- Email: [email protected]
The study of fine and coarse particles, and metallic elements for the daytime and night-time in a suburban area of central Taiwan, Taichung
Chemosphere 41 (2000) 639±644
The study of ®ne and coarse particles, and metallic elements for the daytime and night-time in a suburban area of central Taiwan, Taichung Guor-Cheng Fang a,*, Cheng-Nan Chang b, Yuh-Shen Wu a, Vicky Wang e, Peter Pi-Cheng Fu c, Ding-Guor Yang b, Shun-Chin Chen b, Chia-Chium Chu d a
Department of Environmental Engineering and Health, Hungkuang Institute of Technology, Sha-Lu, Taichung 433, Taiwan, ROC b Department of Environmental Science, Tunghai University, Taichung 407, Taiwan, ROC c Division of Biochemical Toxicology, National Center for Toxicological Research, Jeerson, AR 72079, USA d Chest Medicine, Chien Yu Hospital, Kaohsiung, Taiwan, ROC e Oce of Secretary, Chief Secretary, Hungkuang Institute of Technology, Sha-Lu, Taichung 433, Taiwan, ROC Received 9 July 1999; accepted 11 October 1999
Abstract Daily average concentrations of ®ne and coarse particulates, and TSP samples have been measured simultaneously at daytime and night-time periods by using Universal and PS-1 sampler in a suburban area of central Taiwan from June to August 1998. The samples were analyzed by atomic absorption spectrometry to determine the ®ne and coarse particulate concentrations of metallic elements (Ca, Fe, Mn, Pb, Cu, Zn and Cr). The concentration of PM2:5 and TSP showed a decreased trend for the daytime period. The ®ne particle concentrations were about two times as that of coarse particulate concentrations. The averaged ®ne particulate concentrations at daytime are higher than at night-time. Ca and Fe were mostly in the coarse particulate mode. The correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night periods. Pb showed a similar distribution ratio with Mn for the ®ne to coarse particle ratios at both day and night period. Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. Cu and Zn were mainly in ®ne particles at both day and night period. Fe and Ca consist of the major parts of all the elements. Elemental Mn is the lowest among the rest of the heavy metals. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Metal analysis; Fine particles; Coarse particle; Air sampling
1. Introduction Atmospheric aerosol particles play an important role in our everyday life and in the control of dierent processes in the air (Preining, 1996). It is well documented that environmental eects of aerosol particles depend on *
Corresponding author. Tel.: +886-46318652, extn: 230; fax: +886-43502102. E-mail address: [email protected] (G.-C. Fang).
their size and chemical composition. Thus, solar radiation transfer in the air, cloud-aerosol interactions and biospheric impacts are all determined by the size distribution of particles of dierent composition (Meszaros et al., 1997). Particulate matter with a 50% cut-o diameter of 10 lm (PM10 ) has been associated in epidemiological studies with increased mortality, morbidity and decreased lung function (Saskia et al., 1998). Lam et al. (1999) reported that PM2:5 concentrations constituted about 70% of the total PM10 . Investigators have sug-
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 5 0 7 - X
640
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
gested that trace metals distributed widely throughout the lung on ®ne particles could catalyze the formation of oxidants within the lung, which in turn produce tissue damage (Ghio et al., 1996; Dreher et al., 1997). Industrial processes, such as metal re®ning and fossil fuel combustion, have lead to a substantial increase of trace elements in the atmosphere (WHO, 1987). Heavy metals such as Pb, Cu and Zn mainly found in the particulate phase (Buat-Menard, 1993). They are mainly emitted into the atmosphere by heavy industry, coal burning, metallurgical smelters and automobile trac (Pacyna, 1986; Nriagu and Pacyna, 1988b). The ever increasing dispersion of heavy metals through the atmosphere, water, and soil is a major concern due to their hazardous eect on human health, the possible changes they initiate in natural biochemical processes in all ecosystems, and their inevitable accumulation in the food chain (Van Malderen et al., 1996). The annual total toxicity of all heavy metals mobilized was reported to exceed the combination toxicity of all the radioactive and organic wastes generated each year (Nriagu and Pacyna, 1988a). To understand the particulate and metal element concentrations after transporting long distances from urban area, the study is intended to characterize ®ne (PM2:5 ), coarse (PM2:5±10 ), and inhalable (PM10 ) concentrations and the metal element particulate size fraction in suburban environment during daytime and night-time period. 