Environ. Sci. Technol. 2009, 43, 75–81
Source Identification of Reactive Hydrocarbons and Oxygenated VOCs in the Summertime in Beijing Y I N G L I U , † M I N S H A O , * ,† WILLIAM C. KUSTER,‡ P A U L D . G O L D A N , ‡ X I A O H U A L I , * ,† SIHUA LU,† AND JOOST A. DE GOUW‡ State Joint Key Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking University, Beijing, China, and Chemical Sciences Division, NOAA Earth System Research Laboratory, Boulder, Colorado
Received June 21, 2008. Accepted October 23, 2008.. Revised manuscript received October 3, 2008
It is important to identify the sources of reactive volatile organic compounds (VOCs) in Beijing for effective groundlevel ozone abatement. In this paper, semihourly measurements of hydrocarbons and oxygenated VOCs (OVOCs) were taken at an urban site in Beijing in August 2005. C2-C5 alkenes, isoprene, and C1-C3 aldehydes were determined as “key reactive species” by their OH loss rates. Principal component analysis (PCA) was used to define the major sources of reactive species and to classify the dominant air mass types at the sampling site. Vehicle exhaust was the largest contributor to reactive alkenes. More aged air masses with enriched OVOCs traveled mainly from the east or southeast of Beijing. The OVOC sources were estimated by a least-squares fit approach and included primary emissions, secondary sources, and background. Approximately half of the C1-C3 aldehydes were attributed to secondary sources, while regional background accounted for 21-23% of the mixing ratios of aldehydes. Primary anthropogenic emissions were comparable to biogenic contributions (10-16%).
1. Introduction Air quality has been of increasing concern in Beijing, and much of the attention that has been drawn to its is due to its hosting of the 29th Summer Olympic Games in August of 2008 (1). Although efforts by the municipal government have been successful in cutting SO2 emissions and total suspended particles (TSP) (2), emissions from mobile sources have increased (3). Furthermore, high concentrations of fine particles and ground-level ozone are difficult to control. Hundreds of nonmethane hydrocarbons (NMHCs) and oxygenated volatile organic compounds (OVOCs) play an important role in photochemistry. Reactive alkenes and aromatics had a significant impact on ground-level ozone formation in urban areas including Mexico (4), southern Taiwan (5), Seoul (6), and Beijing (7). Biogenic alkenes and some OVOCs were identified to be mainly responsible for ozone formation in the New England area of the U.S. (8). * Corresponding author phone: (86-10) 62757973; e-mail: mshao@ pku.edu.cn. † Peking University. ‡ NOAA Earth System Research Laboratory. 10.1021/es801716n CCC: $40.75
Published on Web 12/03/2008
2009 American Chemical Society
Thus, it is essential to understand the contributions of these sources of reactive VOCs to design efficient control strategies. However, the sources of reactive VOCs are still unknown in Beijing. Receptor models have been widely used in source apportionments of ambient VOCs (6, 9–17). The results of a positive matrix factorization (PMF) model for the summer of 2005 in Beijing suggested that gasoline-related sources were the major VOC contributor (∼52%) (18). However, one basic weakness of these models is their neglect for chemical losses and productions from the sources to the receptor site. Thus, receptor models including PMF and chemical mass balance (CMB) are not applicable for compounds with short lifetimes or those with large secondary sources. Another model, principal component analysis (PCA), can be used to identify the source of reactive VOCs species by investigating the similarity in variation of these species with the nonreactive species whose sources are better known and tracers of primary emission and secondary production (19). A method was developed to quantify the source contributions to OVOCs by a least-squares fit from a series of equations involving the removal of primary VOCs, their primary emission ratios, and the production of secondary OVOCs (20). In August of 2005, a field campaign was conducted at an urban site in the northwestern region of Beijing. The major goals of this study were to determine the relative role of VOC species on ozone formation and to characterize the dominant sources of VOCs. The focus of this work is to explore the major sources of reactive VOCs species: a principal component analysis (PCA) model was used to identify the dominant sources of VOCs qualitatively, and a least-squares approach was employed for quantification of OVOC sources. Back trajectories were calculated to roughly distinguish the influence of air masses outside Beijing from local emissions.
