Source Apportionment of Ambient Total Suspended Particulates and Coarse Particulate Matter in Urban Areas of Jiaozuo, China

By Feng, Yinchang; Xue, Yonghua; Chen, Xiaohua; Wu, Jianhui; Et al

ABSTRACT

Approximately 750 total suspended particulates (TSPs) and coarse particulate matter (PM^sub 10^) filter samples from six urban sites and a background site and >210 source samples were collected in Jiaozuo City during January 2002 to April 2003. They were analyzed for mass and abundances of 25 chemical components. Seven contributive sources were identified, and their contributions to ambient TSP/PM^sub 10^ levels at the seven sites in three seasons (spring, summer, and winter days) and a “whole” year were estimated by a chemical mass balance (CMB) receptor model. The spatial TSP average was high in spring and winter days at a level of approximately 530 g/m^sup 3^ and low in summer days at 456 g/m^sup 3^; however, the spatial PM^sub 10^ average exhibited little variation at a level of approximately 325 g/m^sup 3^, and PM^sub 10^- to-TSP ratios ranged from 0.58 to 0.81, which suggested heavy particulate matter pollution existing in the urban areas. Apportionment results indicated that geological material was the largest contributor to ambient TSP/PM^sub 10^ concentrations, followed by dust emissions from construction activities, coal combustion, secondary aerosols, vehicle movement, and other industrial sources. In addition, paved road dust and re-entrained dust were also apportioned to the seven source types and found soil, coal combustion, and construction dust to be the major contributors.

INTRODUCTION

Airborne particulate matter (PM) pollution has been one of the public concerns of ambient air quality across China as a result of rapid economic growth in recent decade. Approximately two thirds of approximately 340 cities in the country did not meet the China State Environmental Protection Administration (SEPA) National Ambient Air Quality Standards (NAAQS) limit for total suspended particulate (TSP) or coarse PM (PM^sub 10^) in last 5 yr^sup 1-5^ (100 g/m^sup 3^ for PM^sub 10^, 200 g/m^sup 3^ for TSP, annual average).6 As one of the three common air pollutants monitored nationwide, PM^sub 10^, or even TSP, is and will still be the principal cause for particle pollution in nonattainment areas in China, especially medium- and small-sized cities, although a few studies have begun to focus on fine PM (PM^sub 2.5^; PM≤2.5 m in aerodynamic size) or fine particle pollution in Chinese megacities like Beijing7 and Shanghai. 8 PM^sub 10^ attainment in urban areas requires strictly implementing PM emission control plans or strategies based on survey, monitoring, and scientific research program, such as study on airborne PM source apportionment. This paper exhibited some findings from the PM source apportionment study conducted in Jiaozuo City during 2002-2003.

Jiaozuo City, a medium-sized industrial city located in Central China, had heavy air pollution, mainly caused by TSP and PM^sub 10^. Routine monitoring data from the local environmental authority, Jiaozuo Environmental Protection Bureau (JEPB), showed that the average concentration of TSP reached 452 g/m^sup 3^ in urban areas during 1999-2003, exceeding the Chinese NAAQS limit by 125%. Ambient daily PM^sub 10^ concentration data available from 2003 suggested that PM^sub 10^ was the most prominent air pollutant in 267 days of year 2003. Furthermore, Jiaozuo was ranked by SEPA as “China’s Air- Polluted Cities Top 10″ of year 2004, according to an annual official report from SEPA.9 To identify the major contributing sources and to estimate their contributions to ambient air, this study conducted three intensive ambient TSP/PM^sub 10^ sampling campaigns at seven sites in three periods during January 2002 to March 2003 and collected a variety of potential PM sources in and around the urban areas of Jiaozuo City. Nineteen elemental species, four water-soluble ions, total carbon (TC), and organic carbon (OC) were determined for their abundances. Ambient TSP/PM^sub 10^ at the seven sites and source chemical properties were characterized, and potential major contributors were identified based on local environmental background information and similar experiences from other Chinese cities.10-12 Source contributions to ambient PM^sub 10^ and TSP were estimated using a chemical mass balance receptor model named NKCMB.

(ProQuest-CSA LLC: … denotes formulae omitted.)

EXPERIMENTAL WORK

Study Area Description

Jiaozuo City is situated at Henan Province in Central China (Figure 1, a and b) with 0.8 million inhabitants residing in its urban areas of 69 km^sup 2^. Figure 1c illustrates the detailed terrain where the urban areas are located: the city lies on Northern Henan Plain with the Taihang Mountain Range wandering in the north. Prevailing wind directions are southwest (28%) and northeast (27%), wind speed is high in spring (3.3 m/sec) and low in summer and winter (2.7 m/sec), and approximately 60% of the year is calm (wind speed <0.5 m/sec) or light air (wind speed varying from 0.5 m/sec to 1.5 m/sec) days. Temperature inversion occurs frequently. Such a local topography and meteorology make it difficult for local air pollutants to disperse or move out of this area.

As a newly developing base of chemical, metallurgical, and energy industries, Jiaozuo City has been suffering an extensive industrialization process, with yearly emissions of ~5.8 10^sup 4^ t of coal fly ash, ~5.9 10^sup 4^ t of industrial dusts, ~6.8 10^sup 4^ t of SO^sub 2^, and ~2 10^sup 4^ t of NO^sub 2^ in urban areas. Based on local statistic yearbooks and a pollution source database from JEPB, the major industrial activities potentially impacting on the urban ambient air quality can be summarized as follows. Electricity generation and heating include 11 coal-fired power plants or thermal power plants with a total capacity of more than ~2600 MW and <500 coal-fired industrial boilers run with an inadequate capacity of removing coal fly ash and sulfur dioxide. They consumed more than 4 million t of coal in 2004, releasing ~5 10^sup 4^ t of coal fly ash and <4 10^sup 4^ t of sulfur dioxide to the ambient air. Second, construction material production includes <10 cement factories (~3.5 10^sup 6^ t/yr), 4 ceramic factories (~7.5 10^sup 6^ pieces/yr), and several refractory material (refers to heatresistant materials and products, such as carbide, alumina cement, fiberboard, fire clay, bricks, ceramic fiber insulation, etc.) factories (~1 10^sup 5^ t/yr), and so forth, with annually burned 3 10^sup 5^ t of coal and ~2.8 10^sup 4^ t of dust emitted. Third, regarding chemical production, the chemical industry is a key industry in this area, which produced in 2003 >1 10^sup 5^ t of H^sub 2^SO^sub 4^, 2.5 10^sup 5^ t of NH^sub 3^, 1.2 10^sup 5^ t of NaOH, 3.2 10^sup 5^ t of Na^sub 2^CO^sub 3^, 4 10^sup 4^ t of calcium carbide, ~3 10^sup 6^ pieces of tires, ~2 10^sup 5^ t of polyvinyl chloride, and so on. During the process, air pollutants could be emitted in a considerable volume. Fourth, for the metallurgic industry, during the process of steel making (4.2 10^sup 5^ t/yr), electrolytic aluminum (3 7times; 10^sup 5^ t/yr), AL^sub 2^O^sub 3^ (1.5 10^sup 6^ t/yr), ferrous metal product (1.8 10^sup 4^ t/yr), ZrO (1.25 10^sup 5^ t/yr), TiO^sub 2^ (6.8 104 t/ yr), ZnO (6 10^sup 3^ t/yr), and other air pollutants in significant amounts might also be released to ambient air.

