- Email: [email protected]
Characteristics of organic matter in PM2.5 in Shanghai
Chemosphere 64 (2006) 1393–1400 www.elsevier.com/locate/chemosphere
Characteristics of organic matter in PM2.5 in Shanghai Jialiang Feng a, Chak K. Chan a, Ming Fang b,*, Min Hu c, Lingyan He c, Xiaoyan Tang c a
c
Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China b Institute for Environment and Sustainable Development, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences, Peking University, Beijing, China Received 27 June 2005; received in revised form 5 December 2005; accepted 15 December 2005 Available online 25 January 2006
Abstract Solvent extractable organic compounds (SEOC), organic carbon, elemental carbon and water soluble organic carbon (WSOC) in PM2.5 samples collected in Shanghai, China in 2002 and 2003 were measured to determine the composition and sources of the organic matter in atmospheric aerosols. Distinct seasonal variations were detected with higher concentrations of organic matter in winter. The concentration of total carbon of about 20 lg m 3 in winter was about three times the summer value. About 30% of the total carbon was water soluble. Unresolved complex mixture (UCM) and fatty acids were the most abundant components quantified in SEOC, similar to other Chinese cities previously studied. High ratio of UCM to n-alkanes (U:R) and the composition of triterpanes indicated that engine exhaust was a major source of the airborne organic matter. Emissions from coal burning had more impact in the rural areas, according to the U:R value and PAHs composition. Chemical mass balance (CMB) modeling shows that about half of the organic carbon was from engine exhaust and about 15% was from coal burning. No clear spatial variation in the concentration of the organic matter was found between urban and rural areas. Our results showed that due to the rapid urbanization and relocation of industrial plants from urban areas to rural areas in the past 20 years, air pollution in rural areas is becoming a serious problem in Shanghai and the Yangtze River delta. 2006 Elsevier Ltd. All rights reserved. Keywords: Aerosol; Organic carbon; Water soluble organic carbon; Solvent extractable organic compound; GC–MS; China
1. Introduction Shanghai is the largest commercial and industrial city in China with a population of about 15 million. It has the largest steel mill and petrochemical complex in China. Rapid economic growth in the last two decades has caused soaring energy demand and the annual coal consumption in Shanghai increased from 20 million tons in 1989 to 45 million tons in 2004. Although the annual average concentration of total suspended particulates (TSP) has *
Corresponding author. Tel.: +852 2358 6916; fax: +852 2358 1334. E-mail address: [email protected] (M. Fang).
0045-6535/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.12.026
dropped since the mid-1990s, fine particle loading has remained high (Shanghai Environmental Protection Bureau). The annual PM10 loading in recent years (2001– 2004) is high at 100 lg m 3. Airborne lead pollution in Shanghai has been found to be mainly from the cement and metallurgy industries and coal combustion (Zheng et al., 2004; Chen et al., 2005). Only a small part of the airborne lead (20%) was from automotive emissions. Chemical mass balance (CMB) model analysis of elemental metals showed that coal combustion, construction, vehicle emissions, and steel furnaces were the main contributors of TSP (Shu et al., 2001). Researches on PM2.5 in Shanghai showed that 42% of
1394
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
the mass was from secondary aerosols of ammonium sulfate and nitrate (Ye et al., 2003) and stationary emissions (coal burning) are still believed to be the dominant sources (Yao et al., 2002). A large portion of the PM2.5 mass in Shanghai (40%) is found to be carbonaceous materials (Ye et al., 2003), but data on the detailed speciation of organic species in aerosols is not available. Solvent extractable organic compounds (SEOC) in aerosols contain useful molecular markers that have been successfully used for source apportionment (Simoneit, 1986; Zheng et al., 2000, 2005) and SEOC has been found to be toxic and can cause DNA mutation even at non-lethal dosages (Hsiao et al., 2000). Knowing the composition of the organic matter is also important to understanding the impact of continental aerosols on ocean ecology, because Shanghai is an important starting point of the outflow of continental pollutants to the East China Sea and the Pacific Ocean (Parungo et al., 1994). In this paper, we focus on a detailed study of the abundance and characteristics of SEOC in PM2.5 samples collected in Shanghai on a seasonal basis. The impact of urbanization on the composition of organic aerosols at rural sites will also be discussed. 2. Experimental 2.1. Sampling Samples were collected simultaneously at two sites. One was located on the rooftop of a five-story building, with a height of 15 m, on the campus of Fudan University (FDU). The campus is in the northeast part of the urban area. There is a viaduct with heavy traffic near the campus. The other site was at the Shanghai Observatory (SHO) located on a small hill of 100 m high called Sheshan in the southwestern countryside of Shanghai. It was once used as a background site for air pollution monitoring. There was no direct emission source near the Observatory. The distance between the two sampling site is 40 km. PM2.5 samples were collected during 21–28 November 2002 (high loading season) and 14–21 August 2003 (low loading season) using high-volume samplers and the same procedure as described in our previous publication (Feng et al., 2005). 2.2. SEOC analysis The details of the analysis of SEOC were presented in our previous publications (Zheng et al., 1997; Feng et al., 2005). Briefly, prior to extraction, an internal standard mixture consisting of eicosane-d42, octacosane-d58, phenanthrene-d10, chrysene-d12, perylene-d12 and heptadecanoic acid-d33, was spiked onto the filters. Then, the samples were ultrasonically extracted with three 100-ml aliquots of dichloromethane at room temperature. The combined extract was concentrated to a volume of 2–3 ml, filtered
and reduced further to a volume of 200–300 ll with a stream of high purity N2. The extract was reacted with 14% BF3 in methanol to esterify the free organic acids and then fractionated with a flash column of silica gel into four sub-fractions: aliphatics, PAHs, fatty acids (methyl ester) and alkanols. The total extract and the four fractions were subjected to GC–MS (Finnigan TSQ700 interfaced to a Hewlett Packard Model 5890A gas chromatograph) analyses. The MS was operated in the electron impact mode at 70 eV and the scan range is 50–550 amu. The GC was equipped with a HP-5MS capillary column (30 m · 0.25 mm i.d., film thickness 0.25 lm), with helium as carrier gas. The n-alkanol fraction was converted to trimethylsilyl derivatives by reaction with N,O-bis-(trimethylsilyl) trifluoroacetamide (BSTFA) before analysis. Prior to the GC–MS analysis, hexamethyl benzene was added to all fractions to be used as an internal standard for n-alkanols and to check the recovery of the other three fractions. Alkanes, PAHs and fatty acids were quantified using the corresponding deuterated internal standards having similar chemical characteristics and retention times. 2.3. OC, EC and WSOC measurements Organic and elemental carbon (OC and EC) concentrations of the samples were analyzed by thermal/optical method (Sunset OC/EC analyzer) with the NIOSH temperature program (Birch, 1998). Small samples punched from the high-volume filters (5.8 cm2 each) were extracted with nano-pure water and the concentration of the water soluble organic carbon (WSOC) was measured with a total organic carbon analyzer (Shimadzu TOC-5000 A). Hydrochloric acid was added to each sample before analysis to remove inorganic carbon. 3. Results and discussion 3.1. Loading of OC, EC and WSOC The average OC in November 2002 (16 lg m 3, Table 1) was comparable to the reported winter value (Ye et al., 2003), but the average of August 2003 (3.9 lg m 3, urban and 4.9 lg m 3, rural) was lower than the reported average summer value (10 lg m 3). The occurrence of several high concentration episodes could be an important cause of the higher reported summer value (Ye et al., 2003). TC concentrations of 20 lg m 3 in winter indicated that the pollution level was high in Shanghai even though TC concentrations were lower than in other Chinese cities such as Beijing and Guangzhou (Cao et al., 2003; Duan et al., 2005). The average winter OC/EC ratios (3.9 and 4.5 in urban and rural, respectively) were higher than in summer (2.2 and 2.4, respectively), which could be an effect of lower ambient temperature. The percentage of organic carbon
