Chemosphere 95 (2014) 387–394

Contents lists available at ScienceDirect

Chemosphere journal homepage: www.elsevier.com/locate/chemosphere

An investigation into the relationship between the major chemical components of particulate matter in urban air Yong-Hyun Kim a, Ki-Hyun Kim a,⇑, Chang-Jin Ma b, Zang-Ho Shon c, Chan Goo Park d, Sang-Keun Song e, Chul-Un Ro f, Richard J.C. Brown g a

Department of Environment & Energy, Sejong University, Seoul 143-747, Republic of Korea Department of Environmental Science, Fukuoka Women’s University, Fukuoka 813-8529, Japan Department of Environmental Engineering, Dong-Eui University, Busan 614-714, Republic of Korea d Seoul Metropolitan Institute of Public Health & Environment, Seoul 137-734, Republic of Korea e Department of Earth and Marine Sciences, Jeju National University, Jeju 690-756, Korea f Department of Chemistry, Inha University, YonghyunDong, NamGu, 402-751 Incheon, Republic of Korea g Analytical Science Division, National Physical Laboratory, Hampton Road, Teddington TW11 0LW, UK b c

h i g h l i g h t s  The major components of particulate matter (MCP) were investigated in the urban air.  The behavior of TSP is evaluated to assess the interactive roles between the MCP.  The composition of MCP generally varied in the descending order: anions, OC, cations, EC, and metals.  The measured MCP was able to account for 60% of total TSP composition during the study period.

a r t i c l e

i n f o

Article history: Received 11 June 2013 Received in revised form 4 September 2013 Accepted 10 September 2013 Available online 14 October 2013 Keywords: Cation Anion TSP Metal Industrial Air quality

a b s t r a c t Particulate matter (PM) generally comprises such chemical components as inorganic ions, organic carbon (OC), elemental carbon (EC), and metals. In terms of environmental studies, these major chemical components of particulate matter (MCP) are important in understanding PM distribution, behaviors and source apportionment. In this study, the MCP fractions of total suspended particles (TSP) were measured at an urban residential area in Seoul, Korea from February to December in 2009. The behavior of each individual MCP was studied in order to explain their relationship to environmental conditions and sources. The MCP measured during this study period was able to account between 54% (spring) to 67% (fall) of total TSP composition. During the study period, it was found that the TSP sampled comprised mostly: anions, OC, cations, EC, and metals in decreasing order of abundance. Although such relative ordering seems to remain fairly constant over time, the relative balance of this relationship may be altered by variations in environmental conditions. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Particulate matter (PM) is emitted into the atmosphere by both natural (i.e., crustal matter, sea-salt, etc.) and anthropogenic sources (i.e., industrial processes, transportation, combustion, etc.) (Zheng et al. 2004; Fang et al., 2006; Hwang and Hopke, 2011). As PM contains harmful metals and organic compounds and can adsorb gaseous pollutants from the air, it acts as the main media for the transfer of hazardous pollutants across various environmental reservoirs (Greco et al., 2007). Information concerning the distribution and composition of PM is also important, as the ⇑ Corresponding author. Tel.: +82 2 499 9151; fax: +82 2 3408 4320. E-mail address: [email protected] (K.-H. Kim). 0045-6535/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chemosphere.2013.09.050

individual components may be harmful to human health upon ingestion or inhalation (Seaton et al., 1995; Harrison and Yin, 2000). Megacities have high densities of population and traffic activity. Because emissions of airborne pollutants resulting from high densities of PM-emitting activities cause diverse socio-economic problems, many governments have implemented policies aimed at improving air quality. For example, a policy to control the emission of sulfur from fuel has been the driving force in reducing sulfur dioxide (SO2) in Korea since 1981 (National Institute of Environmental Research (NIER), 2007). The implementation of such polices in Korea has also been helpful in the recent past to reduce concentration levels of certain important pollutants such as PM10 and some PM-bound metallic components like Pb and Cd

388

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

(Kim, 2007a, b; Kim, 2010) as well as certain gaseous components like CO and NOx (Pandey et al., 2008; Kim and Shon, 2011; Shon and Kim, 2011). Similar patterns of abatement have commonly been reported in many developed and developing countries (e.g., Fenger, 1999; Begum et al., 2011; Kim and Shon, 2011). However, it should be noted that levels of change in total suspended particles (TSP) levels, although decreasing gradually since 1990, have not matched the dramatic reductions observed for other pollutants in Korea (e.g., NIER, 2010; Kim, 2010). The important chemical components of TSP in air which are regularly measured by air quality networks are inorganic ions, organic carbon (OC), elemental carbon (EC), and crustal metals (such as Fe, Al, S and Si) (Kang et al., 2009). The remaining fraction of TSP is generally comprised of multiple chemical components at low abundance, and these are generally not routinely quantitatively measured. In order to precisely describe the sources of PM and its behavior in the atmosphere, it is important to have a better understanding of the relationship between the MCPs (Woodruff et al., 1997; Biswas et al., 2009). In this study, the concentrations of TSP and its MCP fraction were measured at an urban residential area (Yang Jae district in Seoul, Korea) during the months of February through December 2009. These measurement data were analyzed to assess the relationship between the MCP fractions and between the MCP and the concurrently measured environmental parameters. The

results of this study will help us extend a better knowledge on the major chemical components of TSP fractions in urban environments and the basic factors influencing the distribution, behavior, and source apportionment of PM under various conditions.