Fugitive dust was another important PM origin to urban ambient air. Huge areas of agricultural land, mainly in the south, bared hills because of coal mining and quarrying in the north, with sandy riverbeds and vacant lots in urban areas become huge reservoirs of PM. Anthropogenic activities, such as construction activity; mining and quarrying of coal, limestone, and other mineral resources; vehicle movement on paved and unpaved roads; and loading and unloading of coal, raw materials, fly ash, and solid wastes, inject dust in a huge amount into ambient air. In addition, vehicular exhaust becomes a new source of importance as a result of a large vehicular population (6 105 vehicles and motorcycles).

Sampling Procedures

Ambient Sampling. Ambient TSP/PM^sub 10^ sampling campaigns were conducted at seven sites listed in Figure 1, b and c, and Table 1. Sites LUB, PF, LTOB, MH, EMC, and RS (see descriptions in Table 1) were located in urban areas, and site MSM was a background site settled at mountain areas ~1 km north of the urban area. Site LUB was located at the downtown area, of which the south contained a busy arterial traffic road and the north contained residential buildings. Site PF was in a factory without PM emission at the urban center, around which were almost residential buildings except for a coal-fired power plant (Jiaozuo Power Plant) with a 1320-MW capacity. Two 210-m stacks are ~2 km west of site PF. Site LTOB is located in the western part of urban areas, and on the south is a busy traffic road, a coal-fired thermal power plant is ~1 km north of it, and Jiaozuo Power Plant is ~3 km south\east of it. Site MH is on the corner of a crossroad in the eastern part of the urban areas, around which are residential or commercial buildings. Site EMC is also on the corner of a crossroad, around which are residential buildings, with industrial chimneys easily observed. Site RS is in the northern part of the urban areas, around which there are no industrial emitters, because this area is a new developing area that used to be agricultural land, and no heavy polluters moved in.

Four precalibrated midvolume PM samplers (type KC6120, Laoshan Electronic Instrument Manufacturer, China) conducted sample collection on each site. Two samplers with PM^sub 10^ inlets collected ambient PM^sub 10^, one using polypropylene filters (90 mm in diameter, Beijing Synthetic Fiber Research Institute, China) for the determination of element species and the other fitted with quartz-fiber filters (90mm in diameter, type 2500QAT-UP, Pall Life Sciences) for the measurement of water-soluble ions and carbon components. The other two samplers with TSP inlets collected ambient TSP, similar to PM^sub 10^ sampling. Each single ambient sample was collected at a flow rate of 100 L min^sup -1^ in an 18-hr period. A total of ~750 ambient TSP/PM^sub 10^ filter samples were obtained during the following three periods: winter days of January 23 to February 1, 2002; summer days of June 11-20, 2002, and spring days of March 29 to April 9, 2003. All of the samples were kept in silicagel desiccators for moisture equilibrium before any analysis.

Source Sampling. In total, ~220 samples were collected from >15 potential sources in and around the urban areas (see Table 2 in detail). Soil samples were taken from croplands, bared hills, and riverbeds at 36 locations where topsoil (between ground surface and 1-2 cm below) and deep soil (at the depth of 10-15 cm under ground surface) samples were collected by grab. Re-entrained dust was defined as PM deposited and accumulated on the surfaces of windowsills, flat roofs, balustrades, and so forth, 5-12 m high above ground, around which there were no obvious PM emitters. Thirty- three locations uniformly distributed in the urban areas were selected for re-entrained dust sample collection. Road dust was swept from representative portions of major paved road surfaces or parking lots at 25 locations with a plastic brush and dustpan. Coal fly ash samples were collected from fly ash storage piles, dust precipitators, and briquette-burning residuals (residential coal combustion) by grab sampling. Diluted stack samples were acquired from two coal-fired power plants: Jiaozuo Power Plant with six 220- MW units and two 210-m stacks and Yanma Power Plant with a 25-MW unit, two 5-MW units, and an 80-m stack, by a dilution sampling system with a typical dilution ratio of ~10:1, like those reported by Hildemann et al.13 and Yu.14 Raw material, clinker, and final product from a cement production line in Jianxing Cement Factory were acquired as cement samples. Samples from other industrial sources such as steel-making, limekiln, caustic soda, graphite product, cryolite, calcium carbide, and refractory material and from fugitive sources like coal gangue (a kind of solid waste during coal mining process) piles, coal piles, quarrying, and so forth were also taken by grab. Source powder/bulk samples, each weighing between 0.5 kg and 3 kg, were stored in labeled polyethylene bags for further analysis.

Soil, road dust, re-entrained dust, and industrial source samples were sieved through a 150-mesh screen (100-m geometric diameter) for TSP size fractions, suspended in a resuspension chamber and sampled through PM^sub 10^ inlets onto polypropylene and quartz-fiber filters. 15,16 An ~4-g TSP size fraction of each source sample was put into a press die or crushing container with an inner diameter of 30 mm and then compressed into a coherent pellet at ~5 t cm^sup -2^ for 2-3 min in a hydraulic press.17,18 The resuspended samples and pressed powder pellets were also stored in silicagel desiccators for moisture equilibrium before any analysis.

Chemical Analysis Procedures

All of the ambient and source samples were measured for mass by a sensitive microbalance (type Metler M5, Switzerland), for 19 elemental species by a wavelength dispersive X-ray fluorescence (WDXRF) spectrometer, TC and OC by an elemental analyzer, and 4 water-soluble ions by ion chromatography (IC). The polypropylene filters (including ambient and resuspended source samples) and pressed powder pellets of source samples were analyzed by a WDXRF spectrometer (type 3080E2, Rigaku Inc., Japan) for the abundances of Na, Mg, Al, Si, P, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Br, Ba, and Pb.18,19 A set of standard reference materials were used for the quantitative analysis, including Sediments Standard Series (GSD1- GSD12), Soil Standard Series (GSS1-GSS8), Rock Standard Series (GSR1- GSR12), coal fly ash (82201), cement (GBW 03201), and so forth. The average standard deviations ranged from 0.30% to 5.88% for the elemental species investigated.18

The ion fractions were determined following the procedure below. One quarter of each TSP/PM^sub 10^ quartz-fiber filter was cut into slices and introduced into a centrifuge tube. Ten milliliters of 18 MΩ deionized water from a water purification system (Millipore) was added into each tube. These tubes were vibrated in an ultrasonic cleaner (type AS3120A, AutoScience Inc) at a frequency of 40 kHz for 25 min and then centrifuged at 5000 rpm for 10 min. The supernatant solutions were decanted into clean tubes for IC determination. Another 10 mL of 18 MΩ deionized water was added to each tube and treated again as described above. For each sample, extraction was performed three times so as to get water-soluble ions in the particles extracted adequately into the solution,20 and the extracts were thoroughly mixed. One milliliter of solution was drawn into a syringe, filtrated by a 13-mm cellulose syringe filter (Xibosh) with pore size <0.22 m, and injected into the IC (Model DX120, Dionex) for the concentrations of Cl^sup -^, NO^sup -^^sub 3^ , SO^sup 2- ^^sub 4^, and NH^sup +^^sub 4^.20-23 Approximately 0.1-g TSP size fraction of each source sample was also analyzed for Cl^sup -^, NO^sup -^^sub 3^, SO^sup 2-^^sub 4^, and NH^sup +^^sub 4^, similar to the procedure described above.