Table 1 Summary of yields and composition of organic matter in PM2.5 from Shanghai, China Site
FDU (urban)
Date
22/11/02 23/11/02 24/11/02 25/11/02 26/11/02 27/11/02 28/11/02
Wind OC EC WSOC n-alkanes PAHs direction (lg m 3) (lg m 3) (lg m 3) a b c Yield CPI Cmax U:R Wax% Yield BeP/ (ng m 3) (ng m 3) (BaP + BeP)
Yield CPI (ng m 3)
C18:1/ Wax% Yield CPIa Cmaxb Wax% C18 (ng m 3)
NE NE NW NW NW NW–SE SE
Average E E E E S SE SE
Average SHO (rural)
22/11/02 23/11/02 24/11/02 25/11/02 26/11/02 27/11/02 28/11/02
NE NE NW NW NW NW–SE SE
Average 15/8/03 16/8/03 17/8/03 18/8/03 19/8/03 20/8/03 21/8/03 Average
E E E E S SE SE
n-alkanol a
4.8 9.3 9.7 17.0 27.8 26.1 16.2
1.2 3.0 3.0 4.0 6.8 5.9 3.9
2.2 3.2 3.4 6.5 11.8 7.4 5.9
32.9 83.3 70.3 175.5 226.2 341.9 116.6
1.3 1.2 1.3 1.0 1.3 1.1 1.3
24 24 24 23 23 23 24
8.2 9.5 6.2 8.6 7.0 6.0 9.1
11.8 4.6 12.1 3.1 11.2 3.7 9.9
7.8 28.9 30.3 62.7 82.0 115.1 39.7
0.83 0.78 0.71 0.71 0.70 0.63 0.75
123.7 304.0 211.2 397.2 668.0 622.9 401.8
12.9 13.7 11.1 13.3 9.8 12.6 11.5
0.10 0.10 0.10 0.07 0.18 0.15 0.08
13.5 12.4 23.0 14.8 25.3 17.3 20.6
16.8 36.8 30.6 40.2 170.9 62.4 65.2
11.2 12.7 7.6 4.7 6.6 8.3 8.3
28 28 30 30 30 30 30
63.0 74.5 65.7 51.0 71.4 77.9 77.2
15.8
4.0
5.8
149.5
1.2
23
7.8
8.0
52.4
0.73
389.8
12.1
0.11
18.1
60.4
8.5
30
68.7
3.4 2.8 6.1 5.0 4.7 2.5 2.6
1.3 1.1 2.7 2.3 2.2 1.6 1.3
1.6 1.3 1.9 1.9 1.6 0.8 0.7
15.6 12.2 39.2 29.6 25.6 20.5 14.5
1.5 1.4 1.2 1.3 1.6 1.4 1.6
31 29 29 29 29 29 31
19.7 24.3 10.7 16.3 22.0 20.8 24.0
20.8 19.9 11.7 15.4 22.9 17.3 23.7
4.3 3.1 13.5 9.2 10.4 6.1 3.3
0.76 0.80 0.78 0.75 0.65 0.79 0.76
108.5 101.3 156.6 167.5 141.0 93.6 106.5
12.5 11.1 10.1 10.8 14.4 17.1 14.8
0.08 0.09 0.08 0.13 0.17 0.14 0.19
14.5 18.6 21.6 17.7 14.1 8.1 14.3
8.2 8.7 11.8 12.2 9.7 4.4 5.5
9.4 9.1 5.9 8.9 8.0 9.9 15.1
28 28 28 28 28 28 28
60.6 71.0 64.6 69.2 59.6 45.6 69.2
3.9
1.8
1.4
22.5
1.4
29
19.7
18.8
7.1
0.75
125.0
13.0
0.13
15.6
8.6
9.5
28
62.8
8.7 13.5 11.4 15.8 29.8 16.3 20.1
2.3 3.4 2.1 3.7 5.8 4.6 3.7
3.5 4.1 4.8 7.0 12.0 6.1 8.7
101.2 159.8 92.9 101.2 182.3 153.6 143.3
1.1 1.0 1.3 1.3 1.4 1.2 1.2
24 24 25 25 24 23 24
5.3 6.6 5.2 7.3 5.5 6.8 6.4
5.1 1.5 12.2 13.0 14.8 8.3 10.2
29.5 52.1 31.7 37.6 125.6 76.3 47.7
0.74 0.75 0.70 0.73 0.64 0.64 0.77
366.2 455.2 306.6 322.8 613.4 463.2 547.7
15.6 17.1 9.4 8.4 7.4 10.1 10.0
0.13 0.07 0.07 0.08 0.09 0.07 0.05
12.5 10.8 27.5 26.0 36.7 25.9 28.0
39.9 51.1 94.4 102.9 198.7 142.5 170.4
12.0 10.2 9.8 6.2 6.9 12.5 14.5
28 30 30 30 30 30 30
72.6 56.8 60.3 67.7 80.0 57.1 64.5
16.5
3.6
6.6
133.5
1.2
24
6.2
9.3
57.2
0.71
439.3
11.1
0.08
23.9
114.3
10.3
30
65.6
6.2 7.4 6.0 5.9 4.2 2.8 1.7
2.8 2.9 2.5 2.5 1.5 1.5 0.6
2.7 2.9 2.7 2.5 2.1 1.1 0.7
24.5 58.2 27.8 35.2 19.8 23.7 10.2
1.7 1.0 1.3 1.2 1.7 1.5 2.8
29 29 29 29 29 29 29
9.7 7.9 9.8 7.9 11.1 15.8 17.0
25.7 5.4 15.0 12.1 27.1 19.6 45.8
9.4 11.6 9.1 12.5 4.5 12.1 1.3
0.72 0.80 0.76 0.85 0.72 0.75 0.77
147.4 188.2 137.1 156.9 97.9 127.9 60.3
13.0 11.5 9.8 10.2 12.8 12.2 12.7
0.16 0.08 0.31 0.11 0.18 0.19 0.23
20.8 20.4 23.4 25.5 21.1 20.8 26.1
11.9 13.4 13.2 8.5 11.3 8.0 5.9
9.2 6.8 5.7 7.8 12.9 13.6 17.8
28 28 28 28 28 28 28
65.9 65.3 62.5 70.3 58.1 66.7 68.6
4.9
2.0
2.1
28.5
1.6
29
11.3
21.5
8.6
0.77
130.8
11.7
0.18
22.6
10.3
10.6
28
65.4
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
15/8/03 16/8/03 17/8/03 18/8/03 19/8/03 20/8/03 21/8/03
n-fatty acids
a
Carbon preference index: for n-alkanes it is expressed as a summation of the odd carbon number homologues divided by a summation of the even carbon number homologues; for n-fatty acids and n-alkanols it is inverted to even-to-odd homologues. b Carbon number of the compound with the highest concentration in a homologous series. c Ratio of unresolved complex mixture to resolved n-alkanes. 1395
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
evaporated at below 250 C (OC1 in the OC/EC thermogram) in the total carbon was obviously higher in the winter samples than in the summer samples. About 30% of the total carbon was water soluble. Higher WSOC at the rural site suggested a larger contribution from secondary organic carbon due to the longer distance between the sampling site and the emission sources. 3.2. Loading of SEOC Quantified SEOC includes n-alkanes, polycyclic aromatic hydrocarbons (PAHs), n-fatty acids and n-alkanols, and molecular markers such as pentacyclic triterpanes. Also quantified was the unresolved complex mixture (UCM) in the aliphatic fraction, which was actually more abundant than the total resolved components. The total yield of resolved SEOC was 163 ng m 3 at the urban site and 178 ng m 3 at the rural site in summer, and 650 and 740 ng m 3, respectively in winter. Among the resolved components, fatty acids were the most abundant. In summer, 80% of the resolved SEOC yield was fatty acids and 70% in winter. About 5% of the resolved SEOC in summer and 8% in winter was PAHs. 3.3. Seasonal variation Distinct seasonal variations were found in the organic matter with higher concentrations in winter. The winter to summer concentration ratios for n-alkanes were 6.7 and 4.8 for urban and rural, respectively, 7.3 and 6.6 for PAHs, 3.1 and 3.4 for n-fatty acids, and 7.0 and 11.1 for n-alkanols. Many of the SEOCs were semi-volatile, contributing to the more pronounced seasonal variation when compared to elemental carbon (2.2 and 1.8 at the urban and rural sites, respectively). High mixing heights are known to lower pollutant concentrations and high mixing heights are usually caused by higher ambient temperatures in summer. Furthermore, gas/particle partitioning is also temperature dependent (Bidleman et al., 1986; Pankow and Bidleman, 1992). So the difference in ambient temperature (26–30 C in summer and 8–12 C in winter during sampling periods) should be the major cause of this seasonal variation. As a coastal city, Shanghai is affected by summer monsoons, which bring in clean oceanic winds that dilute local air pollutants. More precipitation in summer also can remove the pollutants. Low level temperature inversion was also a possible cause for the higher winter concentrations (Ye et al., 2003). 3.4. Spatial variation The expected differences in pollutant concentrations between the urban and rural sampling sites were not found. Actually, the average loading of resolved SEOC was a little higher at the rural site (Table 1). This phenomenon is the consequence of the change in land use when a developing nation modernizes or urbanizes. Due to the rapid increase
in the value of land and to the Shanghai government’s efforts to shut down polluting plants, more than 1500 factories in the city were closed down or relocated to rural areas since the mid-1990s (http://www.sh.xinhua.org/tebiebaodao/tebiebaodao/gongye.htm). Boilers in the urban area now use natural gas or they are retrofitted to burn more efficiently. In addition, new factories are being built in the countryside. About 70% of the added value of industrial output of Shanghai was produced by factories outside the city limits in 2002 (Shanghai Economy Yearbook, 2003). The change in land use has redistributed pollution sources. What was countryside once is now studded with manufacturing plants. As a result, it is difficult to distinguish between rural and urban areas in Shanghai through analysis of air quality data (http://www.sepb.gov.cn). The geography of the two sampling stations caused the urban site to be cleaner than the rural site when the winds were from the sea (east wind in summer and northeast wind in winter), and vice versa when the winds were from the land (south or southeast winds in summer and northwest wind in winter) (Table 1). 3.5. Composition of SEOC 3.5.1. Aliphatic hydrocarbons The concentrations and distributions of n-alkanes (C17– C36) are given in Table 1 and Fig. 1. Carbon preference index (CPI) has been found to be useful in distinguishing the two main categories of n-alkanes, biogenic and fossil fuel residue (Simoneit, 1986; Rogge et al., 1993; Fang et al., 1999; Tareq et al., 2005). Alkanes from biogenic sources have high CPI values (6–9 or above), while alkanes from fossil fuel residue comprise mainly of low carbon number compounds and show no carbon predominance (CPI value of unity). The low CPI values (1.4 at the urban site and 1.6 at the rural site in summer, 1.2 at both sites in winter, Table 1) in Shanghai indicated that fossil fuel residue was the main source of the n-alkanes. The concentration of alkanes from biogenic sources (plant wax) estimated using the method of Simoneit et al. (1991) showed that only about 20% of the alkanes in
25 Winter-rural Winter-urban Summer-rural Summer-urban
20
ng m-3
1396
15
10
5
0 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Carbon number
Fig. 1. Distribution diagrams of n-alkanes average concentrations.