2. Materials and methods 2.1. Site characteristics In order to assess the behavior of the MCPs, the concentrations of OC, EC, metal, and inorganic ions were analyzed along with TSP total mass from the Yang Jae (YJ) site (E–W 127°01.550 , N–S 37°27.510 ) in Seoul, Korea during the months of February through December 2009 (Fig. 1). The site was selected, as it can represent one of the urban background areas consisting dominantly of residential facilities. The city of Seoul has a population of 10 464 051 with 4 116 660 households as of 2009 (Seoul Metropolitan Government (SMG), 2010). Because the YJ site belongs to one of the subdistricts of the Seo Cho district, the basic characteristics of the YJ site can be explored by the general descriptions of the Seo Cho district. According to the SMG (2010), Seo Cho district has a population of 431 131 with a density of 9172 persons per square kilometer. There are a total of 30 registered air pollutants emitting

Fig. 1. Geographical location of Yang Jae (YJ) district in Seoul, Korea (from Google map).

389

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

facilities (man-made sources) in Seo Cho district, excluding domestic and traffic sources. These comprise about 2.81% of such facilities in Seoul (SMG, 2010). In this works a total of 11 field studies were carried out for the collection of daily TSP samples at the YJ site on three consecutive days over 11 periods from February to December 2009. The study duration of 3 d per month period was selected to collect relatively large data sets under acute shortage of skilled manpower (needed to operate multiple instrumental systems) for the limited time period. The specific sampling schedule for each month can be described as: (1) 23–25 February, (2) 9–11 March, (3) 6–8 April, (4) 11–13 May, (5) 9–11 June, (6) 6–8 July, (7) 11–13 August, (8) 14–16 September, (9) 12–14 October, (10) 9–11 November, and (11) 7–9 December. During these sampling periods, a number of the basic meteorological parameters (e.g., temperature (Temp), humidity (RH), wind speed (WS), wind direction (WD), ultraviolet (UV), and solar irradiance (radiation)) were simultaneously measured at same site.

Table 1 Summary of the measurements of TSP and its major constituents (OC, EC, total metals, and ions) at the Yang Jae (YJ) site in Seoul, Korea (February to December 2009).

A. Major chemical components of TSP TSP

3. Results and discussion 3.1. The general patterns of air pollution at the study site In Table 1, a statistical summary is presented to describe the concentration of TSP and its MCP (OC, EC, Fe, Rnon-Fe, RAnion,

75.3 ± 44.3 (66.0)a 9.40–192 (33)b 11.4 ± 4.61 (11.4) 3.64–22.7 (30) 3.85 ± 1.68 (3.90) 1.62–8.17 (30) 2.82 ± 1.49 (2.74) 0.133–6.40 (33) 0.611 ± 0.311 (0.541) 0.0546–1.40 (33) 7.45 ± 5.25 (5.93) 0.690–22.6 (33) 18.1 ± 13.7 (13.6) 0.741–56.1 (33)

OC EC Fe

Rnon-Fe RCation RAnion Element

Unit (lg Sm3)

Ion

Unit (lg Sm3)

B. Individual (identified) chemical components of TSP Fe 2820 ± 1485 (2739)a Na+ 0.564 ± 0.232 (0.567) 133–6401 (33)b 0.197–1.16 (32) As 7.09 ± 4.24 (7.05) 4.20 ± 3.30 (3.34) NHþ 4 0.285–17.6 (28) 0.0997–13.1 (33) B 22.9 ± 12.8 (18.4) K+ 0.509 ± 0.375 (0.410) 7.31–49.4 (31) 0.0578–1.54 (33) 1.97 ± 1.59 (1.53) Ba 102 ± 57.8 (83.4) Ca2+ 14.6–236 (33) 0.169–6.94 (33) Cd 2.33 ± 1.49 (2.19) Mg2+ 0.229 ± 0.183 (0.157) 0.159–6.31 (33) 0.0177–0.851 (33) Co 1.42 ± 0.712 (1.35) F0.0833 ± 0.0972 (0.0280) 0.101–3.42 (32) 9.32E-4–0.346 (33) Cr 10.2 ± 4.81 (9.74) Cl 0.710 ± 0.931 (0.270) 1.36–22.6 (31) 0.0233–3.24 (33) Cu 76.5 ± 48.7 (57.4) NO 7.63 ± 7.18 (5.71) 3 10.9–198 (33) 0.392–27.6 (33) Li 2.77 ± 2.08 (2.09) 9.64 ± 7.67 (7.91) SO2