Punches (15 mm in diameter) from quartz-fiber filters were taken for analysis of OC and TC by an elemental analyzer (type Vario EL, Elementar Analysensysteme GmbH). TC fraction in a punch was first oxidized into CO^sub 2^ in an oxidizing atmosphere (helium with 8% oxygen) at a temperature of 980 C for 90 sec, then reduced at 600 C, and finally TC content in the punch was quantified through the detection of CO^sub 2^ by a thermal conductivity detector.24-27 OC content was also detected following a procedure similar to TC analysis, and the only difference is that the oxidizing temperature was 450 C instead. The precision of the method in terms of the average standard deviation in measurement of TC and OC is 0.40% and 0.68%, respectively.

RESULTS AND DISCUSSION

Ambient PM Concentration Data

Tables 3 and 4 summarize the concentrations of ambient TSP/ PM^sub 10^ and their chemical constitutes found at the seven receptor sites. Ambient PM levels in the urban areas, TSP or PM^sub 10^, were obviously too heavy during this study, both temporally and spatially. Annual TSP concentrations exceeded the mandatory limit for TSP at sites LUB, PF, MH, and EMC by a factor of ~1.75, by ~1.25 at site LTOB and RS, and by 0.61 at site MSM, whereas PM^sub 10^ was ~2.5 times higher than the mandatory annual limit for PM^sub 10^ at sites LUB, PF, LTOB, MH, and EMC, 1.9 times higher at site RS, and 1.5 times higher at site MSM. Annual TSP average and PM^sub 10^ levels reached their highest at site MH and at site LTOB, respectively, and their lowest both at site MSM, the background site. The ratios of PM^sub 10^ to TSP at the seven sites ranged from 0.58 to 0.81, similar to the PM levels in the cities adjacent to Jiaozuo, such as Anyang City, Luoyang City, and Kaifeng City, with PM^sub 10^-to-TSP ratios of 0.71, 0.77, and 0.64 in year 2002, respectively.28

Spring days observed the highest TSP and PM^sub 10^ seasonal concentration averages at site LUB and site LTOB, respectively; summer days found at site EMC and site MH, respectively; and winter days recorded at site PF and site MH, respectively. TSP spatial concentration averages (averages of PM concentrations at the seven sites during a sampling period) were lower in summer days at a level of 456 g/m^sup 3^ and higher in spring and winter days with almost equal values (532 g/m^sup 3^ in spring days and 528 g/m^sup 3^ in winter days). However, spatial averages of PM^sub 10^ changed little with seasons, 325 g/m^sup 3^ in spring days, 324 g/m^sup 3^ in summer days, and 328 g/m^sup 3^ in winter days.

Determined chemical species other than OC accounted for ~50% of the total mass in the ambient TSP samples, ranging between 43% and 62%, and accounted for ~65% in the ambient PM^sub 10^ samples, ranging from 53% to 81%. Chemical species related to crustal material (Al, Si, K, Ca, and Fe), carbon (TC and OC), and ions (NO^sup -^^sub 3^ and SO^sup 2-^^sub 4^) were the most abundant constituents in ambient PM samples, accounting for >90% of the mass of determined species. These findings are similar to many previous studies, for example, Al, Si, Ca, and Fe accounted for the majority of ambient PM^sub 10^ sample mass in Tianjin10; Na, Mg, Al, Si, Ca, and Fe were found the most abundant in ambient TSP samples collected in Qinhuangdao City (a coastal city in North China)11; Ca, Si, Fe, and S were the major constituents of TSP in the lignite-burning area of Western Macedonia, Greece29; high abundances of SiO^sub 2^, AL^sub 2^O^sub 3^, Ca, Fe, Cl^sup -^, NO^sup -^^sub 3^, and SO^sup 2- ^^sub 4^were \observed in TSP samples in Huelva, Spain30; Al, Si, Fe, OC, NO^sup -^^sub 3^, and SO^sup 2-^^sub 4^ had high abundances in ambient PM^sub 10^ at 9 of the 10 sites in the SJVAQS/AUSPEX (San Joaquin Valley Air Quality Study/Atmospheric Utilities Signatures, Predictions, and Experiments program) study31 and in Mexicali and California’s Imperial Valley.32 However, it is different from those reported by Chio et al.26 in that only high abundances of OC, elemental carbon, NO^sup -^^sub 3^, and SO^sup 2-^^sub 4^ were observed in Taichung urban and coastal aerosols in Taiwan.

Very high K abundance was observed in ambient PM, with 1.8-4.2% in TSP and 3.4-7.2% in PM^sub 10^, which are much higher than those in many previous studies.33 Biomass burning, like burning of wheat or corn straw from croplands nearby after harvest time, was a suspected contributor to such high K concentrations in ambient air, as researchers reported. Zinc abundance in ambient PM^sub 10^ at site EMC was 35.7-79.5 times in spring days and 1.4-10.5 times in summer days, higher than those at the rest sites. Zinc abundance in ambient TSP at site EMC (0.18% in spring and 0.66% in summer) was 2.8-21.1 times in spring days and 5.9-23.4 times in summer days, higher than those at the rest six sites. However, no significant differences between zinc abundances at seven sites were observed in winter days. Pollution source database from JEPB recorded that there was a zinc oxide factory ~1.9 km west of site EMC, which could partly account for zinc spatial variability. Local meteorological condition also helped explain why zinc varied spatially. Figure 2 shows wind roses during the sampling campaigns. West-northwest and southwest-west were the prevailing wind directions during spring and summer campaigns, which helped pollutants from the zinc oxide factory disperse to site EMC, whereas wind blew mainly from the northeast during the winter sampling campaign.

Source Profiles

Figures 3 and 4 illustrate averaged source profiles representing the distributions of chemical abundances and their variabilities (minimum-maximum), including TSP and PM^sub 10^ size fractions from potential sources like geological material, coal fly ash, construction dust, industrial sources, and others.