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
summer and less than 10% in winter were from biogenic sources. The winter samples had more low carbon number alkanes. The 6C26 homologues accounted for only 20% of the total alkane in summer but 60% in winter. The carbon number maximum (Cmax) in the summer samples was C29 at both sites, while it was C23 at the urban site and C24 at the rural site in the winter samples. Similar phenomenon was also observed in other cities such as Beijing (Feng et al., 2005) and was attributed to the seasonal difference in ambient temperature. The 6C26 homologues of the alkanes are semi-volatile and will be partitioned between the gas and particle phases (Pankow and Bidleman, 1992). UCM, composed of unresolvable highly branched and cyclic aliphatic hydrocarbons, was thought to be from the incomplete combustion of fossil fuels, and has been used to evaluate the impact of anthropogenic sources (Simoneit, 1986). It is found that vehicular emissions have higher ratio of unresolved to resolved alkanes (U:R) than coal and wood combustion emissions (Kavouras et al., 2001). PM2.5 samples in Shanghai had high U:R ratios (19.7 and 11.3 at the urban and rural sites, respectively, in summer, while 7.8 and 6.2 in winter, Table 1), suggesting the petroleum residue (engine exhaust or traffic) is the main source of the aliphatic organic matter. Aerosols taken at the rural site had lower U:R ratios, indicating a lower contribution from traffic. Triterpanes (hopanes) from C27 to C35 without C28 were detected and quantified by the key ion of m/z 191 (Fig. 2). Hopanes are widely used as markers for fossil fuel residue, especially engine exhaust (Simoneit, 1986; Schauer et al., 1999b). The distribution patterns of hopanes in Shanghai were quite similar between summer and winter, with the most abundant compound being 17a(H),21b(H)-hopane, followed by 17a(H),21b(H)-norhopane. The S/(S + R) ratio at 0.6 for the isomers of 22S and 22R 17a(H), 21b(H)-homohopane suggests that these triterpanes were
Fig. 2. Distribution diagrams of triterpanes average concentrations. (Ts: 18a(H)-22,29,30-trisnorneohopane; Tm: 17a(H)-22,29,30-trisnorhopane; C29ab: 17a(H),21b(H)-norhopane; C29ba: 17b(H),21a(H)-norhopane; C30ab: 17a(H),21b(H)-hopane; C30ba: 17b(H),21a(H)-hopane; C31S: 22S-17a(H), 21b(H)-homohopane; C31R: 22R-17a(H),21b(H)-homohopane; C32S: 22S17a(H),21b(H)-bishomohopane; C32R: 22R-17a(H),21b(H)-bishomohopane).
1397
from high thermally maturated fossil fuels (or products). The distribution of triterpanes was similar with reported values from engine exhaust, but different from coal burning (Oros and Simoneit, 2000; Feng et al., 2005), suggesting that engine exhaust was the main source of these triterpanes in Shanghai. The yield of triterpanes was higher at the urban site (8.0 ng m 3 in summer and 24.2 ng m 3 in winter) than at the rural site (5.4 ng m 3 in summer and 18.7 ng m 3 in winter). The ratio of triterpane yield to non-volatile fossil fuel residue alkanes (sum of the >C26 alkanes that was not biogenic) was 0.7 at the urban site and 0.4 at the rural site, indicating that traffic emissions was more influential in the urban area. 3.5.2. PAHs Benzo[b + k]fluoranthene was the dominant in all samples, followed by benzo[ghi]perylene (BgP), indeno[1,2,3-cd]pyrene (IP), benzo[e]pyrene (BeP) and chrysene/ triphenylene (Fig. 3). One obvious difference in the PAH distributions between summer and winter was the higher low molecular weight (LMW, molecular weight 6228) PAH concentrations in winter. The sum of LMW concentrations was only about 20% of total PAHs in summer while >40% in winter. Similar to what was found on n-alkanes, this was mainly due to the gas-particle partitioning of the semi-volatile LMW at different ambient temperatures and is in agreement with other studies (Bidleman et al., 1986; Zheng et al., 2000; Feng et al., 2005). The BeP/(BeP + BaP) ratio (BaP is benzo[a]pyrene) is used as an indicator for the decay of labile BaP in the atmosphere and thus the aging of aerosols (Nielsen, 1988). Most freshly emitted aerosols have a ratio of 0.5 (Grimmer et al., 1983). The BeP/(BeP + BaP) ratio in Shanghai was >0.7 for both summer and winter (Table 1). Higher ratio in summer indicates the preferential loss
Fig. 3. Distribution diagrams of PAHs average concentrations. (Low molecular weight PAH: (1) Phenanthrene; (2) Anthracene; (3) Fluoranthene; (4) Pyrene; (5) Benzo[ghi]fluoranthene; (6) Benzo[a]anthracene; (7) Chrysene/Triphenylene. High molecular weight PAH: (8) Benzo[b + k]fluoranthene; (9) Benzo[a]fluoranthene; (10) Benzo[e]pyrene; (11) Benzo[a]pyrene; (12) Perylene; (13) Indeno[1,2,3-cd]fluoranthene; (14) Indeno [1,2,3cd]pyrene; (15) Benzo[ghi]perylene; (16) Coronene).
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
of BaP because of the higher ambient temperature and solar radiation also reported by studies in other Chinese cities (Guo et al., 2003; Feng et al., 2005). The high ratio in the winter indicated the impact of the non-local and/or aged aerosols, it was in agreement with the higher WSOC/TC ratio in winter. The IP/(IP + BgP) ratio has been found useful for source apportionment. Aerosols emitted from gasoline vehicles, diesel vehicles, coal combustion were reported to have values of 0.2, 0.37 and 0.56, respectively (Grimmer et al., 1983). In Shanghai, the IP/(IP + BgP) ratio was 0.41 at the urban site and 0.45 at the rural site in summer, and 0.44 and 0.46, respectively in winter. These numbers indicate a mixed source. The lower value at the urban site than the rural site in summer indicated a stronger presence of engine exhaust. 3.5.3. Alkanoic acids C16 and C18 saturated acids were the two largest peaks and they accounted for 50–70% of the total fatty acids (Fig. 4). The strong even carbon number predominance (CPI > 10, Table 1) suggests that the fatty acids were mainly biogenic. TheC22 homologues are from vascular plant wax (Simoneit, 1986). In Shanghai, the contribution from plant wax at the rural site (23% in summer and 24% in winter, Table 1) was slightly higher than at the urban site (16% in sum180
Summer-rural
20
Summer-urban
15
10
5
0 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Carbon number
Fig. 5. Distribution diagrams of n-alkanols average concentrations.
mer and 18% in winter). Similar distribution patterns between summer and winter indicated that the sources of fatty acids had small seasonal variations. 3.5.4. n-Alkanols Normal alkanols of C12–C32 were detected in all samples (Table 1). The distributions of the n-alkanols were similar between summer and winter though the concentrations were quite different (Fig. 5). Bi-modal distribution was found with a main peak at C30 (or C28) and a minor peak at C18 (or C16). Alkanols from vascular plant wax, homologues of >C20 (Simoneit, 1986), accounted for about 65% of the total alkanols. The different Cmax in summer (C28) and winter (C30) suggested that the alkanols were from different vegetations in two seasons.
Winter-urban Summer-rural
140
Summer-urban 120
ng m-3
Winter-urban
3.6. Source apportionment of carbonaceous components
Winter-rural 160
Winter-rural
25
ng m-3
1398
100 80 60 40 20
C32
C31
C30
C29
C28
C27
C26
C25
C24
C23
C22
C21
C20
C19
C18
C18:1
C17
C18:2
C16
C15
C14
C13
C12
0
Fig. 4. Distribution diagrams of n-fatty acids average concentrations. (C18:2 for 9,12-octadecadienoic acid; C18:1 for 9-octadecenoic acid).