2.2. Sampling and analysis The collection of TSP samples was made on a daily basis using a high volume air sampler (SIBATA, HV-1000F, Japan) onto Quartz type PALLFLEX membrane filters. (Prior to sampling, filters were baked at 850 °C for 2 h.) The collection of each TSP sample was initiated at 15:00 and ran for 24 h (flow rate of 800 L min1). These TSP samples were analyzed for their OC, EC, inorganic ion, and metal content. Both unexposed and exposed Quartz type PALLFLEX membrane filters were equilibrated at a temperature of 20 °C and a relative humidity of 50% for a minimum of 24 h prior to weighing. Weighing was performed on a microbalance with 0.001 mg sensitivity. The charge on each filter was neutralized by exposure to a 210 Po ionizing source for at least 30 s prior to the filter being placed on the balance pan. In Table 1S, the experimental conditions for the analysis of ions and metals fraction are summarized. Specific information about the analysis of the TSP samples has been provided in a number of our recent publications made at the same monitoring site (Anthwal et al., 2010; Nguyen et al., 2010; Shon et al., 2012). In order to analyze ion fractions of TSP, filter samples were added to 50 mL distilled water and treated by ultrasonicator (Branson 8510-EDTH) for 30 min. Then, ion fractions were analyzed by ion chromatography (Metrohm 850 Profic professional IC). The concentrations of EC and OC were analyzed by thermal/ optical transmittance Carbon aerosol analyzer (Manufacturer, Sunset Laboratory Inc.). The OC concentration can experience a positive bias because of CaCO3 contained within the blank filters. To eliminate any bias, punched filter pieces were exposed to HCl fumes for 1 h in a desiccator before analysis. For the analysis of metallic components, TSP samples were pretreated with 5 mL of mixed acid (5.55% HNO3/16.75% HCl). Then, 5 mL of 10 ppm Y2O3 was added as a surrogate standard. These samples were then placed in PFA vessels and heated using a microwave digestion system (Tekton, Q45) to 180 °C for 10 min. Extracted solutions were filtered and washed with 5 mL of the above mixed acid solution to produce a final volume of 15 mL. Finally, the metallic components were analyzed using ICP-AES (Spectro DE/CIROS vision, Germany).

Unit (lg Sm3)

Compound

4

Mn Mo Ni Pb Sr V Zn a b

0.146–8.86 (27) 72.9 ± 39.2 (77.5) 3.22–170 (33) 2.99 ± 2.07 (2.33) 0.00460–7.96 (22) 6.61 ± 4.13 (6.03) 0.362–18.1 (33) 73.0 ± 44.3 (63.3) 12.1–173 (32) 10.8 ± 6.47 (10.1) 0.847–28.1 (33) 6.91 ± 4.77 (6.39) 0.221–22.8 (33) 220 ± 146 (189) 15.1–791 (33)

0.262–31.1 (33)

Mean ± SD (median). Range (number of data).

and RCation) measured from the YJ site over the whole study period. Ion fractions were sorted into cations and anions. Metallic components were classified as ferrous and all non-ferrous (combined together) metals. In Table 1b, a statistical summary is also provided for the individual elemental and ion fractions of TSP. As shown in Table 2, the annual mean concentrations of NO2 and PM10 exceeded the regulation guidelines set for ambient air quality by the Korean Ministry of Environment (KMOE) such as: (1) SO2: 0.02 ppm (annual), (2) CO: 9 ppm (8 h), (3) NO2: 0.03 ppm (annual), (4) PM10: 50 lg m3 (annual), (5) O3: 0.06 ppm (8 h), (6) Pb: 0.5 lg m3 (annual), and (7) benzene: 5 lg m3 (annual) (NIER, 2010). The seasonal variation of TSP and its associated chemical components is plotted in Fig. 2. The seasonal variation of TSP is plotted in Fig. 1S in supplementary information (SI). The wind patterns recorded at the study site are also presented in Fig. 1S. In addition, Table 2S also details the basic environmental parameters measured

390

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

Table 2 Summary of concurrently measured environmental (important pollutants and meteorological) parameters at the Yang Jae site in Seoul, Korea from February to December 2009.

a b

Compound

Unit

A. Air pollutants SO2

(ppb)

NO

(ppb)

NO2

(ppb)

NOx

(ppb)

CO

(ppb)

O3

(ppb)

PM10

(lg m3)

PM2.5

(lg m3)

CH4

(ppm)

NCH4

(ppm C)

THC

(ppm C)