2. Experimental 2.1. Sampling site Samples of atmospheric aerosol were collected from June to August in 1998 at a four-story building (12 m height) at Thunghai University (THU). Sampling time was divided into daytime and night-time. The daytime period was de®ned from 8:00 a.m. to 8:00 p.m. and the night-time period was de®ned from 8:00 p.m. to 8:00 a.m. THU, which is located between the urban and the country region, is 12 km from downtown, near to the trac road and surrounded by trees and the campus residents (Fig. 1). Daily dichotomous samples (<2.5 and 2.5±10 lm) were collected on Te¯on ®lters with Universal sampler. TSP (total suspended particulate) samples were concurrently collected on Te¯on ®lters by PS-1 samplers. The Model 310 Universal Air samplerTM (UASTM ) has a design inlet sampling ¯ow rate of 300 l per min (lpm). The sampler is provided with an omnidirectional inlet, a PM10 (10 lm cut) virtual impactor classi®er, either a PM2:5 or PM10 virtual impactor classi®er, a ®ne particle ®lter and a PUF sampler. Air is sampled at 300 lpm (10.6 acfm) from the ambient atmosphere through an omni-directional, cylindrical inlet. Particles greater than 10 lm aerodynamic equivalent
Fig. 1. Map of Taichung suburban sampling site (THU).
diameter are removed from the sampled air stream by the PM10 classi®er and discarded. Particle less than 10 lm ¯ow to the PM2:5 classi®er are located downstream. Particles in the 2.5±10 lm range are collected on a 62 mm ´ 165 mm ®lter and those smaller than 2.5 mm are collected on a 200 mm ´ 250 mm ®nal ®lter. The ®ltered air stream is then directed through the PUF sampler to collect the volatile organic compounds in the ®ltered air stream (Model 310 Universal Air SamplerTM Instruction Manual (USATM), 1996). The PS-1 (GPS1 PUF Sampler, General Metal Work) consists of nine basic assembles: dual chamber, sampling module, ¯ow vent, magnetic gage, voltage variator, elapsed time indicator, pump, seven-day skip timer, exhaust hose and aluminum shelter. The PS-1 is a completed air sampling system designed to simultaneously collect suspended airborne particles at ¯ow rate up to 280 l per min and the ¯ow rate was adjusted to 200 l per min in this study. The Te¯on ®lter paper (diameter 10.2 lm) is used to ®lter the suspended particle in this study. 2.2. Mass measurement In this study, the Te¯on ®lters (62 mm ´ 165 mm and 200 mm ´ 250 mm) were used to collect the coarse (2.5±10 lm) and ®ne (<2.5 lm) particulate concentration, respectively, in central Taiwan. The Te¯on ®lters were previously dried in a desiccator for a 48 h period and weighed to a precision of 10 lg in an analytical balance. After collection, the used ®lters were placed in the desiccator dried for another 48 h period and weighed again in the same analytical balance. The procedure is the same as the previous study (Infante and Acosta, 1991). 2.3. Chemical analysis After sampling, the 62 mm ´ 165 mm ®lters and the 200 mm ´ 250 mm ®lters were cut into appropriate size
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
641
Table 1 Fine, coarse particulate, PM10 and TSP concentrations; PM10 /TSP and ®ne/coarse ratios during daytime, 1998 (lg/m3 ) Date 6/18 6/27 7/2 7/15 7/31 8/3 8/6 8/11 Average
Fine 25.48 50.16 34.62 44.61 26.40 26.54 8.82 26.84 30.43
Coarse 9.94 14.54 9.66 17.29 12.10 8.09 9.70 14.09 11.93
PM10 35.43 64.70 44.28 61.89 38.50 34.63 18.52 40.93 42.36
then digested in 20 ml concentrated nitric acid at 150± 200°C for 2 h. The ®lters were diluted to 25 ml with distilled±deionized water. After ®ltration the solutions were analyzed by A Hitachi Z-5000 series polarized zeeman atomic absorption spectrophotometer. The analysis procedure is the same as the previous study (Fang et al., 1997). Background contamination was routinely monitored by using operational blanks (unexposed ®lters) which were processed simultaneously with ®eld samples. In this study, the background contamination is insigni®cant and can be ignored. Recovery eciencies were calculated and analyzed with at least 10% of the samples spiking with a known amount of metal. The results indicated that the ranges of recovery eciencies among every 10 samples were varied between 95% and 102% and the reproducibility tests were varied between 98% and 105% for all the chemical species.