The sources of the carbonaceous material in the PM2.5 of Shanghai were apportioned using the Chemical mass balance model (CMB8, Watson et al., 1998) with solvent extractable organic compounds as tracers. Source profiles for diesel engine, catalytic and non-catalytic gasoline engine exhausts, meat cooking and cooking with seed oils are from Schauer et al. (1999a,b, 2002a,b), for vegetative detritus is from Rogge et al. (1993), for Chinese kitchen emission is from He et al. (2004), for biomass combustion is from Sheesley et al. (2003), for Chinese coal combustion is from Zheng et al. (2005). Since Si and Al were not measured in this study, road dust resuspension is not included. The species included in the CMB modeling were
Table 2 Source contributions to organic carbon in PM2.5 of Shanghai in percentage Diesel and gasoline exhaust
Coal burning
Kitchen emission
Vegetative detritus
Biomass burning
Others
FDU (urban)
August November
54 50
12 15
9 8
6 4
1 4
18 19
SHO (rural)
August November
45 43
13 15
7 8
8 6
2 5
25 23
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
EC, n-alkanes of C25–C34, steranes of C27–C29, hopanes of C27–C30, PAHs (MW 252 and 276, without BaP), alkanoic acids of C14–C30, cholesterol and b-sitosterol. Considering that non-local source profiles were used, the CMB results were statistically significant with the R2 and chi-square at 0.74–0.78 and 4.1–4.5, respectively. Modeling results in Table 2 show that traffic emissions were the largest contributors to particulate OC. Coal burning emission contributed 15% of the OC. After taking the sulfur (SO2 and sulfate) into account, it was found that pollution caused by coal usage is still the most important in Shanghai. Kitchen emissions accounted for about 8% of the OC, warranting more attention. The unexplained OC (‘‘Others’’ in Table 2) was probably due to secondary organic aerosols. 4. Conclusion No clear spatial variation in SEOC concentrations was observed between the urban and rural sampling sites and this was attributed to the change in land use in Shanghai due to the rapid urbanization of the rural areas in the past 20 years. In an attempt to relieve the city of industrial pollution sources, large-scale relocation of manufacturing plants to rural areas redistributed the pollution sources. Although the air quality in the city has improved, the rural areas are now polluted. The concentrations of carbonaceous matter from the PM2.5 samples collected at a rural site and an urban site in Shanghai in 2002 winter and 2003 summer showed that air pollution is much more severe in winter. Total carbon concentration at 20 lg m 3 showed that air pollution was very severe in winter. The n-alkanes distribution showed that fossil fuel residue was their main source with not more than 20% contribution from plant wax during both seasons. The similarity in triterpane distribution with reported tunnel data and the high U:R ratio of the aliphatic fraction indicated that engine exhaust was a major source of SEOC. IP/(IP + BgP) ratios suggested a mixed source of engine exhaust and coal burning for PAHs. Acknowledgements The authors are grateful for financial support from NSFC/RGC (N_HKUST613/01) and NSFC No. 20131160731, 20177002. The authors are also grateful to Xiaofeng Huang and Yunliang Zhao for their help in collecting the samples. The authors also wish to thank Dr. Jianzhen Yu for her help in OC/EC analysis. References Bidleman, T.F., Billings, W.N., Foreman, W.T., 1986. Vapor-particle partitioning of semi-volatile organic compounds—estimates from field collections. Environ. Sci. Technol. 20, 1038–1043. Birch, M.E., 1998. Analysis of carbonaceous aerosols: interlaboratory comparison. Analyst 123, 851–857.
1399
Cao, J.J., Lee, S.C., Hoa, K.F., Zhang, X.Y., Zou, S.C., Fung, K., Chow, J.C., Watson, J.G., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period. Atmos. Environ. 37, 1451–1460. Chen, J.M., Tan, M.G., Li, Y.L., Zhang, Y.M., Lu, W.W., Tong, Y.P., Zhang, G.L., Yan Li, Y., 2005. A lead isotope record of shanghai atmospheric lead emissions in total suspended particles during the period of phasing out of leaded gasoline. Atmos. Environ. 39, 1245– 1253. Duan, F., He, K., Ma, Y., Jia, Y., Yang, F., Lei, Y., Tanaka, S., Okuta, T., 2005. Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere 60, 355–364. Fang, M., Zheng, M., Wang, F., Chim, K.S., Kot, S.C., 1999. The longrange transport of aerosols from Northern China to Hong Kong—A multi-technique study. Atmos. Environ. 33, 1803–1817. Feng, J.L., Chan, C.K., Fang, M., Hu, M., He, L.Y., Tang, X.Y., 2005. Impact of meteorology and energy structure on solvent extractable organic compounds of PM2.5 in Beijing, China. Chemosphere 61, 623– 632. Grimmer, G., Jacob, J., Naujack, K.W., 1983. Profile of the polycyclic aromatic compounds from crude oils. 3. Inventory by GC, GC/MSPAH in environmental materials. Fresen. Z. Anal. Chem. 316, 29–36. Guo, Z.G., Sheng, L.F., Feng, J.L., Fang, M., 2003. Seasonal variation of solvent extractable organic compounds in the aerosols in Qingdao, China. Atmos. Environ. 37, 1825–1834. He, L.Y., Hu, M., Huang, X.F., Yu, B.D., Zhang, Y.H., Liu, D.Q., 2004. Measurement of emissions of fine particulate organic matter from Chinese cooking. Atmos. Environ. 38, 6557–6564. Hsiao, W.L., MO, Z.Y., Fang, M., Shi, X.M., Wang, F., 2000. Cytotoxicity of PM2.5 and PM2.5–10 ambient air pollutants assessed by the MTT and the Comet assays. Mutat. Res. Gen. Toxicol. Environ. 471, 45–55. Kavouras, I.G., Koutrakis, P., Tsapakis, M., Lagoudaki, E., Stephanou, E.G., Baer, D.V., Oyola, P., 2001. Source apportionment of urban particulate aliphatic and poly-nuclear aromatic hydrocarbons (PAH) using multivariate methods. Environ. Sci. Technol. 35, 2288–2294. Nielsen, T., 1988. The decay of benzo(a)pyrene and cyclopenteno(cd)pyrene in the atmosphere. Atmos. Environ. 22, 2249–2254. Oros, D.R., Simoneit, B.R.T., 2000. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel 79, 515– 536. Pankow, J.F., Bidleman, T.F., 1992. Interdependence of the slopes and intercepts from log–log correlations of measured gas-particle partitioning of vapor pressure. I. Theory and analysis of available data. Atmos. Environ. 26A, 1071–1080. Parungo, F., Nagamoto, C., Zhou, M.Y., Hansen, A.D.A., Harris, J., 1994. Aeolian transport of aerosol black carbon from China to the ocean. Atmos. Environ. 28, 3251–3260. Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R., 1993. Sources of fine organic aerosol. 4. Particulate abrasion products from leaf surfaces of urban plants. Environ. Sci. Technol. 27, 2700–2711. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999a. Measurement of emissions from air pollution sources. 1. C1 through C29 organic compounds from meat charbroiling. Environ. Sci. Technol. 33, 1566–1577. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999b. Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environ. Sci. Technol. 33, 1578–1587. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002a. Measurement of emission from air pollution sources. 1. C–C32 organic compounds from gasoline-powered motor vehicles. Environ. Sci. Technol. 36, 1169–1180. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002b. Measurement of emissions from air pollution sources. 4. C1–C27 organic compounds from cooking with seed oils. Environ. Sci. Technol. 36, 567–575.
1400
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
Sheesley, R.J., Schauer, J.J., Chowdhury, Z., Cass, G.R., Simoneit, B.R.T., 2003. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. J. Geophys. Res. 108 (D9), 4285. doi:10.1029/2002JD002981. Shu, J., Dearing, J.A., Morse, P.A., Yu, L.Z., Yuan, N., 2001. Determining the sources of atmospheric particles in Shanghai, China, from magnetic and geochemical properties. Atmos. Environ. 35, 2615– 2625. Simoneit, B.R.T., 1986. Characterization of organic constituents in aerosols in relation to their origin and transport: A review. Int. J. Environ. Anal. Chem. 23, 207–237. Simoneit, B.R.T., Cardoso, J.N., Robinson, N., 1991. An assessment of terrestrial higher molecular weight lipid compounds in aerosol particulate matter over the south Atlantic from about 30–70 S. Chemosphere 23, 447–465. Tareq, S.M., Tanoue, E., Tsuji, H., Tanaka, N., Ohta, K., 2005. Hydrocarbon and elemental carbon signatures in a tropical wetland: biogeochemical evidence of forest fire and vegetation changes. Chemosphere 59, 1655–1665. Watson, J.G., Robinson, N.F., F ujita, E.M., Chow, J.C., Pace, T.G., Lewis, C., Coulter, T., 1998. CMB8 Applications and validation protocol for PM2.5 and VOCs. Desert Research Institute Document No. 1808.2D1, Desert Research Institute, Reno NV.
Yao, X.H., Chan, C.K., Fang, M., Cadle, S., Chan, T., Mulawa, P., He, K.B., Ye, B.M., 2002. The water-soluble ionic composition of PM2.5 in Shanghai and Beijing, China. Atmos. Environ. 36, 4223– 4234. Ye, B.M., Ji, X.L., Yang, H.Z., Yao, X.H., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2003. Concentration and chemical composition of PM2.5 in Shanghai for a 1-year period. Atmos. Environ. 37, 499–510. Zheng, J., Tan, M.G., Shibata, Y., Tanaka, A., Li, Y., Zhang, G.L., Zhang, Y.M., Shan, Z., 2004. Characteristics of lead isotope ratios and elemental concentrations in PM10 fraction of airborne particulate matter in Shanghai after the phase-out of leaded gasoline. Atmos. Environ. 38, 1191–1200. Zheng, M., Wan, T.S.M., Fang, M., Wang, F., 1997. Characterization of the non-volatile organic compounds in the aerosols of Hong Kong— identification, abundance and origin. Atmos. Environ. 31, 227– 237. Zheng, M., Fang, M., Wang, F., To, K.L., 2000. Characterization of the solvent extractable organic compounds in PM2.5 aerosols in Hong Kong. Atmos. Environ. 34, 2691–2702. Zheng, M., Salmon, L.G., Schauer, J.J., Zeng, L.M., Kiang, C.S., Zhang, Y.H., Cass, G.R., 2005. Seasonal trends in PM2.5 source contributions in Beijing, China. Atmos. Environ. 39, 3967–3976.