2. Materials and Methods 2.1. Sampling Site Description. Measurements of ambient VOCs, CO, NOx, O3, SO2, and PM2.5 were made from the roof of a six-floor building on the campus of Peking University (PKU), located in the northwestern urban area of Beijing. The sampling site (39.99°N, 116.31°E) is situated about 500 m north of the Fourth Ring Road and faces another road 200 m to the east. PKU site was considered as a typical urban site in Beijing city (18).Meteorological data were acquired at a weather station about 500 m northeast of the measurement site. Figure 1a shows the measured meteorological parameters during the field campaign. At the beginning of August 2005, it was mostly sunny with two rain events. Between 10 and 12 August, the weather became hotter and wetter with lower visibility (period (1) in Figure 1), and the average relative humidity increased to 85%, indicating very poor atmospheric ventilation. On 16 and 17 August, typhoon “Matsa” brought thunderstorms to Beijing (period (2) in Figure 1). Clear, cooler, sunny days followed. The relatively stable atmospheric conditions at the end of the month favored photochemical processing. 2.2. VOC Measurements. NMHCs and OVOCs were measured simultaneously using an online gas chromatography-mass spectrometry/flame ionization detection (GCMS/FID) system, which was deployed by the Aeronomy Laboratory (NOAA), U.S. The system is described in detail elsewhere (8). Briefly, samples were collected cryogenically by two parallel sampling channels for 5 min every half-hour; one channel used a FID and one used an electron impact MS. The FID channel was designed for the analysis of C2-C5 alkanes, C2-C4 alkenes, and acetylene on an Al2O3 PLOT column. The MS channel used a moderate polarity DB-624 VOL. 43, NO. 1, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
75
FIGURE 1. Time series of total measured NMHCs and OVOCs, temperature, relative humidity (RH%), precipitation, and wind vectors at PKU during the study period, where (1) indicates the days with high RH% and high temperature and (2) represents the period influenced by typhoon Matsa. After the cold front passed, the weather was clear for the remainder of the experiment. Wind vectors point in the direction the wind is blowing, and vector length is proportional to speed.
FIGURE 2. Backward trajectories for representative days when factors 1 and 2 dominated (red) and when factor 3 dominated (blue) at PKU. column for the detection of heavier hydrocarbons, OVOCs, and halides, including C5-C10 alkanes and alkenes, C6-C9 aromatics, C1-C5 alcohols, and C2-C7 aldehydes and ketones. Detection limits for most compounds were 100 ppbv) during hot and humid days, as indicated by period 1 in Figure 1b; the maximum mixing ratios were recorded on 11 and 12 August and exceeded 160 ppbv. During this period, OVOCs comprised 35.3% of the total VOCs, which was lower than the average level (42.4%), suggesting that the enhanced VOCs resulted from a stagnant boundary layer rather than from photochemical processes. The rain event on 16-17 August cleaned the air, and cooler, dryer days with lower VOC mixing ratios followed. 3.2. Contribution of VOCs to OH Loss Rates. The OH loss rate, which is the product of the VOC mixing ratio and its OH reaction rate coefficient, is used to assess the reactivity of different species (22–24). Although this method cannot quantify the actual ozone produced by VOCs directly (25), it does provide a way to compare the relative importance of gases on daytime photochemistry (8, 26).The OH loss rates for VOCs, CO, NO, NO2, and SO2 in the daytime are discussed: note that methane and formaldehyde were not measured. The value for the OH loss rate of methane assumes a mixing ratio of 2.5 ppm based on a previous measurement at the site (27). The mixing ratio of formaldehyde was calculated by multiplying the measured acetaldehyde by a factor of 2.5, a ratio derived from other measurements in Beijing (28), Hong Kong (29), and Houston (30). VOCs accounted for 50% of the total OH loss frequency, followed by CO (32%) and NO2 (14%). The relative importance of VOC groups to the OH loss rate is shown in Table 1. Anthropogenic alkenes made up a large fraction (∼27%). Formaldehyde alone contributed 17%, which was comparable to the sum of all other measured aldehydes. The relative contribution of biogenic alkenes, including isoprene, pinenes, and limonene, was typically 12%, but increased to 20% in cleaner air masses. The contribution