Geological Sources. Three soil subtypes had been selected for geological or crustal sources during this study, including agricultural soil (JZBN and JZSN), bared hill soil (JZBS and JZSS), and riverbed soil (JZBH and JZSH). All three subtypes of soil samples were designed into two categories: topsoil (JZBN, JZBS, and JZBH) and deep soil (JZSN, JZSS and JZSH) (see Table 2 for descriptions). Mg, Al, Si, Ca, and Fe are abundant constitutes in all of the soil subtypes, together accounting for ~40% in TSP and PM^sub 10^ size fractions, similar to those reported profiles in many previous studies.30,33-36 Al has similar abundances in three soil subtypes (4-5% for TSP and ~6.5% for PM^sub 10^), and so it is with Mg (1.5-2.2%) and Si (6-9.7%). However, Fe in riverbed soil is approximately two times as abundant as that in agricultural and bared hill soil. Ca and TC in bared hill soil (~9% and 7%, respectively) are higher than those in riverbed and agricultural soils (4.5-7% and 3-4.5%, respectively). Substantial variations were observed in the ratios of OC to TC for soils in this study, ranging between 0.18 and 0.48. OC-to-TC ratios are much lower compared with other studies, such as 0.97-0.99 in soil from the Imperial and Mexicali Valleys along the U.S./Mexico border,34 0.91 in geological source from Qalabotjha City in South Africa,37 and >0.9 in Hong Kong soil.36

All three of the subtypes of soil samples were designed into two categories: topsoil (JZBN, JZBS, and JZBH) and deep soil (JZSN, JZSS, and JZSH), provided that topsoil could be significantly impacted by emitters of other source types. The two categories were compared to evaluate whether topsoil samples had been significantly impacted by emitters of other source types and whether they could chemically and physically characterize the soil subtypes so that representative profiles for geological material can be obtained. Comparisons between topsoil and deep soil profiles were conducted in two ways. In one way, log-log scatter plots of the chemical abundances in a topsoil profile against those in a deep soil profile of the same soil subtype are used (see Figure 5). In each plot, the closer the point is to the diagonal line, the closer to each other are two profiles in the abundances of the species that the point represents. The more points that are on or close to the diagonal line, the more similar to each other the two profiles are. Figure 5, a-c, shows JZBS versus JZSS, JZBN versus JZSN, and JZBH versus JZSH for TSP size fraction, respectively, and demonstrate that no significant difference in chemical composition exists between topsoil and deep soil of the same subtype. A similar conclusion can be inferred for the two PM^sub 10^ categories of the same soil subtype from Figure 5, e-g.

In another way, to quantify the similarity between the two categories of a soil subtype, the coefficient of divergence (CD), a self-normalizing parameter used to measure the spread of the data points for two datasets, is used. The CD is defined as follows:

… (1)

where x^sub Ai^, x^sub Bi^ represent the abundance for chemical species i in profile (or source) A and profile (or source) B, respectively, A and B represent two profiles for comparative analysis, and p is the number of investigated chemical species. If CD^sub AB^ approaches zero, source A and B are similar, and if it approaches one, they are significantly different.38,39 In this study, for TSP size fractions, all of the species are investigated except NH^sup +^^sub 4^ and NO^sup -^^sub 3^ because some of their amounts are lower than the method detection limits. The CDs are very close to zero, with 0.177, 0081 and 0.155 for CD^sub JZBS-JZSS^, CD^sub JZBN-JZSN^, and CD^sub JZBH-JZSH^, respectively. For PM^sub 10^ size fractions, all of the species are included except NH^sup +^^sub 4^, and CD^sub JZBS-JZSS^, CD^sub JZBN-JZSN^, and CD^sub JZBH- JZSH^ approach zero with values of 0.290, 0.110, and 0.174. From the CD values, it is also concluded that no significant difference in chemical composition exists between topsoil and deep soil of the same subtype. So comparisons in both ways indicated that topsoil samples did not seem to have been polluted by other PM sources, and both topsoil and deep soil samples still kept chemical characteristics of the soil source type.

Particle size-mass distribution is often used to characterize particles physically. Figure 6 shows the particle size-mass distributions of the three soil subtypes in two categories, for which the mass data was acquired by classifying sieved soil samples (through 150-mesh sieves) into nine aerodynamic size intervals by a dust centrifugal classifier (type YFJ, Chengde Instrument Manufacturer, China). 40 JZBS and JZBN and JZSS and JZSN are very similar in mass-size distribution, respectively. JZBS and JZBN have much higher mass fractions of 6.2-13.3 m than JZSS and JZSN but lower mass fractions when >96.6 m size interval is considered. Riverbed soil samples JZBH and JZSH are quite different from each other and other soil subtypes, with less smaller particles and more larger particles.

According to above analysis of chemical and physical properties, topsoil profiles were chosen to represent soil subtypes, partly because topsoil is a direct emitter to ambient PM. JZBN and JZSN in topsoil category are combined into a composite profile as geological material for input data to source apportionment modeling, because riverbed soil physically differs from agricultural soil and bared hill soil and covers much less area than agricultural soil and bared hill soil do.

Coal Combustion. Coal combustion as a result of electricity generation, heating, water vapor providing, and residential use acts as an important origin of ambient PM. Three composite profiles are constructed for TSP fraction: JZDC for coal-fired power plant stack exhaust, JZBZ for coal fly ash from precipitators in power plants, and JYRM for residues from residential coal burning. Similar profiles are also created for PM^sub 10^ fraction. They are all dominated by Al, Si, Ca, and TC. Al, Si, and Ca vary little between the three profiles, ranging among 15.5-17.4%, 13.3-16.7%, and 2.3- 3.6%, respectively. TC varies 16-fold from 1.1% in residential coal burning residues (JYRM) to 17.9% in coal fly ash from precipitators (JZBZ). TC is more abundant in TSP size fractions than in PM^sub 10^ size fraction, with TC ratios of PM^sub 10^-to-TSP ranging from 0.13 to 0.5.

Construction Material. Construction material is one of the key industries in Jiaozuo City and the basic material for frequent construction activities, which is an important reservoir of fugitive dust. Construction material here includes cement from cement kilns (JXLY), lime from limekiln (JZSHL), and gravel from quarrying (JZST) and construction sites (JZSNN). Four composite profiles are composed, most abundant in Ca, which varies from 24.8% in JXLY to 38.4% in lime. Other major constituents like Al, Si, Fe, and TC have different abundance patterns. Al varies 3-fold from 0.9% in JZSHL to 2.7% in cement product, JZSNN. Si also varies ~3-fold from 3.6% in gravel dust to 9.7% in JZSNN. TC varies 25-fold from 0.42% in JZSNN to 10.6% in JZST. TSP/PM^sub 10^ size fractions have a similar distribution of chemical abundances.

Paved Road Dust and Re-Entrained Dust. Paved roads have been considered as an important emitter to ambient air PM, because of fugitive emissions of road dust, which come from a variety of primary sources, such as trackout, wear of tire and brake, erosion from adjacent areas, pavement ware and spills, and so forth.30,41 Paved road dust, JZDL, is enriched in Ca (>10%), TC (>12.5%), OC (~7\%), Zn (~0.4%) and Pb (0.018%), decreased in Si (~14%), and almost equal in Al (~6%), K (1.5%), Fe (2.7%) and soluble ions, compared with soil. Re-entrained dust is dust that deposited on the surface of roofs, windowsills, balustrades, and so forth. It represents a kind of area source that formed by PM falling out of air and that can be reinjected into ambient air when enough disturbances occur, such as wind blow. Similar findings can be reached as we do with paved road when chemically comparing re- entrained dust (JZYC) with soil.

