The seventh International Conference on Urban Climate,
29 June - 3 July 2009, Yokohama, Japan
THE NO2 BEHAVIOR ANALYSIS IN A ROADSIDE ATMOSPHERE FOR THE
VALIDATION OF THE RSAQSM
1,2
1
1,3
Hiroaki Minoura and Akiyoshi Ito
Petroleum Energy Center, Toranomon, Minato-ku, Tokyo, Japan
2
Toyota Central R&D Labs., Inc., Nagakute, Aichi, Japan
3
Japan Automobile Research Institute, Ibaraki, Japan
Abstract
By the reinforcement of the vehicle emission reduction requirement in Japan, the nitrogen oxides (NOx)
concentration in a roadside atmosphere decreased clearly recently. However, the ambient nitrogen dioxide
(NO2) concentration does not show a substantial decrease.
For the forecast of the improvement of the roadside NO2 by the motor vehicle emission regulation, we
have been developing a roadside air quality simulation model (RSAQSM). The chemical reactions of nitrogen
oxide (NO) oxidation with ozone (O3), photodissociation of NO2, and O3 formation are included in this model.
This study was carried out to clarify the NO2 behavior near roadside and to offer data contributing to the
simulation. Involving a spatial distribution of O3 concentration measurement, temporal and spatial variation of
NO and NO2 was measured by the devices of chemiluminescence method which were prepared for each
species. Concentration-change was measured at the same time in four sites every one second to pay
attention to a percentage of NO2/NOx which varied from route neighborhood to lee side way. The air flow of
each sites was monitored by the ultrasonic 3D anemometer. The effect of NO2 concentration change due to
the air stagnation or background ozone concentration was evaluated.
From the cycle variation of potential ozone (PO) concentration as a function of the elapsed time of the
traffic signal, the direct exhaust NO2 was estimated, and, around Noge intersection of Ring 8, Tokyo, 7.3%
was obtained as primary NO2 emission from vehicles. During an advection from the curbside (0m) to the 20m
remote site, more than 40% of NO oxidized and generate NO2 at 10m height. At 3m height, 7% of NO2 was
generated and O3 was consumed in same quantity at the same time.
Key words: nitric oxide, nitrogen dioxide, ozone, roadside air quality, vehicle emission
INTRODUCTION
The nitrogen oxide (NOx) emission from vehicles amounts about 50% of whole emissions in Tokyo, and it
is considered as primary source (MOE, 2002). Not only vehicle emission regulation every several years but
also riding into regulation to Tokyo was reinforced for reduction of the NOx discharge. Decrease tendency of
ambient NOx concentration is confirmed, however nitrogen dioxide (NO2) concentration is not seen to be
changed. Clapp and Jenkin (2001) showed the local NO2 contributors on the basis of the monitoring data of 14
sites in the UK. One of the contributors is the direct NO2 (f-NO2) emission from vehicles. The aftertreatment of
tailpipe emission (eg. oxidation catalysts) of the diesel vehicle is thought to be one source. Grice et al (2009)
showed that f-NO2 has increased in recent years in Europe. Air quality expert group in the UK published the
fourth report in 2007. The report says that the fraction of NOx emitted as f-NO2 is in excess of 5%, with values
in the range 20-70% for Euro III diesel cars. Possibility to deteriorate for 2005-2010 years was shown in f-NO2
after having shown a slight rise for 2002-2005 years. The second big factor depends on the increase of the
ozone (O3) concentration. The increase in O3 concentration in large area was sometimes reported (Fjæraa
and Hjellbrekke, 2007), and the background O3 concentration exceeds 50ppb during summer season in Japan
(Minoura, 1999). By the increase in the O3 concentration, percentage of the oxidation of the NO increases and,
as a result, raises NO2 concentration. On the other hand, the NO2 is regarded as the cause of asthma and the
lungs functional decline of children, and more reduction of NO2 concentration is demanded (Gauderman et al,
2005, Delfino et al, 2008).
For the purpose of provision of data contributing to environmental policy making through development of
high accuracy Air Quality Model and impact prediction of improved air quality, the Japan Clean Air Program
had started in 1997 by the fund of Ministry of Economy, Trade and Industry of Japan and the cooperation of
the auto industries and the petroleum industries. A tertiary stage of the quinquennium program began, and the
name was changed in JATOP (Japan AuTo-Oil Program) in 2007. A regional scale air quality simulation,
based on CMAQ, and the roadside air quality simulation are developed in this program.