Characteristics of organic matter in PM2.5 in Shanghai Jialiang Feng a, Chak K. Chan a, Ming Fang b,*, Min Hu c, Lingyan He c, Xiaoyan Tang c a
c
Department of Chemical Engineering, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China b Institute for Environment and Sustainable Development, Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences, Peking University, Beijing, China Received 27 June 2005; received in revised form 5 December 2005; accepted 15 December 2005 Available online 25 January 2006
Abstract Solvent extractable organic compounds (SEOC), organic carbon, elemental carbon and water soluble organic carbon (WSOC) in PM2.5 samples collected in Shanghai, China in 2002 and 2003 were measured to determine the composition and sources of the organic matter in atmospheric aerosols. Distinct seasonal variations were detected with higher concentrations of organic matter in winter. The concentration of total carbon of about 20 lg m 3 in winter was about three times the summer value. About 30% of the total carbon was water soluble. Unresolved complex mixture (UCM) and fatty acids were the most abundant components quantified in SEOC, similar to other Chinese cities previously studied. High ratio of UCM to n-alkanes (U:R) and the composition of triterpanes indicated that engine exhaust was a major source of the airborne organic matter. Emissions from coal burning had more impact in the rural areas, according to the U:R value and PAHs composition. Chemical mass balance (CMB) modeling shows that about half of the organic carbon was from engine exhaust and about 15% was from coal burning. No clear spatial variation in the concentration of the organic matter was found between urban and rural areas. Our results showed that due to the rapid urbanization and relocation of industrial plants from urban areas to rural areas in the past 20 years, air pollution in rural areas is becoming a serious problem in Shanghai and the Yangtze River delta. 2006 Elsevier Ltd. All rights reserved. Keywords: Aerosol; Organic carbon; Water soluble organic carbon; Solvent extractable organic compound; GC–MS; China
1. Introduction Shanghai is the largest commercial and industrial city in China with a population of about 15 million. It has the largest steel mill and petrochemical complex in China. Rapid economic growth in the last two decades has caused soaring energy demand and the annual coal consumption in Shanghai increased from 20 million tons in 1989 to 45 million tons in 2004. Although the annual average concentration of total suspended particulates (TSP) has *
Corresponding author. Tel.: +852 2358 6916; fax: +852 2358 1334. E-mail address: [email protected] (M. Fang).
0045-6535/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.12.026
dropped since the mid-1990s, fine particle loading has remained high (Shanghai Environmental Protection Bureau). The annual PM10 loading in recent years (2001– 2004) is high at 100 lg m 3. Airborne lead pollution in Shanghai has been found to be mainly from the cement and metallurgy industries and coal combustion (Zheng et al., 2004; Chen et al., 2005). Only a small part of the airborne lead (20%) was from automotive emissions. Chemical mass balance (CMB) model analysis of elemental metals showed that coal combustion, construction, vehicle emissions, and steel furnaces were the main contributors of TSP (Shu et al., 2001). Researches on PM2.5 in Shanghai showed that 42% of
1394
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
the mass was from secondary aerosols of ammonium sulfate and nitrate (Ye et al., 2003) and stationary emissions (coal burning) are still believed to be the dominant sources (Yao et al., 2002). A large portion of the PM2.5 mass in Shanghai (40%) is found to be carbonaceous materials (Ye et al., 2003), but data on the detailed speciation of organic species in aerosols is not available. Solvent extractable organic compounds (SEOC) in aerosols contain useful molecular markers that have been successfully used for source apportionment (Simoneit, 1986; Zheng et al., 2000, 2005) and SEOC has been found to be toxic and can cause DNA mutation even at non-lethal dosages (Hsiao et al., 2000). Knowing the composition of the organic matter is also important to understanding the impact of continental aerosols on ocean ecology, because Shanghai is an important starting point of the outflow of continental pollutants to the East China Sea and the Pacific Ocean (Parungo et al., 1994). In this paper, we focus on a detailed study of the abundance and characteristics of SEOC in PM2.5 samples collected in Shanghai on a seasonal basis. The impact of urbanization on the composition of organic aerosols at rural sites will also be discussed. 2. Experimental 2.1. Sampling Samples were collected simultaneously at two sites. One was located on the rooftop of a five-story building, with a height of 15 m, on the campus of Fudan University (FDU). The campus is in the northeast part of the urban area. There is a viaduct with heavy traffic near the campus. The other site was at the Shanghai Observatory (SHO) located on a small hill of 100 m high called Sheshan in the southwestern countryside of Shanghai. It was once used as a background site for air pollution monitoring. There was no direct emission source near the Observatory. The distance between the two sampling site is 40 km. PM2.5 samples were collected during 21–28 November 2002 (high loading season) and 14–21 August 2003 (low loading season) using high-volume samplers and the same procedure as described in our previous publication (Feng et al., 2005). 2.2. SEOC analysis The details of the analysis of SEOC were presented in our previous publications (Zheng et al., 1997; Feng et al., 2005). Briefly, prior to extraction, an internal standard mixture consisting of eicosane-d42, octacosane-d58, phenanthrene-d10, chrysene-d12, perylene-d12 and heptadecanoic acid-d33, was spiked onto the filters. Then, the samples were ultrasonically extracted with three 100-ml aliquots of dichloromethane at room temperature. The combined extract was concentrated to a volume of 2–3 ml, filtered
and reduced further to a volume of 200–300 ll with a stream of high purity N2. The extract was reacted with 14% BF3 in methanol to esterify the free organic acids and then fractionated with a flash column of silica gel into four sub-fractions: aliphatics, PAHs, fatty acids (methyl ester) and alkanols. The total extract and the four fractions were subjected to GC–MS (Finnigan TSQ700 interfaced to a Hewlett Packard Model 5890A gas chromatograph) analyses. The MS was operated in the electron impact mode at 70 eV and the scan range is 50–550 amu. The GC was equipped with a HP-5MS capillary column (30 m · 0.25 mm i.d., film thickness 0.25 lm), with helium as carrier gas. The n-alkanol fraction was converted to trimethylsilyl derivatives by reaction with N,O-bis-(trimethylsilyl) trifluoroacetamide (BSTFA) before analysis. Prior to the GC–MS analysis, hexamethyl benzene was added to all fractions to be used as an internal standard for n-alkanols and to check the recovery of the other three fractions. Alkanes, PAHs and fatty acids were quantified using the corresponding deuterated internal standards having similar chemical characteristics and retention times. 2.3. OC, EC and WSOC measurements Organic and elemental carbon (OC and EC) concentrations of the samples were analyzed by thermal/optical method (Sunset OC/EC analyzer) with the NIOSH temperature program (Birch, 1998). Small samples punched from the high-volume filters (5.8 cm2 each) were extracted with nano-pure water and the concentration of the water soluble organic carbon (WSOC) was measured with a total organic carbon analyzer (Shimadzu TOC-5000 A). Hydrochloric acid was added to each sample before analysis to remove inorganic carbon. 3. Results and discussion 3.1. Loading of OC, EC and WSOC The average OC in November 2002 (16 lg m 3, Table 1) was comparable to the reported winter value (Ye et al., 2003), but the average of August 2003 (3.9 lg m 3, urban and 4.9 lg m 3, rural) was lower than the reported average summer value (10 lg m 3). The occurrence of several high concentration episodes could be an important cause of the higher reported summer value (Ye et al., 2003). TC concentrations of 20 lg m 3 in winter indicated that the pollution level was high in Shanghai even though TC concentrations were lower than in other Chinese cities such as Beijing and Guangzhou (Cao et al., 2003; Duan et al., 2005). The average winter OC/EC ratios (3.9 and 4.5 in urban and rural, respectively) were higher than in summer (2.2 and 2.4, respectively), which could be an effect of lower ambient temperature. The percentage of organic carbon
