1

Characteristics of Organic Matter in PM2.5 in Shanghai

Jialiang Feng1, Chak K. Chan1, Ming Fang*, 2

1Department of Chemical Engineering

2Institute for Environment and Sustainable Development

Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong, China

Min Hu, Lingyan He, Xiaoyan Tang

State Key Joint Laboratory of Environmental Simulation and Pollution Control

College of Environmental Sciences

Peking University, Beijing, China

*Corresponding author. Email: fangming@ust.hk, Tel: (852)2358 6916, fax (852)2358 1334

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 at ~20 μg 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, based on the U:R value and PAHs composition. Chemical Mass Balance (CMB)

modeling showed that about one half of the organic carbon was from engine exhaust and

slightly over 20% was from coal burning. No clear spatial variation in the concentration of

This is the Pre-Published Version

2

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.

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 ~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

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 μg 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 the mass was 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

3

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 meters, 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 meters 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 with the same procedure

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

4

et al., 1997; Feng et al., 2005). Briefly, prior to extraction, an internal standard mixture

containing 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 μl 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, polycyclic aromatic hydrocarbons (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 (30m × 0.25 mm i.d., film thickness 0.25 µm),

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

5

measured with a total organic carbon analyzer (Shimadzu TOC-5000A). 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 μg m-3, Table 1) was comparable to the reported

winter value (Ye et al., 2003), but the averages of August 2003 (3.9 μg m-3, urban and 4.9 μg

m-3, rural) were lower than the reported average summer value (~10 μg 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 μg m-3 in winter indicated that the

pollution level was high in Shanghai even though TC concentrations were lower than 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 evaporated at <250 oC (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, PAHs, n-fatty acids, 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

6

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). The difference in ambient temperature (26-30 oC in summer and 8-12 oC in winter

during our sampling periods), therefore, would 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

7

1,500 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

Figure 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 n-alkanes. The

concentration of alkanes from biogenic sources (plant wax) estimated using the method of

Simoneit et al. (1991) showed that only ~20% of the alkanes in summer and <10% in winter

8

were from biogenic sources.

The winter samples had more low carbon number alkanes. The ≤C26 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 ≤C26 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 petroleum residue (engine exhaust or traffic) is the main source of aliphatic

organic matter. Aerosols taken at the rural site had lower U:R ratios, indicating a lower

contribution from traffic.

Triterpanes (hopanes) from C27-C35 without C28 were detected and quantified by the key

ion of m/z 191 (Figure 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 17α(H),21β(H)-hopane, followed by 17α(H),21β(H)-norhopane. The

S/(S+R) ratio at ~0.6 for the isomers of 22S and 22R 17α(H),21β(H)-homohopane suggests

that these triterpanes were from high thermally maturated fossil fuels (or products). The

distribution of triterpanes was similar to reported values from engine exhaust, but different

9

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 dominant in all samples, followed by benzo[ghi]perylene

(BgP), indeno[1,2,3-cd]pyrene (IP), benzo[e]pyrene (BeP) and chrysene/triphenylene (Figure

3). One obvious difference in the PAH distributions between summer and winter was the

higher low molecular weight (LMW, molecular weight ≤228) 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 of BaP because of the higher ambient temperature and solar radiation;

this is also reported by studies in other Chinese cities (Guo et al., 2003; Feng et al., 2005).

The high ratio in the winter indicates the impact of the non-local and/or aged aerosols and is

in agreement with the higher WSOC/TC ratio in winter.

The IP/(IP+BgP) ratio has been found useful for source apportionment. Aerosols emitted

10

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 urban site had a lower value 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 (Figure 4). The strong even carbon number predominance (CPI >10,

Table 1) suggests that the fatty acids were mainly biogenic.

The <C20 homologues of fatty acids are ubiquitous and cooking is one of the important

sources of these compounds (Schauer et al., 1999a; He et al., 2004); while the >C22

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 summer 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 although the concentrations were

quite different (Figure 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

the two seasons.

11

3.6. Source apportionment of carbonaceous components

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 SEOC 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 it is from Rogge et al. (1993), for Chinese kitchen emission it is from He

et al. (2004), for biomass combustion it is from Sheesley et al. (2003), and for Chinese coal

combustion it is from Zheng et al. (2005). Since Si and Al were not measured in this study,

road dust resuspension was not included. The species included in the CMB modeling were

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 β-sitosterol. Considering that non-

local source profiles were used, the CMB results were statistically significant with 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 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 (denoted

as “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.

12

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 μg m-3 showed that air

pollution was very severe in winter.

The n-alkanes distribution showed that fossil fuel residue was the 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.