Parameter 4.28 ± 2.80 (3.96)a 2.00–14.0 (33)b 24.3 ± 26.1 (14.8) 2.46–107 (33) 33.6 ± 17.1 (30.3) 7.17–84.0 (33) 57.7 ± 37.9 (48.3) 9.71–152 (33) 628 ± 230 (600) 288–1100 (33) 19.4 ± 11.6 (18.9) 1.63–48.0 (33) 52.3 ± 31.9 (51.5) 6.50–145 (33) 26.9 ± 17.4 (24.0) 2.57–74.0 (33) 2.14 ± 0.156 (2.12) 1.93–2.43 (33) 0.317 ± 0.111 (0.298) 0.160–0.548 (33) 2.46 ± 0.255 (2.40) 2.09–2.91 (33)

Unit

B. Meteorological parameter Temp (°C) RH

(%)

WS

(m s1)

UV

(mW cm2)

Radiation

(W m2)

15.3 ± 7.62 (16.9) 1.70–29.0 (33) 55.3 ± 14.9 (56.1) 29.0–85.4 (33) 0.899 ± 0.292 (0.800) 0.600–1.76 (33) 0.334 ± 0.218 (0.300) 0.0346–0.850 (33) 138 ± 73.5 (131) 15.2–270 (33)

Mean ± SD (median). Range (number of data).

(A) Seasonal TSP and its MCP concentrations

Dust. Unfortunately, however, our analysis of elemental components did not cover major crustal elements like Si and Al that tend to be more abundant than Fe. Hence, our analysis of TSP composition is expected to underestimate the relative contribution of metallic components. The relative composition (%, in median) of ions (RAnion + RCation) in TSP mass showed a slight seasonal variation with the highest values in summer (35%) and the lowest in fall (28%) mainly due to anion variation. Unlike other chemical components, there was no seasonal variation in relative composition (10%) in cationic species in TSP. The major ion fractions of TSP were the secondary inor þ ganic ions such as SO2 4 , NO3 , and NH4 , whereas all others constituted only minor fractions. Thus, relative summertime enhancement of ions was likely to be related in part to photochemical activity of their gases precursors (HNO3, SO2, and NH3) through gas to particle conversion (ammonium sulfate, (NH4)2SO4 and ammonium nitrate, NH4NO3) under high relative humidity conditions (RH = 67.7%). Meteorological parameters such as temperature and RH can affect ion fractions in TSP via deliquescence (absorbing water) and efflorescence (crystallization) of ammonium sulfate and/or ammonium nitrate and heterogeneous hydrolysis of N2O5 (nitrate formation), especially at night (Pathak et al., 2011). Meanwhile, relative composition of the sum fractions between OC and EC concentrations did not show significant seasonal variation (17–19%) except for fall (27%). 3.2. TSP vs. its major chemical components

(B) TSP and MCP concentrations with weather conditions Fig. 2. Comparison of TSP and MCP measured at the Yang Jae site and their relationship with seasons and weather conditions.

at the YJ site during the study period. In line with general expectation, comparison of these data clearly indicates the relative dominance of both TSP and MCP in winter and spring (compared with summer and autumn). Most importantly, the occurrence of abnormal levels of particles and their chemical components are apparent, especially during spring, reflecting the inflow of Asian

The impact of TSP on the environment and human health can be assessed briefly by the type and concentrations of its chemical components (Cape et al., 2011; Nazir et al., 2011; Tan et al., 2011). Despite the difficulty in accurate identification of all the chemical components of TSP, it is possible to assess the basic relationship between TSP and its chemical components. As a means to explore the effect of meteorological and weather conditions on particle behavior, their concentration data were examined in relation to such variables. If the magnitude of TSP levels is compared across seasons, the highest value is seen during spring (mean 115 lg m3) followed by winter (75 lg m3),

391

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

Table 3 The concentration of TSP and its major identified chemical components (MCP: OC, EC, Fe, Rnon-Fe, RCation, RAnion, and Rnon-X) measured at the Yang Jae district of Seoul, Korea during the study period (February to December 2009; units in lg Sm3 and n = 33).

a

Order

Month

Date

Julian Day

TSP

OC

EC

Fe

Rnon-Fe

RCation

RAnion

RMCP

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

February

09-02-23 09-02-24 09-02-25 09-03-09 09-03-10 09-03-11 09-04-06 09-04-07 09-04-08 09-05-11 09-05-12 09-05-13 09-06-09 09-06-10 09-06-11 09-07-06 09-07-07 09-07-08 09-08-11 09-08-12 09-08-13 09-09-14 09-09-15 09-09-16 09-10-12 09-10-13 09-10-14 09-11-09 09-11-10 09-11-11 09-12-07 09-12-08 09-12-09

54 55 56 68 69 70 96 97 98 131 132 133 160 161 162 187 188 189 223 224 225 257 258 259 285 286 287 313 314 315 341 342 343