Table 1 Summary of yields and composition of organic matter in PM2.5 from Shanghai, China Site
FDU (urban)
Date
22/11/02 23/11/02 24/11/02 25/11/02 26/11/02 27/11/02 28/11/02
Wind OC EC WSOC n-alkanes PAHs direction (lg m 3) (lg m 3) (lg m 3) a b c Yield CPI Cmax U:R Wax% Yield BeP/ (ng m 3) (ng m 3) (BaP + BeP)
Yield CPI (ng m 3)
C18:1/ Wax% Yield CPIa Cmaxb Wax% C18 (ng m 3)
NE NE NW NW NW NW–SE SE
Average E E E E S SE SE
Average SHO (rural)
22/11/02 23/11/02 24/11/02 25/11/02 26/11/02 27/11/02 28/11/02
NE NE NW NW NW NW–SE SE
Average 15/8/03 16/8/03 17/8/03 18/8/03 19/8/03 20/8/03 21/8/03 Average
E E E E S SE SE
n-alkanol a
4.8 9.3 9.7 17.0 27.8 26.1 16.2
1.2 3.0 3.0 4.0 6.8 5.9 3.9
2.2 3.2 3.4 6.5 11.8 7.4 5.9
32.9 83.3 70.3 175.5 226.2 341.9 116.6
1.3 1.2 1.3 1.0 1.3 1.1 1.3
24 24 24 23 23 23 24
8.2 9.5 6.2 8.6 7.0 6.0 9.1
11.8 4.6 12.1 3.1 11.2 3.7 9.9
7.8 28.9 30.3 62.7 82.0 115.1 39.7
0.83 0.78 0.71 0.71 0.70 0.63 0.75
123.7 304.0 211.2 397.2 668.0 622.9 401.8
12.9 13.7 11.1 13.3 9.8 12.6 11.5
0.10 0.10 0.10 0.07 0.18 0.15 0.08
13.5 12.4 23.0 14.8 25.3 17.3 20.6
16.8 36.8 30.6 40.2 170.9 62.4 65.2
11.2 12.7 7.6 4.7 6.6 8.3 8.3
28 28 30 30 30 30 30
63.0 74.5 65.7 51.0 71.4 77.9 77.2
15.8
4.0
5.8
149.5
1.2
23
7.8
8.0
52.4
0.73
389.8
12.1
0.11
18.1
60.4
8.5
30
68.7
3.4 2.8 6.1 5.0 4.7 2.5 2.6
1.3 1.1 2.7 2.3 2.2 1.6 1.3
1.6 1.3 1.9 1.9 1.6 0.8 0.7
15.6 12.2 39.2 29.6 25.6 20.5 14.5
1.5 1.4 1.2 1.3 1.6 1.4 1.6
31 29 29 29 29 29 31
19.7 24.3 10.7 16.3 22.0 20.8 24.0
20.8 19.9 11.7 15.4 22.9 17.3 23.7
4.3 3.1 13.5 9.2 10.4 6.1 3.3
0.76 0.80 0.78 0.75 0.65 0.79 0.76
108.5 101.3 156.6 167.5 141.0 93.6 106.5
12.5 11.1 10.1 10.8 14.4 17.1 14.8
0.08 0.09 0.08 0.13 0.17 0.14 0.19
14.5 18.6 21.6 17.7 14.1 8.1 14.3
8.2 8.7 11.8 12.2 9.7 4.4 5.5
9.4 9.1 5.9 8.9 8.0 9.9 15.1
28 28 28 28 28 28 28
60.6 71.0 64.6 69.2 59.6 45.6 69.2
3.9
1.8
1.4
22.5
1.4
29
19.7
18.8
7.1
0.75
125.0
13.0
0.13
15.6
8.6
9.5
28
62.8
8.7 13.5 11.4 15.8 29.8 16.3 20.1
2.3 3.4 2.1 3.7 5.8 4.6 3.7
3.5 4.1 4.8 7.0 12.0 6.1 8.7
101.2 159.8 92.9 101.2 182.3 153.6 143.3
1.1 1.0 1.3 1.3 1.4 1.2 1.2
24 24 25 25 24 23 24
5.3 6.6 5.2 7.3 5.5 6.8 6.4
5.1 1.5 12.2 13.0 14.8 8.3 10.2
29.5 52.1 31.7 37.6 125.6 76.3 47.7
0.74 0.75 0.70 0.73 0.64 0.64 0.77
366.2 455.2 306.6 322.8 613.4 463.2 547.7
15.6 17.1 9.4 8.4 7.4 10.1 10.0
0.13 0.07 0.07 0.08 0.09 0.07 0.05
12.5 10.8 27.5 26.0 36.7 25.9 28.0
39.9 51.1 94.4 102.9 198.7 142.5 170.4
12.0 10.2 9.8 6.2 6.9 12.5 14.5
28 30 30 30 30 30 30
72.6 56.8 60.3 67.7 80.0 57.1 64.5
16.5
3.6
6.6
133.5
1.2
24
6.2
9.3
57.2
0.71
439.3
11.1
0.08
23.9
114.3
10.3
30
65.6
6.2 7.4 6.0 5.9 4.2 2.8 1.7
2.8 2.9 2.5 2.5 1.5 1.5 0.6
2.7 2.9 2.7 2.5 2.1 1.1 0.7
24.5 58.2 27.8 35.2 19.8 23.7 10.2
1.7 1.0 1.3 1.2 1.7 1.5 2.8
29 29 29 29 29 29 29
9.7 7.9 9.8 7.9 11.1 15.8 17.0
25.7 5.4 15.0 12.1 27.1 19.6 45.8
9.4 11.6 9.1 12.5 4.5 12.1 1.3
0.72 0.80 0.76 0.85 0.72 0.75 0.77
147.4 188.2 137.1 156.9 97.9 127.9 60.3
13.0 11.5 9.8 10.2 12.8 12.2 12.7
0.16 0.08 0.31 0.11 0.18 0.19 0.23
20.8 20.4 23.4 25.5 21.1 20.8 26.1
11.9 13.4 13.2 8.5 11.3 8.0 5.9
9.2 6.8 5.7 7.8 12.9 13.6 17.8
28 28 28 28 28 28 28
65.9 65.3 62.5 70.3 58.1 66.7 68.6
4.9
2.0
2.1
28.5
1.6
29
11.3
21.5
8.6
0.77
130.8
11.7
0.18
22.6
10.3
10.6
28
65.4
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
15/8/03 16/8/03 17/8/03 18/8/03 19/8/03 20/8/03 21/8/03
n-fatty acids
a
Carbon preference index: for n-alkanes it is expressed as a summation of the odd carbon number homologues divided by a summation of the even carbon number homologues; for n-fatty acids and n-alkanols it is inverted to even-to-odd homologues. b Carbon number of the compound with the highest concentration in a homologous series. c Ratio of unresolved complex mixture to resolved n-alkanes. 1395
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
evaporated at below 250 C (OC1 in the OC/EC thermogram) in the total carbon was obviously higher in the winter samples than in the summer samples. About 30% of the total carbon was water soluble. Higher WSOC at the rural site suggested a larger contribution from secondary organic carbon due to the longer distance between the sampling site and the emission sources. 3.2. Loading of SEOC Quantified SEOC includes n-alkanes, polycyclic aromatic hydrocarbons (PAHs), n-fatty acids and n-alkanols, and molecular markers such as pentacyclic triterpanes. Also quantified was the unresolved complex mixture (UCM) in the aliphatic fraction, which was actually more abundant than the total resolved components. The total yield of resolved SEOC was 163 ng m 3 at the urban site and 178 ng m 3 at the rural site in summer, and 650 and 740 ng m 3, respectively in winter. Among the resolved components, fatty acids were the most abundant. In summer, 80% of the resolved SEOC yield was fatty acids and 70% in winter. About 5% of the resolved SEOC in summer and 8% in winter was PAHs. 3.3. Seasonal variation Distinct seasonal variations were found in the organic matter with higher concentrations in winter. The winter to summer concentration ratios for n-alkanes were 6.7 and 4.8 for urban and rural, respectively, 7.3 and 6.6 for PAHs, 3.1 and 3.4 for n-fatty acids, and 7.0 and 11.1 for n-alkanols. Many of the SEOCs were semi-volatile, contributing to the more pronounced seasonal variation when compared to elemental carbon (2.2 and 1.8 at the urban and rural sites, respectively). High mixing heights are known to lower pollutant concentrations and high mixing heights are usually caused by higher ambient temperatures in summer. Furthermore, gas/particle partitioning is also temperature dependent (Bidleman et al., 1986; Pankow and Bidleman, 1992). So the difference in ambient temperature (26–30 C in summer and 8–12 C in winter during sampling periods) should be the major cause of this seasonal variation. As a coastal city, Shanghai is affected by summer monsoons, which bring in clean oceanic winds that dilute local air pollutants. More precipitation in summer also can remove the pollutants. Low level temperature inversion was also a possible cause for the higher winter concentrations (Ye et al., 2003). 3.4. Spatial variation The expected differences in pollutant concentrations between the urban and rural sampling sites were not found. Actually, the average loading of resolved SEOC was a little higher at the rural site (Table 1). This phenomenon is the consequence of the change in land use when a developing nation modernizes or urbanizes. Due to the rapid increase
in the value of land and to the Shanghai government’s efforts to shut down polluting plants, more than 1500 factories in the city were closed down or relocated to rural areas since the mid-1990s (http://www.sh.xinhua.org/tebiebaodao/tebiebaodao/gongye.htm). Boilers in the urban area now use natural gas or they are retrofitted to burn more efficiently. In addition, new factories are being built in the countryside. About 70% of the added value of industrial output of Shanghai was produced by factories outside the city limits in 2002 (Shanghai Economy Yearbook, 2003). The change in land use has redistributed pollution sources. What was countryside once is now studded with manufacturing plants. As a result, it is difficult to distinguish between rural and urban areas in Shanghai through analysis of air quality data (http://www.sepb.gov.cn). The geography of the two sampling stations caused the urban site to be cleaner than the rural site when the winds were from the sea (east wind in summer and northeast wind in winter), and vice versa when the winds were from the land (south or southeast winds in summer and northwest wind in winter) (Table 1). 3.5. Composition of SEOC 3.5.1. Aliphatic hydrocarbons The concentrations and distributions of n-alkanes (C17– C36) are given in Table 1 and Fig. 1. Carbon preference index (CPI) has been found to be useful in distinguishing the two main categories of n-alkanes, biogenic and fossil fuel residue (Simoneit, 1986; Rogge et al., 1993; Fang et al., 1999; Tareq et al., 2005). Alkanes from biogenic sources have high CPI values (6–9 or above), while alkanes from fossil fuel residue comprise mainly of low carbon number compounds and show no carbon predominance (CPI value of unity). The low CPI values (1.4 at the urban site and 1.6 at the rural site in summer, 1.2 at both sites in winter, Table 1) in Shanghai indicated that fossil fuel residue was the main source of the n-alkanes. The concentration of alkanes from biogenic sources (plant wax) estimated using the method of Simoneit et al. (1991) showed that only about 20% of the alkanes in
25 Winter-rural Winter-urban Summer-rural Summer-urban
20
ng m-3
1396
15
10
5
0 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
Carbon number
Fig. 1. Distribution diagrams of n-alkanes average concentrations.