Here comparison analysis is made among paved road dust, re- entrained dust profiles, and receptor profiles, all in percentage of mass. CDs have been calculated between paved road dust and receptor sites (annual average) and between re-entrained dust and receptor sites (annual average), with all of the species investigated except ion components, because ion species abundances are mostly much higher than those in all of the source profiles. CDs between paved road dust and receptor sites range from 0.36 to 0.41 and 0.37 to 0.44 for TSP and PM^sub 10^, respectively. CDs between re-entrained dust and receptor sites vary from 0.38 to 0.43 and 0.42 to 0.47 for TSP and PM^sub 10^, respectively. The CDs indicate that paved road dust and receptor sites and re-entrained dust and receptor sites are chemically similar, to some extent. A similar conclusion also can be made when comparing paved road dust, re-entrained dust, and soil dust. This partly explains why many researchers often or always consider paved road dust as a geological source, as they do with soil dust.16,34-36,42,43

Paved road dust is built up from varieties of primary sources like soil, and re-entrained dust is accumulated from dustfall that originates mainly from primary sources. Thus, they can be seen as temporary sinks of primary sources, having typical properties of an ambient receptor, whereas they both are direct PM emitters to ambient air or a receptor site. That is to say, they have characteristics of both source and receptor. As sources, paved road dust and re-entrained dust account for contributions to ambient PM; however, when their profiles are included in CMB modeling, colinearity often occurs if soil profile is also included. As receptors, paved road dust and re-entrained dust can be apportioned to primary sources. In this study, paved road dust and re-entrained dust are not included as input to the CMB model, but the authors managed to give the contributions of primary sources to them.

Other Sources. Aside from sources described above, there are still many potential industrial sources within the study area. JZGT exhibits the chemical composition in steel making, with ~36.5% Fe and ~9.2% Si. Steel-making profile is significantly different from those PM^sub 10^ profiles of steel production (40% SO^sup 2-^^sub 4^, 11% Fe, 10% OC, 5% K, and 4%Si) reported in U.S. Environmental Protection Agency (EPA) SPECIATE 3.2.44 This may be caused by different sampling procedures and sites. JZBJS shows the chemical constitute abundances of particles emitted from cryolite production, highly abundant in Na (~18.5%) and Al (~9.5%). JZDN from calcium carbide production exhibits high contents of Ca (39%) and TC (10%).

Some other fugitive dust profiles like JZYM and JZBG are also created for coal storage piles and coal gangue piles, respectively. JZYM represents coal dust, characterized with high abundances of Al (6%), Si (8.7%), Ca (2.5%), and TC (31%). TC content in coal dust is much lower than expected and those found in other studies, such as ~50% TC in resuspended coal samples in Wuhan City, China,45 and 75- 88% TC in coal samples in South Africa.37 JZBG exhibits the chemical abundances in coal gangue dust, highly abundant in Si (~19%) and TC (~12%).

Motor vehicle exhaust has been reported as an important contributor to ambient air in urban areas around world. Here in this study, no vehicle exhaust samples were collected, gasoline or diesel. A composite profile combined with gasoline and diesel samples collected in Tianyuan City (capital of Shanxi Province, ~300 km north of Jiaozuo City) is used for source apportionment modeling in stead, because no natural gas-fueled vehicles run in Jiaozuo City in the past or near future. The abundant components in this profile are TC (89.9%), OC (51.7%), and SO^sup 2-^^sub 4^ (3.9%), whereas other constitutes are quite low, with most <1% in abundance,12 which is consistent with previously reported profiles.

It is obvious that soluble ions have much higher abundances (4- 10% SO^sup 2-^^sub 4^, 1.5-4.3% NO^sup -^^sub 3^ , and 0.3-2.2% NH^sup +^^sub 4^) than those in sources sampled during this study (vehicle exhaust has the highest abundances of SO^sup 2-^^sub 4^ (3.9%), NO^sup -^^sub 3^ (0.8%), and NH^sup +^^sub 4^ (0.4%) among the sources). That is to say, mass balance of soluble ions between sources and receptors cannot be achieved unless new contributors with high abundances of soluble ions are found. Because species like NO^sup -^^sub 3^ and SO^sup 2-^^sub 4^ can be formed through gas-to- particle transformation in the air, secondary sulfate and nitrate as potential sources can be introduced to keep mass balance of soluble ions, as previous studies did.29,43,46 In this study, “pure” ammonium bisulfate and ammonium nitrate are used as secondary sulfate and nitrate profiles, with 27% NH^sup +^^sub 4^ and 73% SO^sup 2-^^sub 4^ and 23% NH^sup +^^sub 4^ and 77% NO^sup -^^sub 3^ , respectively.

Source Apportionment Modeling

A CMB receptor model derived from EPA CMB 7.0, NKCMB, was applied to estimate the contributions to ambient TSP and PM^sub 10^ at the seven sites during the sampling periods and a whole year. Soil, coal fly ash, construction dust, steel-making emission, vehicle exhaust, and secondary sulfate and nitrate were included as important sources. They all were identified and their contributions estimated in most cases. Additionally, paved road dust and re-entrained dust were apportioned to primary sources.

Apportionment of Ambient PM. TSP and PM^sub 10^ source contributions for seasonal and annual average are illustrated in Figure 7. Obviously soil is almost the largest contributor at every site during every season and the whole year, accounting for 19-44%. Another two significant contributors were construction dust and coal combustion, accounting for 7-38% and 10-30%, respectively. These three source types constituted the majority of ambient particular matter. Sulfate and nitrate reached a level above 10%, indicating serious secondary aerosol pollution in the urban areas. Vehicular emissions had an estimate of 5-10%. Unknown sources contributed to less than 6% of PM mass. Therefore, abatement of PM emissions from geological sources, construction activities, coal combustion, and vehicular movement is the key to improve the air quality of Jiaozuo City.

In spring days, construction dust was the largest contributor at five sites (LUB, PF, MSM, LTOB, and MH) and the second largest at the rest two sites (EMC and RS) to ambient TSP, ranging spatially between 19.1% and 38.4% and 105 and 226 g/m^sup 3^. Soil dust was the reverse of ambient TSP compared with construction dust, ranging between 19.2% and 30.6% and 99 and 152 g/m^sup 3^. The third largest compared with TSP was coal combustion emission, accounting for 16.4- 23.6%, with the highest concentration at site LTOB (136.7 g/m^sup 3^). Vehicle emissions, secondary sulfate, and nitrate contributed 8.5% and 11.7% in spatial average, respectively, and steel making accounted for 3.4% ambient TSP in spatial average. For PM^sub 10^, soil dust was the largest contributor at five sites (LUB, MSM, LTOB, EMC, and RS) and the second largest at rest two sites (PF and MH), and construction dust contributed reversely when compared with soil dust. Coal combustion, vehicle exhaust, secondary sulfate and nitrate, and steel making accounted for 14%, 7.8%, 9.5%, and 3.1% in spatial average, respectively.