60 88 76 107 65 83 156 171 192 36 107 114 131 52 66 81 42 60 9.4 26.7 116 61.4 68 61.9 66.5 37.1 64.3 16.4 19.3 24.2 112 77.4 35

12.2 11.8 12.6 11.1 9.77 19.6 17.3 20.8 22.7 6.42 14.1 15.5 14.8 6.92 7.01 13 9.61 10.7 3.64 8.92 12 12 9.41 12.4 11.1 6.57 11.7 6.23 5.45 6.08 –a – –

5.39 4.27 4.23 3.14 3.21 6.17 6.08 7.63 8.17 2.41 4.13 4.36 3.97 1.62 1.65 3.76 2.3 1.88 1.67 3.54 3.9 3.9 2.76 4.36 4.64 2.04 4.78 2.43 4.72 2.34 – – –

2.28 2.34 2.63 3.12 3.01 3.59 4.56 5.73 3.82 3.12 2.79 6.4 2.05 0.63 1.77 1.84 1.24 1.22 0.13 1.07 3.11 3.39 2.74 4.01 5.69 2.08 3.94 2.02 1.57 1.43 4.87 2.97 1.89

0.53 0.57 0.53 0.54 1.09 0.72 0.86 1.09 0.73 0.41 0.39 1.4 0.54 0.17 0.34 0.53 0.42 0.38 0.05 0.32 0.78 0.75 0.63 0.72 0.92 0.47 0.89 0.32 0.23 0.21 1.22 0.83 0.56

7.78 10.1 5.29 10.93 3.86 5.93 22.6 16.5 22.4 6.16 4.14 6.7 12.2 3.19 6.42 8.49 4.42 5.39 0.69 1.64 13.6 7.11 7.19 5.34 5.35 5.17 5.49 2.41 2.06 2.66 11 9.34 4.33

19.5 27.7 13.4 28.8 7.86 12.7 56.1 38.1 54.4 13.6 8.8 15.1 33.8 7.38 18.2 24.5 11.2 14.5 0.74 3.05 33.9 18.3 17.8 12.3 12.4 12.9 13 1.52 1.2 3.13 28.2 22.6 9.22

47.7 56.8 38.7 57.6 28.8 48.7 108 89.9 112 32.1 34.4 49.5 67.4 19.9 35.4 52.1 29.2 34.1 6.92 18.5 67.3 45.5 40.5 39.1 40.1 29.2 39.8 14.9 15.2 15.9 45.3 35.7 16.0

March

April

May

June

July

August

September

October

November

December

Not measured.

summer (65 lg m3), and fall (47 lg m3) (Fig. 2A). The relative dominance of PM levels in spring/winter is highly consistent with the patterns observed in many previous studies (e.g., Nguyen et al., 2010). In addition, the behavior of particles can be assessed further by examining their data with three types of simplified weather conditions ([1] clear (no cloud), [2] cloudy, and [3] rainy). If the concentrations of TSP and its MCP are compared with such weather types, the results shown in Fig. 2B comply well with general expectations. More specifically, their concentrations decrease on the order of clean, unclean, and rainy conditions. As seen in Table 2S, the weather conditions of spring season in this study were dominated by ‘‘clear’’ conditions (7 out of all 9 measurements). In Korea, weather conditions of spring and winter are characterized by continental air mass with little rain, while those of summer and fall by oceanic air mass with frequent and heavy rain. As such, the relative enhancement of TSP and its MCP during spring or winter can also be accounted for in this respect to a certain degree. In Table 3, the concentration of TSP and its MCP (OC, EC, Fe, Rnon-Fe, RAnion, and RCation) measured at each month on three consecutive (non-rainy) days are presented. These concentration data were used to make other Tables and Figures for analysis of their behaviors. Fig. 2S depicts the monthly mean concentration of TSP and its chemical components at our study site. As seen in Fig. 2S, the abundance in MCP generally decreases in the order: anions, OC, cations, EC, and metals. Although the fraction of Fe and Rnon-Fe was the lowest of all the measured MCP, their monthly patterns were still comparable to the major chemical components of TSP. This observation thus suggests that they should probably

share common sources between each other. To examine this relationship further, the concentrations of all constituents were normalized against TSP (Fig. 3). The contribution of most TSP components (especially anions and OC) showed large variations over time. The relative composition of MCP measured in this study was similar to those measured from another district (Jeon Nong site) in the Seoul metropolitan area by Kim and Kim (2008). These authors investigated the physico-chemical characteristics of aerosol particulate and reported that PM2.5 consisted of mostly ions (43% for spring and 48% for the remaining seasons), whereas their relative contribution to PM10 mass was less significant during the study period (August 2002–February 2005). It thus clearly confirms that the relative proportion of MCP increases with decreasing particle size. This suggests that the majority of the chemical components which were not measured in this study (Al, Si and other crustal components) are present mostly in the coarse fraction. This is to be expected for crustal contributions produced by natural erosion processes. In addition, the contribution of each chemical component of PM was seen to vary greatly across the seasons. As seen in Fig. 3, the relative composition of MCP, if evaluated throughout the study period, is dominated, in decreasing order of concentration: anion, OC, cation, and EC. The dominant role of ion fractions in the TSP levels was also observed recently at Baeg Nyeong island (Kong et al., 2010). These authors measured the chemical compositions of PM10 and PM2.5 from this remote area which is far away from urban sources. They reported that RAnion and RCation values accounted for 38% and 21% of PM10, respectively, although their contribution tends to vary considerably with