Acknowledgement

The authors are grateful for financial support from NSFC/RGC (N_HKUST613/01) and

NSFC No. 20131160731, 20177002. The authors are also grateful to Messrs 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

semivolatile 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.

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.

13

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 long-range 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/MS-PAH 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. Tox. En. 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

14

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., and 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.T., 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

15

seed oils. Environ. Sci. Technol. 36, 567-575.

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. An. Ch. 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 30-70oS. 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.,

16

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.

17

Figure 1. Distribution diagrams of n-alkanes average concentrations.

Figure 2 Distribution diagrams of triterpanes average concentrations.

(Ts: 18α(H)-22,29,30-trisnorneohopane; Tm: 17α(H)-22,29,30-trisnorhopane; C29αβ: 17α(H),21β(H)-norhopane; C29βα: 17β(H),21α(H)-norhopane; C30αβ: 17α(H),21β(H)-hopane; C30βα: 17β(H),21α(H)-hopane; C31S: 22S-17α(H),21β(H)-homohopane; C31R: 22R-17α(H),21β(H)-homohopane; C32S: 22S-17α(H),21β(H)-bishomohopane; C32R: 22R-17α(H),21β(H)-bishomohopane)

Figure 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[1,2,3-cd]pyrene; 15 Benzo[ghi]perylene; 16 Coronene.)

Figure 4. Distribution diagrams of n-fatty acids average concentrations.

(C18:2 is 9,12-octadecadienoic acid; C18:1 is 9-octadecenoic acid)

Figure 5. Distribution diagrams of n-alkanols average concentrations.

18

0

5

10

15

20

25

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36Carbon number

ng m

-3

Winter-ruralWinter-urbanSummer-ruralSummer-urban

Figure 1.

0

1

2

3

4

5

6

7

Winter-rural Winter-urban Summer-rural Summer-urban

ng m

-3

TsTmC29αβC29βαC30αβC30βαC31SC31RC32SC32R

Figure 2.

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16Compound number

ng m

-3

Winter-ruralWinter-urbanSummer-ruralSummer-urban

Figure 3.

19

0

20

40

60

80

100

120

140

160

180

C12

C13

C14

C15

C16

C17

C18

:2

C18

:1

C18

C19

C20

C21

C22

C23

C24

C25

C26

C27

C28

C29

C30

C31

C32

ng m

-3

Winter-ruralWinter-urbanSummer-ruralSummer-urban

Figure 4.

0

5

10

15

20

25

10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32Carbon number

ng m

-3

Winter-ruralWinter-urbanSummer-ruralSummer-urban

Figure 5.

20

Table 1. Summary of yields and composition of organic matter in PM2.5 from Shanghai, China n-alkanes PAHs n-fatty acids n-alkanol

Site Date Wind Direction

OC (μg m-3)

EC (μg m-3)

WSOC (μg m-3)

Yield (ng m-3) CPI* Cmax

† U:R‡ Wax% Yield (ng m-3)

BeP/ (BaP+BeP)

Yield (ng m-3) CPI* C18:1/C18 Wax% Yield

(ng m-3) CPI* Cmax† Wax%

FDU 22/11/02 NE 4.8 1.2 2.2 32.9 1.3 24 8.2 11.8 7.8 0.83 123.7 12.9 0.10 13.5 16.8 11.2 28 63.0 (Urban) 23/11/02 NE 9.3 3.0 3.2 83.3 1.2 24 9.5 4.6 28.9 0.78 304.0 13.7 0.10 12.4 36.8 12.7 28 74.5