TSP 47.21 72.61 75.11 64.22 59.20 67.80 52.30 53.91 61.54
PM10 /TSP 0.75 0.89 0.59 0.96 0.65 0.51 0.35 0.76 0.68
Fine/coarse 2.56 3.45 3.58 2.58 2.18 3.28 0.91 1.90 2.56
daytime, and the only exception is also on the sampling date of 6 August, which showed that the TSP concentrations were about three times as that of PM10 concentrations. The ratios of ®ne particle concentrations to coarse particle concentrations displayed that the ®ne particle concentrations are almost greater than that of coarse particle concentrations. Table 2 showed that averaged ®ne particle concentrations were mostly higher than coarse particle concentrations except on the date of 15 July, 3 August, and 6 August. In general, the averaged ratio of PM10 /TSP in the night-time was the same as daytime except the date of 3 August, 6 August and 11 August. The ratios of ®ne particle/coarse particle were averaged 1.51. The averaged ®ne particulate and PM10 concentrations at daytime are higher than at night-time (one-sided P-value <0.05, T-test). However, for coarse particulate and TSP concentrations, there were no signi®cant dierences between day and night period (two-sided P-value <0.05, T-test).
3. Results and discussion 3.2. Mass element composition at day and night period 3.1. Ambient air particle mass concentration in suburban Table 1 indicated that the averaged daytime ®ne particulate concentrations were about two times of coarse particulate concentrations except sampling group on the date of 6 August, which exhibit almost equal amount of ®ne and coarse particulate concentrations. The averaged ratios of PM10 /TSP were about 0.68 at
Figs. 2 and 3 show the average metal elements composition in the ®ne and coarse particle mode during the daytime. The results indicated that the elements Fe and Ca contribute the majority of metals composition for both particle modes. The composition of Cr was highest among the rest of the other heavy metal elements (Pb, Mn, Cu, Zn and Cr). Like daytime period, the
Table 2 Fine, coarse particulate, PM10 and TSP concentrations; PM10 /TSP and ®ne/coarse ratios during night-time, 1998 (lg/m3 ) Date 6/18 6/27 7/2 7/15 7/31 8/3 8/6 8/11 Average
Fine 16.47 29.41 48.51 9.82 30.90 6.04 14.22 15.87 21.41
Coarse 14.33 14.62 16.40 14.43 12.84 13.17 15.40 10.4 13.95
PM10 30.81 44.03 64.91 24.25 43.74 19.21 29.62 26.27 35.36
TSP 42.69 53.48 86.89 48.24 50.07 48.12 63.08 56.3 56.11
PM10 /TSP 0.72 0.82 0.75 0.50 0.87 0.40 0.47 0.47 0.63
Fine/coarse 1.15 2.01 2.96 0.68 2.41 0.46 0.92 1.53 1.51
642
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
Fig. 2. Averaged metallic element composition in coarse particle mode at daytime.
Fig. 5. Averaged metallic element composition in coarse particle mode at night-time.
time (Table 1), which suggests the same anthropogenic sources. However, these phenomena had not occurred in the dayetime. The reasons leading to these phenomena need to be further studied. It indicated that they are from the same origin. Table 2 showed the elements Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. These results indicated that these two elements originated from the same emission sources in the suburban area. The coarse particulates of Cu and Zn were positively correlated (R 0.54) at daytime and positively correlated (R 0.42) at night-time, respectively (Table 4). Fig. 3. Average metallic element composition in ®ne particle mode at daytime.