TABLE 1. OH Loss Rates by Compound Class (s-1) VOC groups
OH loss rate, s-1 (mean ( sd)
methane gC2 alkanes alkenes aromatics biogenic alkenes formaldehyde C2-C4 aldehydes C3-C4 ketones C1-C3 alcohols
0.39 0.92 ( 0.53 3.86 ( 2.23 1.47 ( 0.89 1.64 ( 0.69 2.37 ( 1.23 2.42 (1.21 0.39 ( 0.21 0.63 ( 0.28
% 1.4 3.3 13.6 5.2 5.8 8.4 8.6 1.4 2.2
other gases CO NO NO2 SO2
OH loss rate of VOCs total OH loss rate
OH loss rate, s-1 (mean ( sd)
%
9.13 ( 4.76 0.58 ( 0.57 4.08 ( 2.27 0.33 ( 0.29
32.3 2.1 14.5 1.2
14.13 ( 5.49 28.25 ( 11.52
50.0 100
TABLE 2. Parameters for OVOCs from the Fit EROVOC ERprecursor ppbv [ppbv C2H2]-1
species acetaldehyde propanal n-butanal acetone 2-butanone methanol ethanol a
0.47 ( 0.09 0.13 ( 0.02 0.02 ( 0.007 0.24 ( 0.24 0.07 ( 0.02 1.07( 0.03 1.01 ( 0.03
4.3 ( 0.07 1.1 ( 0.02 0.17 ( 0.005 0.47 ( 0.23 0.14 ( 0.20 0a 0a
kOVOC kprecursor 10-12 cm3 molecule-1 s-1 15.8 20 24 0.18 1.22 0.94 3.2
1.9 ( 0.6 2.1 ( 0.7 5.7 ( 0.7 1.6 ( 1 3.2 ( 2 0a 0a
ERbiogenic ppbv [ppbv isoprene]-1
background ppt
0.19 ( 0.03 0.046 ( 0.008 0.011 ( 0.002 0.34 ( 0.03 0.037 ( 0.006 0.50 ( 0.08 0.079 ( 0.05
719 ( 88 167 ( 24 41 ( 6 1710 ( 86 82 ( 19 4310 ( 243 1030 ( 160
The parameter was forced to zero because negative values were obtained from the best fit.
TABLE 3. Results of the source apportionment for measured OVOCs (%) species
r
primary anthropogenic, %
secondary anthropogenic, %
biogenic, %
background, %
acetaldehyde propanal n-butanal acetone 2-butanone methanol ethanol
0.81 0.79 0.83 0.75 0.81 0.75 0.74
16 ( 11 14 ( 13 8(9 30 ( 8 47 ( 9 48 ( 12 74 ( 10
48 ( 15 51 ( 15 57 ( 14 9(4 26 ( 9 0a 0a
13 ( 6 13 ( 6 12 ( 6 18 ( 8 13 ( 7 11 ( 5 4(2
22 ( 9 21 ( 9 23 ( 9 43 ( 11 14 ( 7 41 ( 11 22 ( 10
a
The parameter was forced to zero because negative values were obtained from the best fit.
of gC2 alkanes and C-C3 alcohols were 6.1 and 4.5% on average, respectively. Ketones played minor (∼2.9%) roles in all conditions. Of the measured VOCs, alkenes and aldehydes contributed most to the OH reactivity. In particular, C2-C5 anthropogenic alkenes, isoprene, and C1-C3 aldehydes and aromatics were the key reactive species measured in the urban area of Beijing. Hence, identification of the sources of these species is important for the management of air quality in Beijing.
4. Discussion of VOC Sources 4.1. Source Identification and Air Mass Classification. Because the ability to quantify the contribution of chemical removal processes and to locate secondary sources of reactive NMHCs and OVOCs using the PMF and CMB models is limited, the principal component analysis (PCA) method was used to explain relationships between VOC species and to identify sources (19). Here, PCA was applied for concurrent measurement of NMHCs, OVOCs, and other trace gases (O3, NOx, CO, and SO2) to interpret major sources of VOCs and to determine the dominant air mass types and source regions with the help of a back trajectory model. Six factors were extracted that explained 85.7% of the total variance for 39 compounds. Factor 1 was associated with some short-lived compounds such as ethylene, propylene, 1-butene, and aromatics, in addition to acetylene, CO and NOx (from incomplete combustion), C4-C5 alkanes (gasoline components), and methyl tert-butyl ether (MTBE, gasoline additive). This factor was attributed to local emissions of gasoline
vehicle exhaust. Propanal, n-butanol, i-propanol, and other oxygenated species were also correlated (>0.60) with this factor, indicating that primary anthropogenic emissions also affected urban OVOCs. Factor 2, which appears to reflect the direct emissions of gasoline evaporation, had i-pentane, butanes, MTBE, and C4-C5 alkenes as major contributors with negligible loadings of acetylene, NOx, and CO. Factors 1 and 2 showed relatively fresh emissions from urban areas that accounted for ∼54% of the total variance. Factor 3, explaining approximately 13% of the variance, had high factor loadings (0.7-0.9) of aldehydes and ketones and also correlated with O3, which implied that it was dominated by photochemical processing and influenced by aged air masses. Factor 4 was loaded with ethylene, propylene, C8 aromatics, C4 alkenes, and NOx suggesting direct emissions from diesel exhaust. Factor 5 was heavily correlated with SO2 and CO due to industrial emissions. Factor 6 was dominated by isoprene, indicating a biogenic source. Aldehydes (gC2) and ketones were derived not only from anthropogenic emissions directly, but also from photochemical production. Differences between the scores of the above factors, combined with back trajectory modeling (HYSPLIT) (31), can be used to distinguish the photochemical aged air masses from relatively fresh ones. During 12-14 August, scores of factors 1 and 2 were greater than for factor 3, suggesting that those days were governed by fresh air masses. During 20-27 August, factor 3, associated with aged VOL. 43, NO. 1, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