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
summer and less than 10% in winter were from biogenic sources. The winter samples had more low carbon number alkanes. The 6C26 homologues accounted for only 20% of the total alkane in summer but 60% in winter. The carbon number maximum (Cmax) in the summer samples was C29 at both sites, while it was C23 at the urban site and C24 at the rural site in the winter samples. Similar phenomenon was also observed in other cities such as Beijing (Feng et al., 2005) and was attributed to the seasonal difference in ambient temperature. The 6C26 homologues of the alkanes are semi-volatile and will be partitioned between the gas and particle phases (Pankow and Bidleman, 1992). UCM, composed of unresolvable highly branched and cyclic aliphatic hydrocarbons, was thought to be from the incomplete combustion of fossil fuels, and has been used to evaluate the impact of anthropogenic sources (Simoneit, 1986). It is found that vehicular emissions have higher ratio of unresolved to resolved alkanes (U:R) than coal and wood combustion emissions (Kavouras et al., 2001). PM2.5 samples in Shanghai had high U:R ratios (19.7 and 11.3 at the urban and rural sites, respectively, in summer, while 7.8 and 6.2 in winter, Table 1), suggesting the petroleum residue (engine exhaust or traffic) is the main source of the aliphatic organic matter. Aerosols taken at the rural site had lower U:R ratios, indicating a lower contribution from traffic. Triterpanes (hopanes) from C27 to C35 without C28 were detected and quantified by the key ion of m/z 191 (Fig. 2). Hopanes are widely used as markers for fossil fuel residue, especially engine exhaust (Simoneit, 1986; Schauer et al., 1999b). The distribution patterns of hopanes in Shanghai were quite similar between summer and winter, with the most abundant compound being 17a(H),21b(H)-hopane, followed by 17a(H),21b(H)-norhopane. The S/(S + R) ratio at 0.6 for the isomers of 22S and 22R 17a(H), 21b(H)-homohopane suggests that these triterpanes were
Fig. 2. Distribution diagrams of triterpanes average concentrations. (Ts: 18a(H)-22,29,30-trisnorneohopane; Tm: 17a(H)-22,29,30-trisnorhopane; C29ab: 17a(H),21b(H)-norhopane; C29ba: 17b(H),21a(H)-norhopane; C30ab: 17a(H),21b(H)-hopane; C30ba: 17b(H),21a(H)-hopane; C31S: 22S-17a(H), 21b(H)-homohopane; C31R: 22R-17a(H),21b(H)-homohopane; C32S: 22S17a(H),21b(H)-bishomohopane; C32R: 22R-17a(H),21b(H)-bishomohopane).
1397
from high thermally maturated fossil fuels (or products). The distribution of triterpanes was similar with reported values from engine exhaust, but different from coal burning (Oros and Simoneit, 2000; Feng et al., 2005), suggesting that engine exhaust was the main source of these triterpanes in Shanghai. The yield of triterpanes was higher at the urban site (8.0 ng m 3 in summer and 24.2 ng m 3 in winter) than at the rural site (5.4 ng m 3 in summer and 18.7 ng m 3 in winter). The ratio of triterpane yield to non-volatile fossil fuel residue alkanes (sum of the >C26 alkanes that was not biogenic) was 0.7 at the urban site and 0.4 at the rural site, indicating that traffic emissions was more influential in the urban area. 3.5.2. PAHs Benzo[b + k]fluoranthene was the dominant in all samples, followed by benzo[ghi]perylene (BgP), indeno[1,2,3-cd]pyrene (IP), benzo[e]pyrene (BeP) and chrysene/ triphenylene (Fig. 3). One obvious difference in the PAH distributions between summer and winter was the higher low molecular weight (LMW, molecular weight 6228) PAH concentrations in winter. The sum of LMW concentrations was only about 20% of total PAHs in summer while >40% in winter. Similar to what was found on n-alkanes, this was mainly due to the gas-particle partitioning of the semi-volatile LMW at different ambient temperatures and is in agreement with other studies (Bidleman et al., 1986; Zheng et al., 2000; Feng et al., 2005). The BeP/(BeP + BaP) ratio (BaP is benzo[a]pyrene) is used as an indicator for the decay of labile BaP in the atmosphere and thus the aging of aerosols (Nielsen, 1988). Most freshly emitted aerosols have a ratio of 0.5 (Grimmer et al., 1983). The BeP/(BeP + BaP) ratio in Shanghai was >0.7 for both summer and winter (Table 1). Higher ratio in summer indicates the preferential loss
Fig. 3. Distribution diagrams of PAHs average concentrations. (Low molecular weight PAH: (1) Phenanthrene; (2) Anthracene; (3) Fluoranthene; (4) Pyrene; (5) Benzo[ghi]fluoranthene; (6) Benzo[a]anthracene; (7) Chrysene/Triphenylene. High molecular weight PAH: (8) Benzo[b + k]fluoranthene; (9) Benzo[a]fluoranthene; (10) Benzo[e]pyrene; (11) Benzo[a]pyrene; (12) Perylene; (13) Indeno[1,2,3-cd]fluoranthene; (14) Indeno [1,2,3cd]pyrene; (15) Benzo[ghi]perylene; (16) Coronene).
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
of BaP because of the higher ambient temperature and solar radiation also reported by studies in other Chinese cities (Guo et al., 2003; Feng et al., 2005). The high ratio in the winter indicated the impact of the non-local and/or aged aerosols, it was in agreement with the higher WSOC/TC ratio in winter. The IP/(IP + BgP) ratio has been found useful for source apportionment. Aerosols emitted from gasoline vehicles, diesel vehicles, coal combustion were reported to have values of 0.2, 0.37 and 0.56, respectively (Grimmer et al., 1983). In Shanghai, the IP/(IP + BgP) ratio was 0.41 at the urban site and 0.45 at the rural site in summer, and 0.44 and 0.46, respectively in winter. These numbers indicate a mixed source. The lower value at the urban site than the rural site in summer indicated a stronger presence of engine exhaust. 3.5.3. Alkanoic acids C16 and C18 saturated acids were the two largest peaks and they accounted for 50–70% of the total fatty acids (Fig. 4). The strong even carbon number predominance (CPI > 10, Table 1) suggests that the fatty acids were mainly biogenic. The
Summer-rural
20
Summer-urban
15
10
5
0 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Carbon number
Fig. 5. Distribution diagrams of n-alkanols average concentrations.
mer and 18% in winter). Similar distribution patterns between summer and winter indicated that the sources of fatty acids had small seasonal variations. 3.5.4. n-Alkanols Normal alkanols of C12–C32 were detected in all samples (Table 1). The distributions of the n-alkanols were similar between summer and winter though the concentrations were quite different (Fig. 5). Bi-modal distribution was found with a main peak at C30 (or C28) and a minor peak at C18 (or C16). Alkanols from vascular plant wax, homologues of >C20 (Simoneit, 1986), accounted for about 65% of the total alkanols. The different Cmax in summer (C28) and winter (C30) suggested that the alkanols were from different vegetations in two seasons.
Winter-urban Summer-rural
140
Summer-urban 120
ng m-3
Winter-urban
3.6. Source apportionment of carbonaceous components
Winter-rural 160
Winter-rural
25
ng m-3
1398
100 80 60 40 20
C32
C31
C30
C29
C28
C27
C26
C25
C24
C23
C22
C21
C20
C19
C18
C18:1
C17
C18:2
C16
C15
C14
C13
C12
0
Fig. 4. Distribution diagrams of n-fatty acids average concentrations. (C18:2 for 9,12-octadecadienoic acid; C18:1 for 9-octadecenoic acid).
The sources of the carbonaceous material in the PM2.5 of Shanghai were apportioned using the Chemical mass balance model (CMB8, Watson et al., 1998) with solvent extractable organic compounds as tracers. Source profiles for diesel engine, catalytic and non-catalytic gasoline engine exhausts, meat cooking and cooking with seed oils are from Schauer et al. (1999a,b, 2002a,b), for vegetative detritus is from Rogge et al. (1993), for Chinese kitchen emission is from He et al. (2004), for biomass combustion is from Sheesley et al. (2003), for Chinese coal combustion is from Zheng et al. (2005). Since Si and Al were not measured in this study, road dust resuspension is not included. The species included in the CMB modeling were
Table 2 Source contributions to organic carbon in PM2.5 of Shanghai in percentage Diesel and gasoline exhaust
Coal burning
Kitchen emission
Vegetative detritus
Biomass burning
Others
FDU (urban)
August November
54 50
12 15
9 8
6 4
1 4
18 19
SHO (rural)
August November
45 43
13 15
7 8
8 6
2 5
25 23
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
EC, n-alkanes of C25–C34, steranes of C27–C29, hopanes of C27–C30, PAHs (MW 252 and 276, without BaP), alkanoic acids of C14–C30, cholesterol and b-sitosterol. Considering that non-local source profiles were used, the CMB results were statistically significant with the R2 and chi-square at 0.74–0.78 and 4.1–4.5, respectively. Modeling results in Table 2 show that traffic emissions were the largest contributors to particulate OC. Coal burning emission contributed 15% of the OC. After taking the sulfur (SO2 and sulfate) into account, it was found that pollution caused by coal usage is still the most important in Shanghai. Kitchen emissions accounted for about 8% of the OC, warranting more attention. The unexplained OC (‘‘Others’’ in Table 2) was probably due to secondary organic aerosols. 4. Conclusion No clear spatial variation in SEOC concentrations was observed between the urban and rural sampling sites and this was attributed to the change in land use in Shanghai due to the rapid urbanization of the rural areas in the past 20 years. In an attempt to relieve the city of industrial pollution sources, large-scale relocation of manufacturing plants to rural areas redistributed the pollution sources. Although the air quality in the city has improved, the rural areas are now polluted. The concentrations of carbonaceous matter from the PM2.5 samples collected at a rural site and an urban site in Shanghai in 2002 winter and 2003 summer showed that air pollution is much more severe in winter. Total carbon concentration at 20 lg m 3 showed that air pollution was very severe in winter. The n-alkanes distribution showed that fossil fuel residue was their main source with not more than 20% contribution from plant wax during both seasons. The similarity in triterpane distribution with reported tunnel data and the high U:R ratio of the aliphatic fraction indicated that engine exhaust was a major source of SEOC. IP/(IP + BgP) ratios suggested a mixed source of engine exhaust and coal burning for PAHs. Acknowledgements The authors are grateful for financial support from NSFC/RGC (N_HKUST613/01) and NSFC No. 20131160731, 20177002. The authors are also grateful to Xiaofeng Huang and Yunliang Zhao for their help in collecting the samples. The authors also wish to thank Dr. Jianzhen Yu for her help in OC/EC analysis. References Bidleman, T.F., Billings, W.N., Foreman, W.T., 1986. Vapor-particle partitioning of semi-volatile organic compounds—estimates from field collections. Environ. Sci. Technol. 20, 1038–1043. Birch, M.E., 1998. Analysis of carbonaceous aerosols: interlaboratory comparison. Analyst 123, 851–857.
