure to ozone. Ozone can react with chemicals such as limonene, pinene, and styrene to produce
secondary air pollutants, including formaldehyde,
ultrafine particles, and highly reactive free radicals,
which may add to the health burden on people. In
addition, the great variability of these products’ contributions to room ozone concentrations and personal
exposure concentrations, especially from duplicate product units, indicates potential quality control issues
with their manufacture. Several products appeared to
be made from low-quality materials and parts, and/or
were loosely assembled. Table 1 shows that only six of
the products tested have electrical safety marks, indicating that the others have not been tested and certified
under existing electrical safety programs; thus, they
may pose potential safety issues. Actually, one personal air purifier caused an electric shock when
touched, which shut down the ozone monitor during
testing. And finally, high levels of ozone from these
products may also deteriorate materials indoors. Due
to its high chemical reactivity, ozone can cause damage
to rubber and elastomers (Rowe et al., 1986). It can
cause certain sensitive dyes and artists’ pigments to
fade (Beloin, 1973; Shaver et al., 1983). Thus, use of
these ozone-generating products indoors, such as the
refrigerator air purifiers, may increase the costs of
maintenance and replacement for materials and appliances with materials sensitive to ozone.
Except for UL Standard 867 for Electrostatic Air
Cleaners and UL Standard 2823 for Electronic Equipment, there are no industry test standards for ozone
emissions from the types of consumer products tested
in this study. In addition, there are no consumer product regulations that limit ozone emissions from these
products. Product designs with lower emissions, industry test standards to certify low-emitting products,
and/or regulations limiting ozone emissions from these
products appear to be needed to fill these gaps in consumer protection.
Additionally, research on additional models of the
types of products tested as well as other ozone-emitting
products is warranted. Only a small group of products
that emit ozone either intentionally or as a by-product
of their functions were tested in this study. The prevalence of ozone emissions from these products and the
high levels of ozone produced by some products indicate a need to investigate other products that may generate ozone, such as spas and hot tubs that use ozone
Ozone from consumer products and home appliances
generators for water purification, toilets equipped with
ozone sanitizers, bathroom sanitizers, and ozone appliances used to deodorize sports equipment.
Limitations
Due to limited resources, only a selected set of products could be tested in this pilot level study. There are
more products that may emit ozone either intentionally
or as a by-product of their functions, but it was not
feasible to test all of them in this study. Because Room
1 (the air monitoring shelter) used to determine emission rates was not available when duplicate product
units were received, the interunit variability of emission
rates was not determined. In addition, all of the products were tested when they were new; how ozone emissions from these products change after repeated usage
is unknown.
Conclusions
Nine of 17 products tested in this study emitted ozone
while operating, and some of them produced concentrations several times greater than the level of the 1-h
California AAQS of 0.09 ppm. Additionally, despite
short durations of use, three of the tested products can
contribute a significant fraction of total daily exposure
to ozone at levels that can impact health. Thus, the use
of some products tested in this study may present a
health risk to the public. Further research and actions,
such as product design changes, testing to industry
standards, and/or regulations limiting ozone emissions
from these products, appear to be needed to reduce
ozone exposures from the use of these products.
Acknowledgements
We thank James Pham and Mac McDougall of the
Monitoring and Laboratory Division (MLD) of the
California Air Resources Board for calibrating the
instruments, and Patrick Vaca and Joseph Guerrero
of the MLD for providing a facility and assistance
for this study. We also thank Deborah Drechsler of
the Research Division of the California Air
Resources Board for reviewing the health effect
assessments.
Supporting Information
Additional Supporting Information may be found
online in the supporting information tab for this article:
Figure S1. Instrument collocation data for the 2B
ozone monitors used in this study.
Figure S2. Personal exposure concentrations to ozone
during the operation of facial steamers.
Figure S3. Ozone emission profiles of RAP1 at a distance of 2 inches.
Figure S4. Personal exposure concentrations to ozone
when the refrigerator door was opened after 30-min
operation of refrigerator air purifiers.
Table S1. Inter-unit variability of room ozone concentrations and personal exposure concentrations from
duplicate product units.