77
4.2.2. Estimation of Isoprene from Biogenic Emission (ISOPsource). It is assumed that OVOCs derived from biogenic sources are proportional to ISOPsource in the third term of eq 1 and that biogenic sources for isoprene and photochemical production of methyl vinyl ketone (MVK) and methacrolein (MACR) dominate their ambient mixing ratios at the PKU site during the summertime. Thus, before the regression for OVOCs is calculated, ISOPsource during the daytime should be roughly extrapolated from the observed ratios of [MACR]/ [isoprene] and [MVK]/[isoprene]. First, the processing time of isoprene experienced in an air mass was estimated from eqs 3-1 and 3-2 (36–38). The initial isoprene was then derived from the reaction time and the observed mixing ratio of isoprene. ISOP + OH f 0.63HCHO + 0.32MVK + 0.23MACRk1 ) FIGURE 3. Relationship between measured [MVK+MACR] isoprene.
initial
(ISOPsource)
and
air masses, was comparable to or larger than one standard deviation, indicating a greater contribution of photochemical processes. Figure 2 compares the backward trajectories during periods influenced by fresh emission air masses and those affected by more aged/processed air masses. The air masses dominated by factors 1 and 2 usually came from southern suburban Beijing or the Hebei Province, representing fresher urban emissions. More aged air masses originated from the east or southeast and traveled through the neighboring Tianjin Municipality and Hebei Province. Emissions from sources outside Beijing, having undergone chemical reactions during transport, added more secondary products. The contributions of primary and secondary sources to OVOC mixing ratios are explored in detail in Section 4.2 4.2. Quantification of OVOC Sources. 4.2.1. Methodology. As mentioned above, current receptor models are not applicable to short-lived compounds or secondary products from photochemical reactions. A method was developed for determining the sources of ambient OVOCs (20).The measured OVOC mixing ratios were divided into four terms, i.e., primary anthropogenic emissions, secondary anthropogenic OVOCs, primary biogenic sources, and background. In eq 1, EROVOC and ERprecursor refer to the primary anthropogenic emission ratios of OVOC and its precursor versus acetylene, and ERbiogenic represents the coefficient of OVOC relative to isoprene from biogenic emissions (ISOPsource). The photochemical age (∆t) can be indirectly estimated from the observed decay of two hydrocarbons (32–34). Here, the measured ratios between toluene and benzene were used in eq 2. The parameters kOVOC and kC2H2 are the rate coefficients for the reaction with OH radicals (35), and other parameters including EROVOC, ERprecursor, ERbiogenic, kprecursor, and [background] are estimated from a least-squares fit. The major limitations of this method lie in the mixing effects between air masses with different ages and origins. The photochemical ages calculated by eq 2 have restrictions for species with higher OH reactivity than toluene (20). [OVOC] ) EROVOC × [C2H2] × exp(-(KOVOC)[OH]∆t) + Kprecussior × ERprecussior × [C2H2] × KOVOC - Kprecussior exp(-Kprecussior[OH]∆t) - exp(-KOVOC[OH]∆t) + exp(-KC2H2[OH]∆t) ERisogenic × (isoprenesource) + [background] (1) ∆t )
1 [toluence] × ln | [OH](ktoluence - kbenzene) benzene t-0 [toluence] ln | benzene t-t
[ (
)
(
78
9
)] (2)
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 1, 2009
1.0 × 10-10cm-3s-1 MACR + OH f productsk2 ) 3.3 × 10-11cm-3s-1 MVK + OH f productsk3 ) 1.9 × 10-11cm-3s-1 0.23k1 [MACR] t ) (1 - e(k2-k1)[OH]avg ) [ISOP] (k2 - k1)
(3-1)
0.32k1 [MVK] t ) (1 - e(k1-k3)[OH]avg ) [ISOP] (k3 - k1)
(3-2)