For the forecast of the atmospheric environment improvement of the NO2 by the motor vehicle emission
regulation, we are developing a roadside air quality simulation model (RSAQSM). The chemical reactions of
nitrogen oxide (NO) oxidation with O3, photodissociation of NO2, and O3 formation are going to introduce in
this model.
The behavior of ambient NO2 near a trunk road is very complicated; the air flow involving a wake flow by
* Corresponding author address: Hiroaki Minoura, Toyota Central R&D Labs., Inc., Nagakute, Aichi, 4801192, Japan, e-mail: [email protected]
The seventh International Conference on Urban Climate,
29 June - 3 July 2009, Yokohama, Japan
vehicle running perturb the NO2 concentration, and the direct emission from vehicles and NO oxidation with
ambient O3 gives a quick variation, as mention above. Real-time measurement using a laser absorption
technique is one of the ideal method. Shorter et al (2005) used the tunable infrared laser differential
absorption spectrometer and showed the NO2 emission from in-use buses. However, the observation of the
spatial distribution is also important to understand the behavior of the NO2, and the simultaneous multi-point
observation is unable with such an expensive device.
This study was carried out to clarify the NO2 behavior near roadside and offer data contributing to the
simulation. Involving a spatial distribution of O3 concentration, wind, atmospheric temperature, humidity, and
solar radiation, temporal and spatial variation of NO and NO2 was measured by the devices of
chemiluminescence method of the high-speed response which were prepared by each chemical species.
OBSERVATION
Sampling lines and the equipment component is shown in
Figure 1. Two NOx monitor of the same model (Thermoelectron
company, Model 42) was used to obtain real-time measurement of
the NO2. As shown in Figure 1, a sampling line was changed by the
solenoid valves to measure concentrations of NO, NOx and O3 by 3
altimetry (1, 3, and 10m from the ground) in one spot. We changed
the solenoid valves every ten minutes and measured the
concentrations of 3 levels repeatedly. To avoid the chemical
reaction with NO and O3 in the sampling line, two surge pumps
were used, and the sampled air was introduced into the measuring
monitors within one second. Air flow (u, v, and w) of each levels
were monitored by the ultrasonic 3D anemometer (Kaijo, SAT550).
Figure 1 Sampling line and
As shown in Figure 2, four sites (a south curbside (0m) “Site A”,
measurement system
20m south remote site “Site B”, 100m south remote site “Site C”,
and a north curbside “Site D”) were set up in the direction that was
perpendicular to the trunk road (Ring No.8; average traffic volume
-1
is approximately 2,400 vehicles hr , and the type ratio of heavy
duty vehicle is 11%). The observation Site A, B and C was located
in Noge Park, Setagaya, Tokyo, and there were no large stationary
sources of NOx around this area. The surroundings was residential
area of three stories buildings at the highest and it was the place
where the pollutant of the road flows into the park ideally without
disturbing air flow. Atmospheric temperature and relative humidity
was monitored (Vaisala, HMP45D) at C point of the 3m height
above the ground. Global solar radiation was monitored
(Ogasawara Keiki Seisakusho, model MS42) at the same point.
Eight sets of the NOx monitor involving the sampling line were
checked each difference with the standard gas. The data was
revised in consideration of correction factor provided by this
approval. The NO2 concentration was obtained from difference of
NOx concentration and the NO concentration. We confirmed
response of the NOx monitor by giving standard gas of the fixed
Figure 2 Location of the observation
quantity with a syringe in a purified air instantly. When we injected
points
NO gas, the concentration-change of the shape that resembled the
Weibull distribution of about 1.5sec in half width was seen. On the other hand, half width spread through about
2 times when we carried out similar experiment in NOx. The half width broadening by the length of the
sampling line was not seen. It is thought that the half width broadening seen in NOx was caused by the
adsorption to the catalyst surface when we deoxidize NOx into NO (private communication with a technician of
Thermoelectron company). We cannot get NO2 concentration from a simple difference in order that there was
slip for response time of NO and the NOx even if we measure the same pollutant gas. Because the highest
correlation was provided when we delayed observed data of the NO for about two seconds and compared it
with data of the NOx, and a provided NO2 level was reasonable, we analyzed results by this method.
The simultaneous observation was carried out from March 4 to March 7 in 2008. All data was recorded in
PC in succession every one second. By using the ultrasonic 3D anemometer, the effect of NO2 concentration
change due to the air stagnation was evaluated. Only data of the consecutive lee wind condition of more than
three minutes was analyzed. The lee wind condition was defined as the wind direction of 19 degrees and the
deflection angle of the wind direction within 45 degrees, which perpendicular to the road. Value of the north
curbside point was assumed as a background value at first step. However, because the values of the north
curbside point was seemed to be affected by the building nearby, and high concentration of NOx was seen
several times even when the measured wind direction showed north, we gave up to use the North curbside
data.