1399
Cao, J.J., Lee, S.C., Hoa, K.F., Zhang, X.Y., Zou, S.C., Fung, K., Chow, J.C., Watson, J.G., 2003. Characteristics of carbonaceous aerosol in Pearl River Delta Region, China during 2001 winter period. Atmos. Environ. 37, 1451–1460. Chen, J.M., Tan, M.G., Li, Y.L., Zhang, Y.M., Lu, W.W., Tong, Y.P., Zhang, G.L., Yan Li, Y., 2005. A lead isotope record of shanghai atmospheric lead emissions in total suspended particles during the period of phasing out of leaded gasoline. Atmos. Environ. 39, 1245– 1253. Duan, F., He, K., Ma, Y., Jia, Y., Yang, F., Lei, Y., Tanaka, S., Okuta, T., 2005. Characteristics of carbonaceous aerosols in Beijing, China. Chemosphere 60, 355–364. Fang, M., Zheng, M., Wang, F., Chim, K.S., Kot, S.C., 1999. The longrange transport of aerosols from Northern China to Hong Kong—A multi-technique study. Atmos. Environ. 33, 1803–1817. Feng, J.L., Chan, C.K., Fang, M., Hu, M., He, L.Y., Tang, X.Y., 2005. Impact of meteorology and energy structure on solvent extractable organic compounds of PM2.5 in Beijing, China. Chemosphere 61, 623– 632. Grimmer, G., Jacob, J., Naujack, K.W., 1983. Profile of the polycyclic aromatic compounds from crude oils. 3. Inventory by GC, GC/MSPAH in environmental materials. Fresen. Z. Anal. Chem. 316, 29–36. Guo, Z.G., Sheng, L.F., Feng, J.L., Fang, M., 2003. Seasonal variation of solvent extractable organic compounds in the aerosols in Qingdao, China. Atmos. Environ. 37, 1825–1834. He, L.Y., Hu, M., Huang, X.F., Yu, B.D., Zhang, Y.H., Liu, D.Q., 2004. Measurement of emissions of fine particulate organic matter from Chinese cooking. Atmos. Environ. 38, 6557–6564. Hsiao, W.L., MO, Z.Y., Fang, M., Shi, X.M., Wang, F., 2000. Cytotoxicity of PM2.5 and PM2.5–10 ambient air pollutants assessed by the MTT and the Comet assays. Mutat. Res. Gen. Toxicol. Environ. 471, 45–55. Kavouras, I.G., Koutrakis, P., Tsapakis, M., Lagoudaki, E., Stephanou, E.G., Baer, D.V., Oyola, P., 2001. Source apportionment of urban particulate aliphatic and poly-nuclear aromatic hydrocarbons (PAH) using multivariate methods. Environ. Sci. Technol. 35, 2288–2294. Nielsen, T., 1988. The decay of benzo(a)pyrene and cyclopenteno(cd)pyrene in the atmosphere. Atmos. Environ. 22, 2249–2254. Oros, D.R., Simoneit, B.R.T., 2000. Identification and emission rates of molecular tracers in coal smoke particulate matter. Fuel 79, 515– 536. Pankow, J.F., Bidleman, T.F., 1992. Interdependence of the slopes and intercepts from log–log correlations of measured gas-particle partitioning of vapor pressure. I. Theory and analysis of available data. Atmos. Environ. 26A, 1071–1080. Parungo, F., Nagamoto, C., Zhou, M.Y., Hansen, A.D.A., Harris, J., 1994. Aeolian transport of aerosol black carbon from China to the ocean. Atmos. Environ. 28, 3251–3260. Rogge, W.F., Hildemann, L.M., Mazurek, M.A., Cass, G.R., Simoneit, B.R., 1993. Sources of fine organic aerosol. 4. Particulate abrasion products from leaf surfaces of urban plants. Environ. Sci. Technol. 27, 2700–2711. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999a. Measurement of emissions from air pollution sources. 1. C1 through C29 organic compounds from meat charbroiling. Environ. Sci. Technol. 33, 1566–1577. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 1999b. Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environ. Sci. Technol. 33, 1578–1587. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002a. Measurement of emission from air pollution sources. 1. C–C32 organic compounds from gasoline-powered motor vehicles. Environ. Sci. Technol. 36, 1169–1180. Schauer, J.J., Kleeman, M.J., Cass, G.R., Simoneit, B.R.T., 2002b. Measurement of emissions from air pollution sources. 4. C1–C27 organic compounds from cooking with seed oils. Environ. Sci. Technol. 36, 567–575.
1400
J. Feng et al. / Chemosphere 64 (2006) 1393–1400
Sheesley, R.J., Schauer, J.J., Chowdhury, Z., Cass, G.R., Simoneit, B.R.T., 2003. Characterization of organic aerosols emitted from the combustion of biomass indigenous to South Asia. J. Geophys. Res. 108 (D9), 4285. doi:10.1029/2002JD002981. Shu, J., Dearing, J.A., Morse, P.A., Yu, L.Z., Yuan, N., 2001. Determining the sources of atmospheric particles in Shanghai, China, from magnetic and geochemical properties. Atmos. Environ. 35, 2615– 2625. Simoneit, B.R.T., 1986. Characterization of organic constituents in aerosols in relation to their origin and transport: A review. Int. J. Environ. Anal. Chem. 23, 207–237. Simoneit, B.R.T., Cardoso, J.N., Robinson, N., 1991. An assessment of terrestrial higher molecular weight lipid compounds in aerosol particulate matter over the south Atlantic from about 30–70 S. Chemosphere 23, 447–465. Tareq, S.M., Tanoue, E., Tsuji, H., Tanaka, N., Ohta, K., 2005. Hydrocarbon and elemental carbon signatures in a tropical wetland: biogeochemical evidence of forest fire and vegetation changes. Chemosphere 59, 1655–1665. Watson, J.G., Robinson, N.F., F ujita, E.M., Chow, J.C., Pace, T.G., Lewis, C., Coulter, T., 1998. CMB8 Applications and validation protocol for PM2.5 and VOCs. Desert Research Institute Document No. 1808.2D1, Desert Research Institute, Reno NV.
Yao, X.H., Chan, C.K., Fang, M., Cadle, S., Chan, T., Mulawa, P., He, K.B., Ye, B.M., 2002. The water-soluble ionic composition of PM2.5 in Shanghai and Beijing, China. Atmos. Environ. 36, 4223– 4234. Ye, B.M., Ji, X.L., Yang, H.Z., Yao, X.H., Chan, C.K., Cadle, S.H., Chan, T., Mulawa, P.A., 2003. Concentration and chemical composition of PM2.5 in Shanghai for a 1-year period. Atmos. Environ. 37, 499–510. Zheng, J., Tan, M.G., Shibata, Y., Tanaka, A., Li, Y., Zhang, G.L., Zhang, Y.M., Shan, Z., 2004. Characteristics of lead isotope ratios and elemental concentrations in PM10 fraction of airborne particulate matter in Shanghai after the phase-out of leaded gasoline. Atmos. Environ. 38, 1191–1200. Zheng, M., Wan, T.S.M., Fang, M., Wang, F., 1997. Characterization of the non-volatile organic compounds in the aerosols of Hong Kong— identification, abundance and origin. Atmos. Environ. 31, 227– 237. Zheng, M., Fang, M., Wang, F., To, K.L., 2000. Characterization of the solvent extractable organic compounds in PM2.5 aerosols in Hong Kong. Atmos. Environ. 34, 2691–2702. Zheng, M., Salmon, L.G., Schauer, J.J., Zeng, L.M., Kiang, C.S., Zhang, Y.H., Cass, G.R., 2005. Seasonal trends in PM2.5 source contributions in Beijing, China. Atmos. Environ. 39, 3967–3976.