Table S2. Data inputs for total exposure assessment.
Table S3. Percentage of total daily exposure to ozone
from the use of several tested products for four hypothetical exposure scenarios.
Data S1. Total Exposure Assessment.
References
Baptista, F.J., Bailey, B.J., Randall, J.M.
and Meneses, J.F. (1999) Greenhouse
ventilation rate: theory and measurement
with tracer gas techniques, J. Agric. Eng.
Res., 72, 363–374.
Bates, D.V., Bell, G.M., Burnham, C.D.,
Hazucha, M., Mantha, J., Pengelly, L.D.
and Silverman, F. (1972) Short-term
effects of ozone on the lung, J. Appl.
Physiol., 32, 176–181.
Beloin, N.J. (1973) Fading of dyed fabrics
exposed to air pollutants, Textile Chem.
Color., 5, 29–34.
Boeniger, M.F. (1995) Use of ozone
generating devices to improve indoor
air quality, Am. Ind. Hyg. Assoc. J., 56,
590–598.
Britigan, N., Alshawa, A. and Nizkorodov,
S.A. (2006) Quantification of ozone levels
in indoor environments generated by ionization and ozonolysis air purifiers, J. Air
Waste Manag. Assoc., 56, 601–610.
CARB (2005) Review of the california ambient air quality standard for ozone, Staff
report to the California Air Resources
Board. Available at: http://www.arb.ca.
gov/research/aaqs/ozone-rs/rev-staff/
rev-staff.htm#Summary
CARB (2006) Evaluation of ozone emissions from portable indoor ‘air cleaners’ that intentionally generate ozone,
Staff report to the California Air
Resources Board. Available at: http://
www.arb.ca.gov/re
search/indoor/o3 g-rpt.pdf
CARB (2008) Evaluation of ozone emissions
from portable indoor air cleaners: electrostatic precipitators and ionizers, Staff
report to the California Air Resources
Board. Available at: http://www.arb.ca.
gov/research/indoor/esp_report.pdf
CARB (2016) California certified air cleaning devices, Last updated on March 3,
2016. Available at: http://www.arb.ca.
gov/research/indoor/aircleaners/certified.htm
Cole, E.C. (2003) Gas-phase ozone: assessment of biocidal properties for the indoor
environment – a critical review, Appl.
Biosaf., 8, 112–117.
Coleman, B.K., Lunden, M.M., Destaillats,
H. and Nazaroff, W.W. (2008) Secondary
organic aerosol from ozone-initiated reactions with terpene-rich household products, Atmos. Environ., 42, 8234–8245.
Destaillats, H., Sleiman, M. and Fisk, W.J.
(2014) Evaluation of pollutant emissions
from portable air cleaners, Final Report
to California Air Resources Board:
Contract No. 10-320. Available at: http://
www.arb.ca.gov/research/apr/past/
10-320.pdf
Foarde, K.K., Vanosdell, D.W. and Steiber,
R.S. (1997) Investigation of gas-phase
ozone as a potential biocide, Appl. Occup.
Environ. Hyg., 12, 535–542.
11
Zhang & Jenkins
Folinsbee, L.J., Drinkwater, B.L., Bedi, J.F.
and Horvath, S.M. (1978) The Influence
of Exercise on the Pulmonary Function
Changes Due to Exposure to Low Concentrations of Ozone, New York, NY, Academic Press, 125–145.
Horvath, S.M., Gliner, J.A. and MatsenTwisdale, J.A. (1979) Pulmonary function and maximum exercise responses following acute ozone exposure, Aviat.
Space Environ. Med., 50, 901–905.
Ibald-Mulli, A., Wichmann, H.E., Kreyling, W. and Peters, A. (2002) Epidemiological evidence on health effects
of ultrafine particles, J. Aerosol. Med.,
15, 189–201.
Leovic, K.W., Sheldon, L.S., Whitaker,
D.A., Hetes, R.G., Calcagni, J.A.
and Baskir, J.N. (1996) Measurement of
indoor air emissions from dry-process
photocopy machines, J. Air Waste
Manag. Assoc., 46, 821–829.