24/11/02 NW 9.7 3.0 3.4 70.3 1.3 24 6.2 12.1 30.3 0.71 211.2 11.1 0.10 23.0 30.6 7.6 30 65.7 25/11/02 NW 17.0 4.0 6.5 175.5 1.0 23 8.6 3.1 62.7 0.71 397.2 13.3 0.07 14.8 40.2 4.7 30 51.0 26/11/02 NW 27.8 6.8 11.8 226.2 1.3 23 7.0 11.2 82.0 0.70 668.0 9.8 0.18 25.3 170.9 6.6 30 71.4 27/11/02 NW-SE 26.1 5.9 7.4 341.9 1.1 23 6.0 3.7 115.1 0.63 622.9 12.6 0.15 17.3 62.4 8.3 30 77.9 28/11/02 SE 16.2 3.9 5.9 116.6 1.3 24 9.1 9.9 39.7 0.75 401.8 11.5 0.08 20.6 65.2 8.3 30 77.2 Average 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 15/8/03 E 3.4 1.3 1.6 15.6 1.5 31 19.7 20.8 4.3 0.76 108.5 12.5 0.08 14.5 8.2 9.4 28 60.6 16/8/03 E 2.8 1.1 1.3 12.2 1.4 29 24.3 19.9 3.1 0.80 101.3 11.1 0.09 18.6 8.7 9.1 28 71.0 17/8/03 E 6.1 2.7 1.9 39.2 1.2 29 10.7 11.7 13.5 0.78 156.6 10.1 0.08 21.6 11.8 5.9 28 64.6 18/8/03 E 5.0 2.3 1.9 29.6 1.3 29 16.3 15.4 9.2 0.75 167.5 10.8 0.13 17.7 12.2 8.9 28 69.2 19/8/03 S 4.7 2.2 1.6 25.6 1.6 29 22.0 22.9 10.4 0.65 141.0 14.4 0.17 14.1 9.7 8.0 28 59.6 20/8/03 SE 2.5 1.6 0.8 20.5 1.4 29 20.8 17.3 6.1 0.79 93.6 17.1 0.14 8.1 4.4 9.9 28 45.6 21/8/03 SE 2.6 1.3 0.7 14.5 1.6 31 24.0 23.7 3.3 0.76 106.5 14.8 0.19 14.3 5.5 15.1 28 69.2 Average 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

SHO 22/11/02 NE 8.7 2.3 3.5 101.2 1.1 24 5.3 5.1 29.5 0.74 366.2 15.6 0.13 12.5 39.9 12.0 28 72.6 (Rural) 23/11/02 NE 13.5 3.4 4.1 159.8 1.0 24 6.6 1.5 52.1 0.75 455.2 17.1 0.07 10.8 51.1 10.2 30 56.8

24/11/02 NW 11.4 2.1 4.8 92.9 1.3 25 5.2 12.2 31.7 0.70 306.6 9.4 0.07 27.5 94.4 9.8 30 60.3 25/11/02 NW 15.8 3.7 7.0 101.2 1.3 25 7.3 13.0 37.6 0.73 322.8 8.4 0.08 26.0 102.9 6.2 30 67.7 26/11/02 NW 29.8 5.8 12.0 182.3 1.4 24 5.5 14.8 125.6 0.64 613.4 7.4 0.09 36.7 198.7 6.9 30 80.0 27/11/02 NW-SE 16.3 4.6 6.1 153.6 1.2 23 6.8 8.3 76.3 0.64 463.2 10.1 0.07 25.9 142.5 12.5 30 57.1 28/11/02 SE 20.1 3.7 8.7 143.3 1.2 24 6.4 10.2 47.7 0.77 547.7 10.0 0.05 28.0 170.4 14.5 30 64.5 Average 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 15/8/03 E 6.2 2.8 2.7 24.5 1.7 29 9.7 25.7 9.4 0.72 147.4 13.0 0.16 20.8 11.9 9.2 28 65.9 16/8/03 E 7.4 2.9 2.9 58.2 1.0 29 7.9 5.4 11.6 0.80 188.2 11.5 0.08 20.4 13.4 6.8 28 65.3 17/8/03 E 6.0 2.5 2.7 27.8 1.3 29 9.8 15.0 9.1 0.76 137.1 9.8 0.31 23.4 13.2 5.7 28 62.5 18/8/03 E 5.9 2.5 2.5 35.2 1.2 29 7.9 12.1 12.5 0.85 156.9 10.2 0.11 25.5 8.5 7.8 28 70.3 19/8/03 S 4.2 1.5 2.1 19.8 1.7 29 11.1 27.1 4.5 0.72 97.9 12.8 0.18 21.1 11.3 12.9 28 58.1 20/8/03 SE 2.8 1.5 1.1 23.7 1.5 29 15.8 19.6 12.1 0.75 127.9 12.2 0.19 20.8 8.0 13.6 28 66.7 21/8/03 SE 1.7 0.6 0.7 10.2 2.8 29 17.0 45.8 1.3 0.77 60.3 12.7 0.23 26.1 5.9 17.8 28 68.6 Average 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

* Carbon Preference Index: for n-alkanes it is expressed as the summation of odd carbon number homologues divided by the summation of even carbon number homologues; for n-fatty acids and n-alkanols it is the reverse. † Carbon number of the compound with the highest concentration in a homologous series. ‡ Ratio of unresolved complex mixture to resolved n-alkanes.

21

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)

Aug. 54% 12% 9% 6% 1% 18% Nov. 50% 15% 8% 4% 4% 19%

SHO (Rural)

Aug. 45% 13% 7% 8% 2% 25% Nov. 43% 15% 8% 6% 5% 23%