Figure 3 shows a plot of the initial isoprene (ISOPsource) versus the measured values [MVK +MACR] for different processing times (∆tbiogenic) of isoprene. Assuming that the 24 h average concentration of OH is 2.5 × 106 molecules · cm-3, the processing times of isoprene were calculated to be between 0.4 and 4.2 h with the bulk of the points near 2.4 h. A good correlation between ISOPsource and MVK + MACR was obtained where the slope of the regression lines tended to be larger at higher processing times. Since the slope is related to the yield of MVK + MACR from isoprene and further oxidation of those two products, data points away from the dashed line are likely due to chemical losses of MVK and MACR and/or the influence of continuous emissions of isoprene. 4.2.3. Source Contribution to OVOCs. Table 2 summarizes the parameters for each OVOC species. Calculated emission ratios in Beijing are lower than those on the northeast coast of the U.S. (20), which may result from higher acetylene mixing ratios in Beijing. The emission ratios of acetaldehyde from gasoline vehicles, diesel trucks, and wood combustion were 0.18 (39), 5.4 (40), and 1.43 (41), respectively. As gasoline vehicles account for the dominant fraction (∼85%) of vehicles in Beijing, the acetaldehyde emission ratio averaged over the total vehicles will be close to those from gasoline vehicles. The emission ratio of acetaldehyde derived from the fit was 0.47 ( 0.09, indicating that the primary acetaldehyde in Beijing was influenced by gasoline exhaust and biomass burning. In addition, solvent evaporation is also a possible source for primary OVOCs. The average source contributions to OVOCs are listed in Table 3, which show that aldehydes were largely secondary products, whereas ketones and alcohols were predominantly primary emissions. About half of the aldehydes were attributed to secondary formation with a smaller fraction from background (∼22%); the primary anthropogenic and biogenic contributions were comparable (10-16%). Such results for aldehydes are similar to those measured in the New England area (20). Background contributed most to acetone (43%), followed by primary anthropogenic (30%) and biogenic (18%) contributions, and a lower portion (9%) of acetone resulted from a secondary anthropogenic source which differs from the results in the New England area (20). Unlike acetone, methyl ethyl ketone (MEK) had the largest fraction of anthropogenic sources (47%), with a lower loading (26%)
FIGURE 4. Comparisons of measured and calculated concentrations of acetaldehyde, MEK, and methanol between 15 and 25 August 2005. from secondary sources. The two alcohols listed were dominated by primary sources with negligible secondary sources. Ethanol had the most significant primary anthropogenic contribution. More than 40% of the acetone and methanol were attributed to background which implies that the entire region around Beijing is a source region of these compounds. Most of the calculated OVOCs are close to their measured concentrations at the selected time, as described in Figure 4. The contribution of primary anthropogenic emissions increased in the early morning and late evening, whereas secondary and biogenic contributions reached their maximum at noon. The calculated results underestimated the peaks of aldehydes at noon on 15 August, presumably because the misty and muggy weather led to the accumulation of pollutants. As mentioned in Section 3.1, OVOC mixing ratios decreased with the rainfall event between 16 and 17 August. On 18-19 August, OVOCs increased with sunny weather, and primary anthropogenic emissions became the most dominant sources of OVOCs. With the abundance of sunshine at midday, aldehydes at noon were mostly attributed to secondary anthropogenic sources.
Therefore, alkenes and aldehydes are the most reactive VOC groups in terms of OH reactivity in Beijing. Most reactive alkenes were primarily from mobile sources. The vehicle population in Beijing has increased at an average annual rate of 14.5% since 1997. In August of 2003, it exceeded 2 million and was estimated to reach 3.25 million in 2008 (3). Such a large vehicle fleet will have a significant negative effect on the air quality of Beijing. About half-of gC2 aldehydes are formed in photochemical processes. It is also worth noting that emissions from the greater Beijing metropolitan area affected the ambient level of reactive VOCs, especially the secondary species, which suggests that additional control measures for Beijing neighborhoods should be taken into consideration to reduce ozone levels.