In summer days, soil dust was the largest contributor at almost all of the sites to both TSP and PM^sub 10^, ranging spatially between 26.9% and 39.7% and 105 and 174 g/m^sup 3^ for TSP and 25.8% and 41.5% and 95 and 168 g/m^sup 3^ for PM^sub 10^. Construction dust as the second largest source varied spatially between 19.3% and 32.6% and 82 and 171 g/m^sup 3^ for TSP and 16.6% and 26.1% and 53 and 106 g/m^sup 3^ for PM^sub 10^. Coal combustion emission accounted for 13-23.6% and 64-118 g/m^sup 3^ for TSP and 10.2-18.3% and 36-71 g/m^sup 3^ for PM^sub 10^. Vehicle exhaust, secondary sulfate and nitrate, and steel making contributed 6.4%, 7.2%, and 2.4% in spatial average to TSP and 9.9%, 10.3%, and 1.4% in spatial average to PM^sub 10^, respectively.

In winter days, soil dust remained the largest portions at almost all of the sites, ranging spatially between 24.6% and 33.2% and 121 and 186 g/m^sup 3^ for TSP and 21.3% and 38.7% and 62 and 124 g/ m^sup 3^ for PM^sub 10^. Coal combustion came up as the second largest, with contributions of 24.5-30.3% and 14-206 g/m^sup 3^ for TSP and 21.83-28.9% and 53-103 g/m^sup 3^ for PM^sub 10^. Construction dust as the second largest source varied spatially between 14.5% and 25.2% and 55 and 165 g/m^sup 3^ for TSP and 7.2% and 21.1% and 22 and 90 g/m^sup 3^ for PM^sub 10^. Vehicle exhaust, secondary sulfate and nitrate, and steel making contributed 8.4%, 10.1%, and 1.5% in spatial average to TSP and 12.2%, 15.6%, and 0.4% in spatial average to PM^sub 10^, respectively. Such a low contribution of steel making to ambient PM^sub 10^ was attributed to the fact that steel making was not identified by modeling at three sites (LTOB, EMC, and RS).

According to the seasonal variation of source contributions discussed above, contributions of some sources varied significantly with seasons. The contribution of coal combustion in winter was obviously higher than those in the other two seas\ons, which was attributed to heavier coal consumption for domestic heating in the cold season. Construction dust reached its largest percent contributions in spring days, for which a good explanation was not found. Major sources like soil dust, construction dust, and coal combustion contributed highly varied concentrations at different sites.

Apportionment of Paved Road Dust and Re-entrained Dust. Paved road dust and re-entrained dust are also apportioned to six source types by CMB modeling. Paved road dust and re-entrained dust are usually considered as sources because of their emissions to ambient air but seldom as receptors in almost all of the previous studies except in some literature.10,11,47 No detailed descriptions about how to apportion a “receptor” like paved road dust or re-entrained dust to primary sources are found in previous literatures. This paper describes why and how we can use CMB modeling to accomplish such an apportionment and gives the source contribution estimates to paved road dust and re-entrained dust.

Here we show a mathematical deduction for why such an apportionment can be done by using a CMB model, which is illustrated as follows, taking paved road dust for example. Given that the mass of paved road dust is M (grams), number of sources is n, jth source contributes mass Mj (grams) to M, then mass balance between the n sources and paved road dust can be expressed as follows:

… (2)

Mass is also balanced for every chemical species, expressed as follows:

… (3)

where F^sub Ri^ and F^sub ji^ are the abundances (g g^sup -1^) of ith species in paved road dust and in jth source, respectively. Divide both sides of eq 3 by M, and we can get the following:

… (4)

where M^sub j^/M is the contribution fraction of jth source to paved road dust. Let S^sub j^ = M^sub j^/M, then eq 4 can be rewritten as follows:

… (5)

Equation 5 is exactly the mathematical expression for a CMB model.48,49 Thus, paved road dust apportionment can be performed by a CMB model. The only difference between apportionments of ambient PM and paved road dust by CMB modeling is that paved road dust becomes a receptor, and the source contribution estimates are mass fractions (g/g), not mass concentrations (g/m^sup 3^). Additionally, an EPA CMB 7.0 receptor profile for paved road dust should be created by filling the fields of chemical concentrations and their uncertainties with chemical abundances and their uncertainties in paved road dust, and setting TOT (a species of the of paved road dust chemical profile) as 1, according to eq 2.

Six sources were identified by CMB modeling that contributed to paved road dust and re-entrained dust in this study. Their contributions were estimated in Figures 8 and 9. Soil (>50%), construction dust (~24%), and coal combustion (~15%) accounted for the major content of paved road dust and re-entrained dust. For paved road dust, secondary sulfate, and nitrate were not identified, but instead tire wear from EPA SPECIATE 3.244 was estimated to account for ~1%. For re-entrained dust, soil, construction dust, and coal combustion account for 46.3%, 18.5%, and 26.4% in TSP fraction and 55%, 17.2%, and 20.5% in PM^sub 10^ fraction, respectively. Secondary sulfate was identified to have an estimate of ~2%.

SUMMARY AND CONCLUSIONS

During this study, very high mass concentrations were observed at seven sites in the urban areas of Jiaozuo City, with annual spatial averages of 492 g/m^sup 3^ for TSP and 326 g/m^sup 3^ for PM^sub 10^, which by far exceeded the Chinese NAAQS limits. The spatial TSP average was high in spring and winter days at a level of ~530 g/ m^sup 3^ and low in summer days at 456 g/m^sup 3^; however, the spatial PM^sub 10^ average exhibited little variation at a level of ~325 g/m^sup 3^. Obvious site-to-site variations of mass concentrations were also observed. PM^sub 10^-to-TSP ratios ranged from 0.58 to 0.81. All of these facts suggested that heavy PM pollution existed in the urban areas of Jiaozuo City.

Many types of potential PM sources were sampled and determined for local source profiles. Soil sources were physically and chemically characterized in detail. Topsoil and deep soil profiles of a soil subtype showed no significant difference in chemical composition. Paved road dust and re-entrained dust were found to have similar chemical composition as soil dust, which partly explains why many researchers often or always considered road dust as geological sources. In addition, the receptor characteristics of paved road dust and re-entrained dust were discussed other than their properties as source.

Source apportionment modeling results suggest that soil dust, construction dust, and coal combustion are three major contributors to urban ambient PM pollution, and they totally accounted for >70%, each varying temporally and spatially. Related industries, such as electricity generation, ferrous and nonferrous mineral mining, and processing, ought to be abated, and fugitive dust from geological sources, construction sites, and paved roads or streets should be controlled strictly. Vehicle emission should also be kept under control because of its contribution and its harm to human health. This paper illustrated how to use CMB modeling to apportion paved road dust and re-entrained dust to their sources and found that soil, construction dust, and coal fly ash were the most important contributors to them in the urban areas of Jiaozuo City. All of the above findings help us understand the local sources and their contributions to serious particle pollution levels and also provide valuable insights for planning future monitoring and controlling strategies for airborne PM pollution in this area.

ACKNOWLEDGMENTS

This study was financially supported by the National Technology Supporting, Jiaozuo Environmental Protection Bureau, Henan Province, People’s Republic of China. The authors are grateful to the staff from Environmental Monitoring Center of Jiaozuo City for helping conduct ambient and source sampling during the study.