The seventh International Conference on Urban Climate,
29 June - 3 July 2009, Yokohama, Japan
RESULTS AND ANALSYS
A temporal variation in the NOx concentration was seen depending on a traffic volume of Ring 8. Because
the traffic volume was controlled by the signal of the intersection nearby, the NOx concentration variation was
seen corresponding with the signal cycle of 140 seconds. Figure 3 shows the mean values of 1m height
provided in the lee wind condition for the observation period as a function of an elapsed time of the signal. As
shown in Figure 3, temporal variation of the high periodic repeatability of 140 seconds was observed in NOx,
NO, and O3. With increase in NOx concentration, decrease tendency in O3 concentration was seen, and some
NO reacted with O3 was suggested. The periodic variation in curbside was the most remarkable and the space
attenuation was confirmed.
Figure 3 Time variation of gas concentrations observed 1m height during one signal cycle
NO2 concentration in a roadside atmosphere depends on two factors. The first is the direct emission (fNO2) from vehicles and the second is the oxidation from NO from vehicle and O3 from ambient. When there is
none of f-NO2, the potential ozone concentration (PO = NO2 + O3) keeps constant value. It became clear that
PO showed temporal variation from results of Figure 3. It was thought that the quantity of PO variation in one
period was equivalent to f-NO2, and a ratio with the quantity of NOx was estimated as 7.3% on the basis of the
curbside data. NO2 of 7.3% was thought to be included in motor vehicle emission, when we assume observed
NOx was caused by vehicle emission. Yao et al (2005) measured f-NO2 in a long tunnel under low O3
condition in Hong Kong, and obtained 2%. Grice et al (2009) showed that f-NO2 of petrol-fuelled vehicles was
less than 5%, and that of the diesel vehicles was 10-12%. And the diesel vehicle with the latest after treatment
systems was thought to be higher value of f-NO2. Abbott (2005) showed a similar tendency.
Table 1 shows the mean value of NO2/NOx obtained whole
Table 1 The mean value of
observation period under the lee wind condition, and it shows space
NO2/NOx
dependency. 0m in Table 1 shows Site A. The reason why NO2/NOx
values increase as the distance from the road is NO2 oxidation
generation. A change in NO2/NOx was seen by the height in curbside
data. In Site C, NO2/NOx became uniform by height, and it was thought
that the influence by the oxidation of the NO from Ring 8 terminated at
this point. The NO2/NOx changed by ambient O3 concentration, and it
was found to be increased according to the ambient O3 concentration.
To estimate oxidation formation of the NO2 between Site A and Site B, the value in Site C was assumed as
the background concentration and the diffusion coefficient between the two sites was estimated from a ratio of
the NOx concentration, as follows;
# NOx(SiteB)"NOx(SiteC) &
theDiffusionCoefficient = %
(
$ NOx(SiteA)"NOx(SiteC) '
(1)
With this diffusion coefficient, gas concentration in Site B was estimated by using the concentration
measured in Site A only with diffusion change without chemical change. The difference with the estimated
!
concentration
and the observed concentration provides a portion of the chemical reaction. Table 2 shows the
concentration differences and the rate of changes by the chemical reaction during an advection from Site A to
Site B. The NO2 generation and the NO loss (or the O3 loss) should become the equivalent value theoretically.
At 10m height, the NO loss and the NO2 generation was almost equal, and more than 40% of NO oxidized and
generate NO2. As for O3, the difference of the concentration was not observed between the sites. This result
suggests that the equivalent portion of the consumed O3 for the NO oxidation was supplied from the sky.
However, the NO2 generation was lower than the NO loss at 1m height and 3m height as shown in Table 2. At
3m height, 7% of NO2 was generated and the same value of O3 was consumed. Unknown NO consumption
was seen approximately 6% at 3m height without the diffusion process. At 1m height, none of the NO2 was
generated, and NO of 16% disappeared. The tendency that NO2 generation increases according to height was
The seventh International Conference on Urban Climate,
29 June - 3 July 2009, Yokohama, Japan
seen. This is suggested a lot of supply of O3 from higher sky
generates NO2 much more. The NO is supplied from vehicles which
are running on the ground, on the other hand, O3 is supplied from the
sky. Table 2 shows that a supply of O3 was insufficient lower than 3m
height environment, and the oxidation reaction was not enough during
an advection from the south curbside to the 20m remote point.