Acknowledgments This study was supported by the National Natural Science Foundation of China (40575059) and National Basic Research Program of China (2002CB410801). We thank Jie-tai Mao, Zhi-jun Wu, and Bin Wang for gases and meteorological data. We also thank Yuan Liu, Xin Li, Xin Xie, Qiao-qiao Wang, and Yuan-zi Wu from PKU. VOL. 43, NO. 1, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
79
Supporting Information Available The results of principal component analysis. This material is available free of charge via the Internet at http:// pubs.acs.org.
Literature Cited (1) Streets, D. G.; Fu, J. H. S.; Jang, C. J.; Hao, J. M.; He, K. B.; Tang, X. Y.; Zhang, Y. H.; Wang, Z. F.; Li, Z. P.; Zhang, Q.; Wang, L. T.; Wang, B. Y.; Yu, C. Air quality during the 2008 Beijing Olympic Games. Atmos. Environ. 2007, 41 (3), 480–492. (2) Hao, J. M.; Wang, L. T. Improving urban air quality in China: Beijing case study. J. Air Waste Manage. Assoc. 2005, 55 (9), 1298–1305. (3) Hao, J. M.; Wang, L. T.; Li, L.; Hu, J. N.; Yu, X. C. Air pollutants contribution and control strategies of energy-use related sources in Beijing. Sci. China, Ser. D: Earth Sci. 2005, 48 (Supp. II), 138–146. (4) Blake, D. R.; Rowland, F. S. Urban leakage of liquefied petroleum gas and its impact on Mexico City air quality. Science 1995, 269 (5226), 953–956. (5) Chang, C. C.; Chen, T. Y.; Lin, C. Y.; Yuan, C. S.; Liu, S. C. Effects of reactive hydrocarbons on ozone formation in southern Taiwan. Atmos. Environ. 2005, 39 (16), 2867–2878. (6) Na, K.; Moon, K. C.; Kim, Y. P. Source contribution to aromatic VOC concentration and ozone formation potential in the atmosphere of Seoul. Atmos. Environ. 2005, 39 (30), 5517–5524. (7) Shao, M.; Fu, L. L.; Liu, Y.; Lu, S. H.; Zhang, Y. H.; Tang, X. Y. Major reactive species of ambient volatile organic compounds (VOCs) and their sources in Beijing. Sci. China, Ser. D: Earth Sci. 2005, 48 (Supp. II), 147–154. (8) Goldan, P. D.; Kuster, W. C.; Williams, E.; Murphy, P. C.; Fehsenfeld, F. C.; Meagher, J. Nonmethane hydrocarbon and oxy hydrocarbon measurements during the 2002 New England Air Quality Study. J. Geophys. Res. , [Atmos.] 2004, 109, D21309. doi:10.1029/2003JD004455. (9) Fujita, E. M.; Watson, J. G.; Chow, J. C.; Lu, Z. Validation of the chemical mass balance receptor model applied to hydrocarbon source apportionment in the Southern California air quality study. Environ. Sci. Technol. 1994, 28 (9), 1633–1649. (10) Liu, Y.; Shao, M.; Zhang, J.; Fu, L. L.; Lu, S. H. Distributions and source apportionment of ambient volatile organic compounds in Beijing city, China. J. Environ. Sci. Health, Part A: Toxic/ Hazard. Subst. Environ. Eng. 2005, 40 (10), 1843–1860. (11) Vega, E.; Mugica, V.; Carmona, R.; Valencia, E. Hydrocarbon source apportionment in Mexico City using the chemical mass balance receptor model. Atmos. Environ. 2000, 34 (24), 4121– 4129. (12) Watson, J. G.; Chow, J. C.; Fujita, E. M. Review of volatile organic compound source apportionment by chemical mass balance. Atmos. Environ. 2001, 35 (9), 1567–1584. (13) Lai, C. H.; Chen, K. S.; Ho, Y. T.; Peng, Y. P.; Chou, Y. M. Receptor modeling of source contributions to atmospheric hydrocarbons in urban Kaohsiung, Taiwan. Atmos. Environ. 2005, 39 (25), 4543–4559. (14) Guo, H.; Wang, T.; Blake, D. R.; Simpson, I. J.; Kwok, Y. H.; Li, Y. S. Regional and local contributions to ambient non-methane volatile organic compounds at a polluted rural/coastal site in Pearl River Delta, China. Atmos. Environ. 