IMPLICATIONS

Properties of ambient TSP/PM^sub 10^ at seven sites and potential PM sources were characterized and source contributions were estimated in Jiaozuo, a medium-sized city in China, during January 2002 to April 2003. This paper carefully compared topsoil sample with deep soil sample of the same soil in aspects of chemical abundances and physical size distribution and found no significant difference between them. Road dust and re-entrained dust as “receptors” were also discussed here because they were important sources in many previous studies, and they were apportioned to other sources by CMB modeling. The above findings may be helpful for other source apportionment studies.

REFERENCES

1. State Environmental Protection Administration. Report on the State of the Environment in China 2000; available at http:// www.sepa.gov.cn/english/SOE/soechina2000/english/atmospheric/ atmospheric_e.htm (accessed 2006).

2. State Environmental Protection Administration. Report on the State of the Environment in China 2001; available at http:// www.sepa.gov.cn/english/SOE/soechina2001/english/2-air.htm (accessed 2006).

3. State Environmental Protection Administration. Report on the State of the Environment in China 2002; available on State Environmental Protection Administration of China Web site, http:// www.zhb.gov.cn/english/SOE/soechina2002/air.htm (accessed 2006).

4. State Environmental Protection Administration. Report on the State of the Environment in China 2003; available at http:// www.zhb.gov.cn/english/SOE/soechina2003/air.htm (accessed 2006).

5. State Environmental Protection Administration. Report on the State of the Environment in China 2004; available at http:// www.zhb.gov.cn/english/SOE/soechina2004/air.htm (accessed 2006).

6. State Environmental Protection Administration. Ambient Air Quality Standard (GB3095-1996); State Environmental Protection Administration of China; available at http://www.zhb.gov.cn/ image20010518/5298.pdf (accessed 2006).

7. Sun, Y.L.; Zhuang, G.S.; Wang, Y.; Han, L.H.; Guo, J.H.; Dan, M.; Zhang, W.J.; Wang, Z.F.; Hao, Z.P. The Air-Borne Particulate Pollution in Beijing-Concentration, Composition, Distribution and Sources; Atmos. Environ. 2004, 38, 5991-6004.

8. Ye, B.M.; Jia, X.L.; Yang, H.Z.; Yao, X.H.; Chanb, C.K.; Cadlec, H.S.; Chan, T.; Mulawa, P.A. Concentration and Chemical Composition of PM^sub 2.5^ in Shanghai for a 1-Year Period; Atmos. Environ. 2003, 37, 499-510.

9. State Environmental Protection Administration. Annual Report on Urban Environmental Management and Comprehensive Improvement of Environmental Protection Key Cities in 2004; State Environmental Protection Administration of China; available at http:// www.sepa.gov.cn/eic/650497471566315520/20050602/8237_3.shtml (accessed 2006).

10. Dai, S.G.; Zhu, T.; Zeng, Y.S.; Fu, X.Q.; Liao, Y.M. Source Apportionment for Tianjin Urban Aerosol in Heating Season (in Chinese with abstract in English); China Environ. Sci. 1986, 6, 24- 30.

11. Zhu, T.; Bai, Z.P.; Chen, W. Source Apportionment of Air Particulate in Qinhuangdao City (in Chinese with abstract in English); Res. Environ. Sci. 1995, 8, 49-55.

12. Liang, L.M.; Zhu, T. Source Apportionment of Airborne Particulate in Taiyuan City (Report); Press of Nankai University: Tianjing, People’s Republic of China, 2002.

13. Hildemann, L.M.; Cass, G.R.; Markowski, G.R. A Dilution Stack Sampler for Collection of Organic Aerosol Emission: Design, Characterization and Field Tests; Aerosol Sci. Technol. 1989, 10, 193-204.

14. Yu X.N. A Dilution Stack Sampling System for Point Sources: Design and Application, Dissertation for a Ph.D. Degree; Peking University: Peking, People’s Republic of China, 2004; pp 20-28.

15. Chow, J.C.; Watson, J.G.; Houck, J.E.; Pritchett, L.C.; Rogers, C.F.; Frazier, C.A.; Egami, R.T.; Ball, B.M. A Laboratory Resuspension Chamber to Measure Fugitive Dust Size Distributions and Chemical \Compositions; Atmos. Environ. 1994, 28, 3463-3481.

16. Gillies, J.A.; O’Connor C. M.; Mamane Y.; Gertler, A.W. Chemical Profiles for Characterising Dust Sources in an Urban Area, Western Nevada, USA; Z.Geomorph. N.F. 1999, 116(Suppl.-Bd),19-44.

17. Latour, T.E. Analysis of Rocks Using X-Ray Fluorescence Spectrometry; Rigaku J. 1989, 6, 3-9.

18. Liu F.Y.; Guo, G. Determination of Twenty-One Elements in the Refly Dust by X-Ray Fluorescence Spectrometry (in Chinese with abstract in English); Urban Environ. Urban Ecol. 1996, 9, 45-47.

19. Watson, J.G.; Chow, J.C.; Frazier, C.A. X-Ray Fluorescence Analysis of Ambient Air Samples. In: Landsberger, S., Creatchman, M., Eds.; Elemental Analysis of Airborne Particles. Gordon and Breach: New York, NY, 1999; pp 67-96.

20. Yu, C.Y. Determination and Characterization of Water-Soluble Species of PM^sub 2.5^ in Beijing; Dissertation for a Ph.D. degree; Tsinghua University: Tsinghua, People’s Republic of China, 2004; pp 39-41.

21. Carvalho, L.R.F., Souza, S.R.; Martinis, B.S.; Korn, M. Monitoring of the Ultrasonic Irradiation Effect on the Extraction of Airborne Particle Matter by Ion Chromatography; Anal. Chim. Acta; 1995, 317, 171-179.

22. Zhang, Z.S.; Lu, Y. The Determination of F^sup -^, Cl^sup – ^, NO^sup -^ ^sub 3^ and SO^sup 2-^^sub 4^ on the Total Suspended Particles in Air by Ion Chromatography (in Chinese with abstract in English). Chinese J. Chromatogr. 1999, 17, 313-314.

23. Talebi, S.M.; Abedi, M. Determination of Atmospheric Concentrations of Inorganic Anions by Ion Chromatography Following Ultrasonic Extraction; J. Chromatogr. A 2005, 1094, 118-121.

24. Chi, X.G.; Di, Y.A.; Dong, S.P.; Liu, X.D. Determination of Organic Carbon and Elemental Carbon in Atmospheric Aerosol Samples (in Chinese with abstract in English); Environ. Monit. China 1999, 15, 11-13.

25. Liu, X.D.; Chi, X.G.; Duan, F.K.; Dong S.P.; Yu, T. Determination of the Organic Carbon and Elemental Carbon in Chinese Urban Aerosol by Using CHN Elemental Analyzer; Aerosol Sci. 2000, 31(Suppl. 1), S240-S241.

26. Chio, C.P.; Cheng, M.T.; Wang, C.F. Source Apportionment to PM^sub 10^ in Different Air Quality Conditions for Taichung Urban and Costal Areas, Taiwan; Atmos. Environ. 2004, 38, 6893-6905.