392

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

(A) Major TSP chemical components (other than metals) (n=33)

(B) Metal only (Fe vs. Σnon-Fe) (n=33) Fig. 3. Monthly trend of TSP normalized concentrations of PM chemical components (%) at the Yang Jae site.

the varying concentrations of PM10, possibly indicating the presence of several contribution sources in remote locations. 3.3. Factors controlling the major chemical components of particulate matter (MCP) in TSP To learn more about the relationship between MCPs, their temporal trends were analyzed after the TSP-normalization of the monthly mean values. Fig. 3 shows that during this study period relative pattern of MCP was fairly constant through most of the study period prior to Julian day 300. Considering that the relative composition of MCP cannot fully explain all the TSP components, it may be important to consider the relationship between MCP and unidentified particulate components (UPC). To address the relationship between MCP and UPC, the latter is computed crudely as the difference between TSP and MCP. As shown in Fig. 3S, the relative contribution between MCP and UPC can be compared using their concentrations across the seasons. According to this analysis, the MCP can account for 54% (spring) to 67% (fall) of TSP composition. Although our analysis focused on inorganic particulates such as inorganic ions, metals, and EC, the organic material (OM) which can constitute a large fraction of the particles in the atmosphere have not been measured. In addition, we did not measure all elemental components such as major crustal elements (e.g., Si and Al). Therefore, the unclaimed portion of MCP (40%) should represent mostly from these organic bound components as well as some major crustal components.

As a simple means to assess the relationship between TSP and MCP and between the different chemical components of MCPs, we conducted correlation analysis using all the measurement data. Fig. 4 depicts the results of a correlation analysis between the concentrations of TSP and its chemical components (MCP: RAnion, RCation, OC, EC, Fe, and Rnon-Fe) across four seasons. As shown in Fig. 4A, the MCP exhibited strong correlations with TSP during the study period. Note that the correlation analysis was made for all data (n = 33) acquired during this study period. Exceptionally, OC and EC exhibited weak inverse correlations only during winter (slope (n = 6) = 0.0146 (OC) and 0.0418 (EC)). RAnion had the highest slope values (spring = 0.3169, summer = 0.3034, fall = 0.2708, and winter = 0.2492) against TSP across the four seasons. If one considers the correlation between TSP and its major chemical components, ion fractions showed strong correlations ((1) RAnion (n = 33): R2 = 0.793 and p-value = 3.80E12 and (2) RCation (n = 33): R2 = 0.804 and pvalue = 1.69E12). In contrast, metallic components were not so strong as ion species ((1) Fe (n = 33): R2 = 0.387 and p-value = 1.10E-4 and (2) Rnon-Fe (n = 33): R2 = 0.351 and p-value = 2.78E–4). This probably indicates the different types of source for these components. (Anions and cations: secondary aerosol formation from primary emitted gaseous. Metals: primary emission from point sources and non-exhaust traffic emissions.) However, all major chemical components generally showed significant correlations with p-values below 0.01, when all data (n = 33) were put together for the comparison. Such relationships between TSP and the ion fractions were even maintained during summer ((1) RAnion (n = 9): R2 = 0.9525 and p-value = 6.95E6 and (2) RCation (n = 9): R2 = 0.945 and p-value = 1.16E5). As seen in this study, similar patterns between TSP and its major chemical components can also be found from some previous studies. For instance, Wang et al. (2006) collected the TSP aerosol samples at two urban sites (Fudan and Taopu) in Shanghai, China in four seasons from September 2003 to January 2005. Concentrations of 15 ions (10 anions plus 5 cations) in TSP were measured from a total of 202 samples. If the correlation analysis is conducted between the concentration of TSP and ion fractions (RAnion and RCation) measured in this study, the ions had a good correlation with TSP ((1) RAnion (n = 6): R2 = 0.8778 and p-value = 5.84E3 and (2) RCation (n = 6): R2 = 0.6012 and p-value = 7.00E2). In contrast, an opposite trend can also be found from other studies. Fang et al. (2007) collected ambient air samples at Taichung airport site in central Taiwan for the analysis of TSP and nine associated ions (4 anions and 5 cations) during September to December 2005. These authors reported insignificant correlations between concentrations of TSP and ion fractions. The results of correlation analysis between TSP and nine ions showed that the R2 values for 12 samples were 0.0144 (RAnion) and 0.0008 (RCation) with respective p-values of 0.71 and 0.93. As such, comparison of different data sets obtained from different studies suggests the possible existence of different source contributions at different sites and alludes to the presence of complicated situations in which the relationship between the mass concentration of PM and its relative composition shows 9 significant variation.