2006, 40 (13), 2345– 2359. (15) Zhao, W. X.; Hopke, P. K.; Karl, T. Source identification of volatile organic compounds in Houston, Texas. Environ. Sci. Technol. 2004, 38 (5), 1338–1347. (16) Xie, Y. L.; Berkowitz, C. M. The use of positive matrix factorization with conditional probability functions in air quality studies: An application to hydrocarbon emissions in Houston, Texas. Atmos. Environ. 2006, 40 (17), 3070–3091. (17) Kim, E.; Brown, S. G.; Hafner, H. R.; Hopke, P. K. Characterization of non-methane volatile organic compounds sources in Houston during 2001 using positive matrix factorization. Atmos. Environ. 2005, 39 (32), 5934–5946. (18) Song, Y.; Shao, M.; Liu, Y.; Lu, S. H.; Kuster, W.; Goldan, P.; Xie, S. D. Source apportionment of ambient volatile organic compounds in Beijing. Environ. Sci. Technol. 2007, 41 (12), 4348– 4353. (19) Millet, D. B.; Donahue, N. M.; Pandis, S. N.; Polidori, A.; Stanier, C. O.; Turpin, B. J.; Goldstein; A. H. Atmospheric volatile organic compound measurements during the Pittsburgh Air Quality Study: Results, interpretation, and quantification of primary and secondary contributions J. Geophys. Res., [Atmos.] 2005, 110, D07S07, doi: 10.1029/2004JD004601. 80
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 43, NO. 1, 2009
(20) De Gouw, J. A.; Middlebrook, A. M.; Warneke, C.; Goldan, P. D.; Kuster, W. C.; Roberts, J. M.; Fehsenfeld, F. C.; Worsnop, D. R.; Canagaratna, M. R.; PszennyA. A. P. Budget of organic carbon in a polluted atmosphere: Results from the New England Air Quality Study in 2002 J. Geophys. Res., [Atmos.] 2005, 110, D16305, doi: 10.1029/2004JD005623. (21) Kuster, W. C.; Jobson, B. T.; Karl, T.; Riemer, D.; Apel, E.; Goldan, P. D.; Fehsenfeld, F. C. Intercomparison of volatile organic carbon measurement techniques and data at La Porte during the TexAQS2000 Air Quality Study. Environ. Sci. Technol. 2004, 38 (1), 221–228. (22) Goldan, P. D.; Kuster, W. C.; Fehsenfeld, F. C. Nonmethane hydrocarbon measurements during the Tropospheric OH Photochemistry Experiment. J. Geophys. Res., [Atmos.] 1997, 102 (D5), 6315–6324. (23) Riemer, D.; Pos, W.; Milne, P.; Farmer, C.; Zika, R.; Apel, E.; Olszyna, K.; Kliendienst, T.; Lonneman, W.; Bertman, S.>. Observations of nonmethane hydrocarbons and oxygenated volatile organic compounds at a rural site in the southeastern United States. J. Geophys. Res., [Atmos.] 1998, 103 (D21), 28111– 28128. (24) Winkler, J.; Blank, P.; Glaser, K.; Gomes, J. A. G.; Habram, M.; Jambert, C.; Jaeschke, W.; Konrad, S.; Kurtenbach, R.; Lenschow, P. Ground-based and airborne measurements of nonmethane hydrocarbons in BERLIOZ: Analysis and selected results. J. Atmos. Chem. 2002, 42 (1), 465–492. (25) Carter, W. P. L. Development of ozone reactivity scales for volatile organic compounds. J. Air Waste Manage. Assoc. 1994, 44 (7), 881–899. (26) Jobson, B. T.; Berkowitz, C. M.; Kuster, W. C.; Goldan, P. D.; Williams, E. J.; Fesenfeld, F. C.; Apel, E. C.; Karl, T.; Lonneman, W. A.; Riemer, D. Hydrocarbon source signatures in Houston, Texas: Influence of the petrochemical industry J. Geophys. Res., [Atmos.] 2004, 109, D24305, doi: 10.1029/2004JD004887. (27) Su, F. Study on the variation and sources of volatile organic compounds in Beijing (in Chinese). Doctoral dissertation, Peking University, 2003. (28) Pang, X. B.; Mu, Y. J. Seasonal and diurnal variations of carbonyl compounds in Beijing ambient air. Atmos. Environ. 2006, 40 (33), 6313–6320. (29) Ho, K. F.; Lee, S. C.; Louie, P. K. K.; Zou, S. C. Seasonal variation of carbonyl compound concentrations in urban area of Hong Kong. Atmos. Environ. 