27. Dong, S.P.; Liu, X.D.; Qi, H.; Zhang, T.; Willy, M. Direct Determination of Element Carbon in the Atmospheric Aerosols by CHN Elemental Analyzer (in Chinese with abstract in English); Environ. Monit. China 2004, 20, 20-23.

28. Li, W.Q. The Pollution Characteristics of TSP and PM^sub 10^ in Aatmospheric Aerosol in Several Important Cities of Henan (in Chinese with abstract in English); Henan Sci. 2004, 22, 714-717.

29. Samara, C;. Chemical Mass Balance Source Apportionment of TSP in a Lignite-Burning Area of Weastern Macedonia, Greece; Atmos. Environ. 2005, 39, 6430-6443.

30. Querol, X.; Alastuey, A.; Rosa, J.D.; Campa, S.A.; Plana, F.; Ruiz, C.R. Source Apportionment Analysis of Atmospheric Particulates in an Industrialized Urban Site in Southwestern Spain; Atmos. Environ. 2002, 36, 3113-3125.

31. Chow, J.C.; Watson, J.G.; Lu, Z.Q.; Lowenthal, D.H.; Frazier, C.A.; Solomon, P.A.; Thuillier, R.H.; Magliano, K. Descriptive Analysis of PM^sub 2.5^ and PM^sub 10^ and Regionally Representative Locations during SJVAQS/AUSPEX; Atmos. Environ. 1996, 30, 2079- 2112.

32. Chow, J.C.; Watson, J.G.; Green, M.C.; Lowenthal, D.H.; Bates, B;. Oslund, W.; Torres, G. Cross-Border Transport and Spatial Variability of Suspended Particles in Mexicali and California’s Imperial Valley; Atmos. Environ. 2000, 34, 1833-1843.

33. Chow, J.C.; Watson, J.G.; Lowenthal, D.; Countess, R.J. Sources and Chemistry of PM^sub 10^ Aerosol in Santa Barbara County, CA; Atmos. Environ. 1995, 30, 1489-1499.

34. Watson, J.G.; Chow, J.C. Source Characterization of Major Emission Sources in the Imperial and Mexicali Valleys along the US/ Mexico Border; Sci. Total Environ. 2001, 276, 33-47.

35. Vega, E.; Mugica, V.; Reyes, E.; Snchez, G.; Chow, J.C.; Watson, J.G. Chemical Composition of Fugitive Dust Emitters in Mexico City; Atmos. Environ. 2001, 35, 4033-4039.

36. Ho, K.F.; Lee, S.C.; Chow, J.C.; Watson, J.G. Characterization of PM^sub 10^ and PM^sub 2.5^ Source Profiles for Fugitive Dust in Hong Kong; Atmos. Environ. 2003, 37, 1023-1032.

37. Engelbrecht, J.P.; Swanepoel, L.; Chow, J.C.; Watson, J.G.; Egami, R. The Comparison of Source Apportionment from Residential Coal and Low-Smoke Fuels, Using CMB Modeling, in South Africa; Environ. Sci. Policy 2002, 5, 157-167.

38. Wonphatarakul, V.; Friedlander, S.K.; Pinto, J.P. A Comparative Study of PM^sub 2.5^ Ambient Aerosol Chemical Databases; Environ. Sci. Technol. 1998, 32, 3926-3934.

39. Zhang, Z.Q.; Friedlander, S.K. A Comparative Study of Chemical Databases for Fine Particle Chinese Aerosols; Environ. Sci. Technol. 2000, 34, 4687-4694.

40. Liu, T.G.; Wang, Y.J. The United Application of the Dust Centrifugal Classifier (YFJ Type) and the Standard Sieves (in Chinese with abstract in English); J. Beijing Inst. Light Ind. 1997, 15, 36-40.

41. Chow, J.C.; Watson, J.G. Fugitive Emissions Add to Air Pollution; Environ. Protect. 1992, 3, 26-31.

42. Chow, J.C.; Watson, J.G.; Kuhns, H.; Etyemezian, V.; Lowenthal, D.H.; Crow, D.; Kohl, S.D.; Engelbrecht, J.P.; Green, M.C. Source Profiles for Industrial, Mobile and Area Sources in the Big Bend Regional Aerosol Visibility and Observational Study; Chemosphere 2004, 54, 185-208.

43. Watson, J.G.; Chow, J.C.; Lu, Z.Q.; Fujita, E.M.; Lowenthal, D.H.; Lawson, D.R. Chemical Mass Balance Source Apportionment of PM^sub 10^ during the Southern California Air Quality Study; Aerosol Sci. Technol. 1994, 21, 1-36.

44. U.S. Environmental Protection Agency Software; SPECIATE Version 3.2; Released on November 3, 2002; available at http:// www.epa.gov/ttn/chief/software/speciate/(accessed 2006).

45. Zelenka, M.P.; Wilson, W.E.; Chow, J.C.; Lioy, P.J. A Combined TTFA/CMB Receptor Modeling Approach and Its Application to Air Pollution Sources in China; Atmos. Environ. 1994, 28, 1425- 1435.

46. Watson, J.G.; Chow, J.C. Review of PM^sub 2.5^ and PM^sub 10^ Apportionment for Fossil Fuel Combustion and Other Sources by the Chemical Mass Balance Receptor Model; Energy Fuels 2002, 16, 222- 260.

47. Feng, Y.C.; Bai, Z.P.; Zhu, T. The Principle and Application of Improved Source Apportionment Technique of Atmospheric Particulate Matter (in Chinese with abstract in English); China Environ. Sci. 2002, 23(Suppl.), 106-108.

48. Watson, J.G.; Cooper, J.A.; Huntzicker, J.J. The Effective Variance Weighting for Least Squares Calculations Applied to the Mass Balance Receptor Model; Atmos. Environ. 1984, 18, 1347-1355.

49. Receptor Model Technical Series, Volume III (1989 Revision); EPA-450/4-90-004; U.S. Environmental Protection Agency: Research Triangle Park, NC, 1990; pp A-3-A-9.

Yinchang Feng, Yonghua Xue, Xiaohua Chen, Jianhui Wu, Tan Zhu, and Zhipeng Bai

State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control, Nankai University, Tianjin, People’s Republic of China

Shengtang Fu and Changju Gu

Jiaozuo Environmental Monitoring Center, Jiaozuo City, People’s Republic of China

About the Authors

Yinchang Feng, Yonghua Xue, Xiaohua Chen, Jianhui Wu, Tan Zhu, and Zhipeng Bai are in the State Environmental Protection Key Laboratory of Urban Ambient Air Particulate Matter Pollution Prevention and Control at College of Environmental Science and Engineering, Nankai University. Shengtang Fu and Changju Gu are in Jiaozuo Environmental Monitoring Center. Address correspondence to Yinchang Feng, College of Environmental Science and Engineering, Nankai University, Tianjin 300071, China; phone: 86-22-23503397; fax: 86-22-23503397; e-mail: fengyc@nankai.edu.cn.

Copyright Air and Waste Management Association May 2007

(c) 2007 Journal of the Air & Waste Management Association. Provided by ProQuest Information and Learning. All rights Reserved.