Maddalena, R., McKone, T., Destaillats, H.,
Rusell, M., Hodgson, A. and Perino, C.
(2011) Quantifying pollutant emissions
from office equipment: a concern in
energy-efficient buildings, California
Energy Commission, PIER EnergyRelated Environmental Research, CEC500-2011-046
Mason, M.A., Sparks, L.E., Moore, S.A.,
Dolgov, I. and Perry, R.B. (2000) Characterization of ozone emissions from air
cleaners equipped with ozone generators
and sensor and feedback control circuitry, In: Engineering Solutions to Indoor
Air Quality Programs Symposium, VIP98, A&WMA, 254–269.
12
McConnell, R., Berhane, K., Gilliland, F.,
London, S.J., Islam, T., Gauderman,
W.J., Avol, E., Margolis, H.G. and
Peters, J.M. (2002) Asthma in exercising
children exposed to ozone: a cohort
study, Lancet, 359, 386–391.
Morrison, G., Shaughnessy, R. and Siegel, J.
(2014) In-duct air cleaning devices: ozone
emission rates and test methodology,
Final Report to California Air Resources
Board: Contract No. 09-342. Available at:
http://www.arb.ca.gov/research/apr/past/
09-342.pdf
Nazaroff, W.W. and Weschler, C.J. (2004)
Cleaning products and air fresheners:
exposure to primary and secondary air
pollutants, Atmos. Environ., 38, 2841–
2865.
NTP (2011) Report on Carcinogens, 13th
edn, Research Triangle Park, NC,
National Toxicology Program, U.S.
Department of Health and Human Services. Available at: http://ntp.niehs.nih.
gov/ntp/roc/content/profiles/formalde
hyde.pdf
Oberd€
orster, G. (2000) Pulmonary effects of
inhaled ultrafine particles, Int. Arch.
Occup. Environ. Health, 74, 1–8.
Oberd€
orster, G., Oberd€
orster, E. and
orster, J. (2005) Nanotoxicology:
Oberd€
an emerging discipline evolving from
studies of ultrafine particles, Environ.
Health Perspect., 113, 823–839.
Peters, A., Wichmann, H.E., Tuch, T., Heinrich, J. and Heyder, J. (1997) Respiratory
effects are associated with the number of
ultrafine particles, Am. J. Respir. Crit.
Care Med., 155, 1376–1383.
Poppendieck, D.G., Rim, D. and Persily,
A.K. (2014) Ultrafine particle removal
and ozone generation by in-duct electrostatic precipitators, Environ. Sci. Technol., 48, 2067–2074.
Rowe, R.D., Lauraine, C.G., Peterson,
D.C., Miller, C., Adamas, R.M., Oliver,
W.R. and Hogo, H. (1986) The benefits
of air pollution control in California,
Final report to the California Air
Resources Board. Available at: http://
www.arb.ca.gov/research/apr/past/a2118-32_exsum.pdf
Shaver, C.L., Cass, G.R. and Druzik, J.R.
(1983) Ozone and the deterioration of
works of art, Environ. Sci. Technol., 17,
748–752.
Singer, B.C., Coleman, B.K., Destaillats, H.,
Hodgson, A.T., Lunden, M.M., Weschler, C.J. and Nazaroff, W.W. (2006)
Indoor secondary pollutants from cleaning product and air freshener use in the
presence of ozone, Atmos. Environ., 40,
6696–6710.
Tung, T.C.W., Niu, J.L., Burnett, J. and Hung,
K. (2005) Determination of ozone emission
from a domestic air cleaner and decay
parameters using environmental chamber
tests, Indoor Built Environ., 14, 29–37.
U.S. EPA (2013) Integrated science assessment of ozone and related photochemical
oxidants, Final Report to U.S. Environmental Protection Agency. Available at:
http://cfpub.epa.gov/ncea/isa/recordisplay.cfm?deid=247492
Weschler, C.J. (2000) Ozone in indoor environments: concentration and chemistry,
Indoor Air, 10, 269–288.