2002, 36 (8), 1259–1265. (30) Wert, B. P.; Trainer, M.; Fried, A.; Ryerson, T. B.; Henry, B.; Potter, W. ; Angevine, W. M.; Atlas, E.; Donnelly, S. G.; Fehsenfeld, F. C. ; et al., Signatures of terminal alkene oxidation in airborne formaldehyde measurements during TexAQS 2000 J. Geophys. Res., [Atmos.] 2003, 108, (D3), doi: 10.1029/2002JD002502. (31) Chen, M.; Talbot, R.; Mao, H. T.; Sive, B. ; Chen, J. J.; Griffin, R. J. , Air mass classification in coastal New England and its relationship to meteorological conditions J. Geophys. Res., [Atmos.] 2007, 112, (D10), D10S05, doi: 10.1029/2006JD007687. (32) Blake, N. J.; Penkett, S. A.; Clemitshaw, K. C. Estimates of atmospheric hydroxyl radical concentrations from the observed decay of many reactive hydrocarbons in well-defined urban plumes. J. Geophys. Res., [Atmos.] 1993, 98 (D2), 2851–2864. (33) McKeen, S. A.; Trainer, M.; Hsie, E. Y.; Tallamrayu, R. K.; Liu, S. C. On the indirect determination of atmospheric OH radical concentrations from reactive hydrocarbon measurements. J. Geophys. Res., [Atmos.] 1990, 95 (D6), 7493–7500. (34) Roberts, J. M.; Fehsenfeld, F. C.; Liu, S. C.; Bollinger, M. J.; Hahn, C.; Albritton, D. L.; Sievers, R. E. Measurements of aromatic hydrocarbon ratios and NOx concentrations in the rural troposphere: Observation of air mass photochemical aging and NOx removal. Atmos. Environ. 1984, 18, 2421–2432. (35) Atkinson, R.; Arey, J. Atmospheric degradation of volatile organic compounds. Chem. Rev. 2003, 103 (12), 4605–4638. (36) Stroud, C. A.; Roberts, J. M.; Goldan, P. D.; Kuster, W. C.; Murphy, P. C.; Williams, E. J.; Hereid, D.; Parrish, D.; Sueper, D.; Trainer, M.; Fehsenfeld, F. C.; Apel, E. C.; Riemer, D.; Wert, B.; Henry, B.; Fried, A.; Martinez-Harder, M.; Harder, H.; Brune, W. H.; Li, G.; Xie, H.; Young, V. L. Isoprene and its oxidation products, methacrolein and methylvinyl ketone, at an urban forested site during the 1999 Southern Oxidants Study. J. Geophys. Res., [Atmos.] 2001, 106 (D8), 8035–8046. (37) Apel, E. C.; Riemer, D. D.; Hills, A.; Baugh, W.; Orlando, J.; Faloona, I.; Tan, D.; Brune, W.; Lamb, B. Westberg, et al., Measurement and interpretation of isoprene fluxes and isoprene, methacrolein, and methyl vinyl ketone mixing ratios at the PROPHET site during the 1998 Intensive. Journal of Geophysical Research-Atmospheres 2002, 107, (D3), 4034, doi: 10.1029/2000JD000225.
(38) Roberts, J. M.; Marchewka, M.; Bertman, S. B.; Goldan, P.; Kuster, W.; de Gouw, J.; Warneke, C.; Williams, E.; Lerner, B. ; Murphy, P.; Apel, E.; Fehsenfeld, F. C. Analysis of the isoprene chemistry observed during the New England Air Quality Study (NEAQS) 2002 intensive experiment. J. Geophys. Res., [Atmos.] 2006, 111, (D23), D23S12, doi: 10.1029/2006JD007570. (39) Schauer, J. J.; Kleeman, M. J.; Cass, G. R.; Simoneit, B. R. T. Measurement of emissions from air pollution sources. 5. C-1C-32 organic compounds from gasoline-powered motor vehicles. Environ. Sci. Technol. 2002, 36 (6), 1169–1180.
(40) Schauer, J. J.; Kleeman, M. J.; Cass, G. R.; Simoneit, B. R. T. Measurement of emissions from air pollution sources. 2. C-1 through C-30 organic compounds from medium duty diesel trucks. Environ. Sci. Technol. 1999, 33 (10), 1578–1587. (41) Schauer, J. J.; Kleeman, M. J.; Cass, G. R.; Simoneit, B. R. T. Measurement of emissions from air pollution sources. 3. C-1C-29 organic compounds from fireplace combustion of wood. Environ. Sci. Technol. 2001, 35 (9), 1716–1728.
ES801716N
VOL. 43, NO. 1, 2009 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
81