4. Conclusions In this study, various aspects of the relationship between TSP and its chemical components were evaluated. We confirmed that the MCP measured during this study period accounted for about 60% of TSP. In order to assess the relationship between TSP and its chemical components, their correlations were evaluated. The results indicated that most chemical components of TSP (RAnion, RCation, OC, and EC) maintained good correlations with each other. As expected, the anion and cation compositions were very

393

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

(A) All data measured this study period (n = 33)

(B) Spring (n = 9)

(C) Summer (n = 9)

(D) Fall (n = 9)

(E) winter (n = 6)

Fig. 4. Results of correlation analysis between TSP and its individual chemical components (MCP: RAnion, RCation, OC, EC, Fe, and Rnon-Fe).

strongly correlated in compliance with the requirement for charge balance. The mass concentration of TSP in ambient air during this study was most strongly influenced by the ion fractions, especially anions, relative to the other major chemical components.

In order to evaluate the properties of MCP as a key chemical component of TSP, their temporal trends were evaluated in relation to their contribution to TSP composition. The results of this study confirm that the observed concentrations of TSP are most sensitively affected by ion fractions, although all chemical components

394

Y.-H. Kim et al. / Chemosphere 95 (2014) 387–394

of MCP show a clear relationship with TSP mass concentration. The different patterns of relationship between TSP mass concentration and the MCP chemical components most probably indicate the different sources of these constituents: anions and cations come from secondary aerosol formation from primary emitted gases, whereas metals are from primary emissions from point sources and non-exhaust traffic emissions. If our analysis had determined all quantifiable chemical components of particles, we would have reduced further the unknown portion of the TSP fraction. In addition, it should also be considered that the carbon analyzer (TOT) applied in this study is likely to underestimate EC compared to other techniques (e.g., thermal/ optical reflectance (TOR) method). Although the gap between TSP and its MCP fractions can be narrowed by more complete and rigorous assessment of the latter, the analysis of MCP conducted in this study suggests that the existence of an unquantifiable particulate fraction remains an important factor in understanding the composition of airborne particles and is an issue that all mass closure models need to address. Acknowledgements This research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and Future Planning (Grant No 2013-004624). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.chemosphere. 2013.09.050. References Anthwal, A., Jung, K., Kim, H.J., Bae, I.-S., Kim, K.-H., 2010. Polycyclic aromatic hydrocarbons in ambient air at four urban locations in Seoul, Korea. Fresenius Environ. Bull. 19 (7), 1356–1368. Begum, B.A., Hossain, A., Saroar, G., Biswas, S.K., Nasiruddin, Md., Nahar, N., Chowdury, Z., Hopke, P.K., 2011. Sources of carbonaceous materials in the airborne particulate matter of Dhaka. Asian J. Atmos. Environ. 5, 237–246. Biswas, S., Verma, V., Schauer, J.J., Sioutas, C., 2009. Chemical speciation of PM emissions from heavy-duty diesel vehicles equipped with diesel particulate filter (DPF) and selective catalytic reduction (SCR) retrofits. Atmos. Environ. 43, 1917–1925. Cape, J.N., Cornell, S.E., Jickells, T.D., Nemitz, E., 2011. Organic nitrogen in the atmosphere—where does it come from? A review of sources and methods. Atmos. Res. 102 (1–2), 30–48. Fang, G.-C., Wu, Y.-S., Rau, J.-Y., Huang, S.-H., 2006. Traffic aerosols (18 nm 5 particle size 5 18 lm) source apportionment during the winter period. Atmos. Res. 80 (4), 294–308. Fang, G.-C., Wu, Y.-S., Lee, W.-J., Lin, I.-C., Chou, T.-Y., 2007. Ambient air particulates and related ionic species study at Taichung airport, Taiwan. Environ. Forensics 8, 53–62. Fenger, J., 1999. Urban air quality. Atmos. Environ. 33, 4877–4900.