Furthermore, there were many elements promoting deposition such
as ground surface or the trees and leaves near by, and it is thought
that those elements made the discrepancies between the values of
the NO loss and the NO2 generation.
Table 2 Concentration differences
and rate of changes (parenthesis)
by the reaction during an
advection from the south curbside
point to the 20m remote point
CONCLUSIONS
1. With two NOx monitor of the high-speed response, the
concentration variation of NO and NOx at roadside was measured. Furthermore, a diffusion process of the
pollutants from vehicles and the NO oxidation process with ambient O3 were able to be understood by using
multipoint-simultaneous observation.
2. Concentration-change depending on signal cycle was seen, and cyclical variation arranged in an elapsed
time of the signal was able to be observed. The cycle variation depending on the signal was confirmed by the
20m remote site (SiteB).
3. From the cycle variation of potential ozone (PO) concentration, the direct exhaust NO2 (f-NO2) was
estimated, and, 7.3% of f-NO2 was obtained around the Noge intersection of Ring 8. It was observed that the
spatial tendency of NO2/NOx ratio grew large in proportion to the distance from the road.
4. The NO2/NOx ratio depended on O3 concentration, and the increase tendency with O3 concentration was
seen. During an advection from the south curbside (Site A) to Site B, more than 40% of NO oxidized and
generated NO2 at 10m height. At 3m height, 7% of NO2 was generated and the same value of O3 was
consumed. Unknown NO consumption was seen approximately 6% at 3m height. At 1m height, NO2 did not
generate, but there was unknown NO consumption of 16%. This is suggested a lot of supply of O3 from higher
sky generates NO2 at higher level, and the deposition process disappeared the NO at lower level.
ACKNOWLEDGEMENT
This research was performed as a part of JATOP (a vehicle and fuel diversification by the fuel technology,
CO2 reduction, program for atmospheric improvements) which the Petroleum Energy Center carried out for
auto industry and collaboration of the petroleum industry.
REFERENCES
Abbott, J., 2005. Primary nitrogen dioxide emissions from road traffic: analysis of monitoring data, AEA
Technology, National Environmental Technology Centre, Report AEAT-1925.
Air quality expert group, 2007. Trends in Primary Nitrogen Dioxide in the UK, ISBN 978-0-85521-179-0,
pp.139
Clapp, L.J., Jenkin, M.E., 2001. Analysis of the relationship between ambient levels of O3, NO2, and NO as
a function of NOx in the UK, Atmospheric Environment, 35, 6391-6405.
Delfino, R.J., Staimer, N., Tjoa, T., Gillen, D., Kleinman, M.T., Sioutas, C., Cooper ,D., 2008. Personal and
Ambient Air Pollution Exposures and Lung Function Decrements in Children with Asthma, Envionmental
Health Perspectives, 116, 550-558.
Fjæraa, A.M., Hjellbrekke, A.G., 2007. EMEP Co-operative Programme for Monitoring and Evaluation of
the Long-range Transmission of Air Pollutants in Europe, Ozone measurements 2005, EMEP/CCC-Report
2/2007, O-99074
Gauderman, W.J., Avol, E., Lurmann, F., Kuenzli, N., Gilliland, F., Peters, J., McConnell, R., 2005.
Childhood Asthma and Exposure to Traffic and Nitrogen Dioxide, Epidemiology, 16, 1-7.
Grice, S., Stedman, J., Kent, A., Hobson, M., Norris, J., Abbott, J., Cooke, S., 2009. Recent trends and
projections of primary NO2 emissions in Europe, Atmospheric Environment, 43, 2154-2167.
Ministry of the Environment, 2002. A guide of the NOx / PM Law for vehicles (in Japanese).
Minoura, H., 1999. Some characteristics of surface ozone concentration observed in an urban atmosphere,
Atmospheric Research, 51, 153-169.
Shorter, J.H., Herndon, S., Zahniser, M.S., Nelson, D.D., Wormhoudt, J., Demerjian, K.L., Kolb, C.E., 2005.
Real-Time Measurements of Nitrogen Oxide Emissions from In-Use New york City Transit Buses Using a
Chase Vehicle, Environmental Science & Technology, 39, 7991-8000.
Yao, X., Lau, N.T., Chan, C.K., Fang, M., 2005. The use of tunnel concentration profile data ot determine
the ratio of NO2/NOx directly emitted from vehicles, Atmospheric Chemistry and Physics Discussions, 5,
12723-12740