Greco, S.L., Wilson, A.M., Spengler, J.D., Levy, J.I., 2007. Spatial patterns of mobile source particulate matter emissions-to-exposure relationships across the United States. Atmos. Environ. 41, 1011–1025. Harrison, R.M., Yin, J., 2000. Particulate matter in the atmosphere: which particle properties are important for its effects on health? Sci. Total Environ. 249, 85– 101. Hwang, I.J., Hopke, P.K., 2011. Comparison of source apportionment of PM2.5 using PMF2 and EPA PMF version 2. Asian J. Atmos. Environ. 5, 86–96. Kang, C.-H., Kim, W.-H., Ko, H.-J., Hong, S.-B., 2009. Asian dust effects on total suspended particulate (TSP) compositions at Gosan in Jeju Island, Korea. Atmos. Res. 94 (2), 345–355. Kim, K.-H., 2007a. Airborne lead concentration levels on the Korean peninsula between 1991 and 2004. Atmos. Environ. 41, 809–824. Kim, K.-H., 2007b. Airborne cadmium in the major monitoring locations in Korea between 1991 and 2004. Atmos. Environ. 41, 4380–4395. Kim, Y.P., 2010. Analysis of the trend of atmospheric PM10 concentration over the Seoul Metropolitan Area between 1999 and 2008. Korean Assoc. Environ. Impact Assess. 19, 59–74. Kim, S.-D., Kim, C.-H., 2008. The physico-chemical character of aerosol particle in Seoul Metropolitan Area. Seoul Dev. Inst. 9 (3), 23–33. Kim, K.-H., Shon, Z.-H., 2011. Nationwide shift in CO concentration levels in urban areas of Korea after 2000. J. Hazard. Mater. 188, 235–246. Kong, B.J., Kim, S.K., Park, J.S., Lee, S.D., Kim, J.C., Kim, J.S., Lee, D.W., Lee, S.J., 2010. The characteristics of particle composition at the Baengnyeong Island. J. Korea Soc. Environ. Admin. 16 (2), 97–105. National Institute of Environmental Research (NIER), 2007. Annual Report of Environment in Korea 2007. National Institute of Environmental Research (NIER), 2010. Annual Report of Air Quality in Korea 2009. Nazir, R., Shaheen, N., Shah, M.H., 2011. Indoor/outdoor relationship of trace metals in the atmospheric particulate matter of an industrial area. Atmos. Res. 101 (3), 765–772. Nguyen, H.T., Park, C.G., Kim, J.A., Lee, J.S., Lee, J.H., Kim, K.-H., 2010. Concentrations of TSP-bound metals in four urban residential locations in Seoul, Korea. Sci. World J. 10, 742–759. Pandey, S.K., Kim, K.-H., Chung, S.-Y., Cho, S.-J., Shon, Z.-H., 2008. Long term study of NOx behavior at urban roadside and background locations in Seoul, Korea. Atmos. Environ. 42 (4), 606–621. Pathak, R.K., Wang, T., Wu, W.S., 2011. Nighttime enhancement of PM2.5 nitrate in ammonia-poor atmospheric conditions in Beijing and Shanghai: Plausible contributions of heterogeneous hydrolysis of N2O5 and HNO3 partitioning. Atmos. Environ., 45, pp. 1183–1191. Seaton, A., MacNee, W., Donaldson, K., Godden, D., 1995. Particulate air pollution and acute health effects. Lancet 345, 176–178. Seoul Metropolitan Government, 2010. Seoul Statistical Yearbook. Periodical report, pp. 438–444. Shon, Z.-H., Kim, K.-H., 2011. Impact of emission control strategy on NO2 in urban areas of Korea. Atmos. Environ. 45, 808–812. Shon, Z.-H., Kim, K.-H., Song, S.K., Chae, Y.Z., Park, C.K., Jung, K., 2012. Fractionation of secondary organic carbon in aerosol in the megacity of Seoul, Korea. Atmos. Environ. 50, 225–233. Tan, J., Guo, S., Ma, Y., Duan, J., Cheng, Y., He, K., Yang, F., 2011. Characteristics of particulate PAHs during a typical haze episode in Guangzhou, China. Atmos. Res. 102 (1–2), 91–98. Wang, Y., Zhuang, G., Zhang, X., Huang, K., Xu, C., Tang, A., Chen, J., An, Z., 2006. The ion chemistry, seasonal cycle, and sources of PM2.5 and TSP aerosol in Shanhai. Atmos. Environ. 40, 2935–2952. Woodruff, T.J., Grillo, J., Schoendorf, K.C., 1997. The relationship between selected causes of postneonatal infant mortality and particulate air pollution in the United States. Environ. Health Perspect. 105, 608–612. Zheng, J., Tan, M., Shibata, Y., Tanaka, A., Li, Y., Zhang, G., Zhang, Y., Shan, Z., 2004. Characteristics of lead isotope ratios and elemental concentrations in PM10 fraction of airborne particulate matter in Shanghai after the phase-out of leaded gasoline. Atmos. Environ. 38, 1191–1200.

An investigation into the relationship between the major chemical components of particulate matter in urban air.

Particulate matter (PM) generally comprises such chemical components as inorganic ions, organic carbon (OC), elemental carbon (EC), and metals. In ter...
2MB Sizes 0 Downloads 0 Views