Fig. 4. Averaged metallic element composition in ®ne particle mode at night-time.
average metal element composition in the ®ne and coarse particle mode during the night-time period was also studied (Figs. 4 and 5). The results also indicated that the crustal elements Fe and Ca consist of the major parts of all the elements. Element Mn is the lowest composition among the rest of the heavy metals. Correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night period, respectively (Table 3). The correlation coecient of ®ne particulate Zn and Mn was 0.53 at night-
4. Conclusions Results of measurements presented in this paper give the following conclusions: 1. The averaged ®ne particle concentrations were mostly higher than coarse particle concentrations for both daytime and night-time in the suburban sampling site. 2. The averaged ratio of PM10 /TSP in the night-time ( 0.68) was about the same as daytime ( 0.63). The averaged ®ne particulate and PM10 concentrations at daytime are higher than at night-time (onesided P-value < 0.05, T-test). 3. As for coarse particulate and TSP concentrations, there were no signi®cant dierences between day and night period (two-sided P-value < 0.05, T-test). 4. Ca and Fe were mostly in the coarse particulate mode and correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night period, respectively. It indicated that they are from the same origin. 5. Pb showed similar distribution ratios with Mn for the ®ne/coarse particle ratios at both day and night period. Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. These results indicated that these two elements originated from the same emission sources in the suburban area.
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
643
Table 3 Correlation matrix of coarse particulate metallic elements in central Taiwan during daytime and night-time period Daytime Ca Fe Mn Pb Cu Zn Cr
Night-time
Ca
Fe
Mn
Pb
Cu
Zn
Cr
Ca
Fe
Mn
Pb
Cu
Zn
Cr
1
0.63 1
)0.47 0.10 1
0.05 0.34 0.38 1
0.14 0.09 0.42 0.32 1
)0.06 )0.65 0.10 0.11 0.54 1
)0.05 0.14 0.07 0.07 0.12 0.02 1
1
0.69 1
0.22 0.16 1
0.36 )0.20 0.78 1
0.02 )0.21 )0.36 )0.27 1
0.04 )0.46 0.30 0.50 0.42 1
0.14 )0.22 0.21 0.23 )0.04 )0.32 1
Table 4 Correlation matrix of ®ne particulate metallic elements in central Taiwan during daytime and night-time period Daytime Ca Fe Mn Pb Cu Zn Cr
Night-time
Ca
Fe
Mn
Pb
Cu
Zn
Cr
Ca
Fe
Mn
Pb
Cu
Zn
Cr
1
0.09 1
)0.37 )0.23 1
)0.4 )0.38 0.78 1
)0.83 )0.20 0.39 0.39 1
0.09 0.27 )0.13 )0.37 )0.78 1
0.03 0.08 )0.38 )0.31 )0.15 0.22 1
1
)0.26 1
)0.20 0.14 1
)0.51 )0.46 0.61 1
0.09 0.19 )0.03 )0.50 1
)0.20 0.39 0.53 )0.01 0.45 1
0.17 0.02 )0.25 )0.53 0.05 0.16 1
6. Cu and Zn were mainly in ®ne particle mode for both day and night period. The coarse particulates of Cu and Zn were positively correlated (R 0.54) at daytime and positively correlated (R 0.42) at nighttime, respectively. 7. The composition of Cr was highest among the rest of the other heavy metal elements (Pb, Mn, Cu, Zn and Cr). Element Mn is the lowest composition among the rest of the heavy metals.
Acknowledgements The ®nancial support provided by the Hungkuang Institute of Technology of Humanities and Science Council, through a research contract (HKHSC-88-01) is gratefully appreciated.
References Buat-Menard, P., 1993. Global change in atmospheric metal cycles. In: Global Atmospheric Chemical Change. Elsevier, Amsterdam, pp. 271±311. Dreher, K.L., Jaskot, R.H., Lehmann, J.R., Richards, J.H., McGee, J.K., Ghio, A.J., Costa, D.L.J., 1997. Toxicol. Environ. Health 50, 285±305. Fang, G.-C., Cheng, M.-T., Chang, C.-N., 1997. Monitoring and modeling the mass, heavy metal and ion species dry
deposition in central Taiwan. J. Environ. Sci. Health A 32 (8), 2183±2199. Ghio, A.J., Stonehuerner, J., Prithchard, R.J., Piantadosi, C.A., Quigley, D.R., Dreher, K.L., Costa, D.L., 1996. Inhalation Toxicol. 8, 479±494. Infante, R., Acosta, I.L., 1991. Size distribution of trace metals in Ponce, Puerto Rico air particulate matter. Atmos. Environ. B 25, 121±131. Lam, C.K., Leung, D.Y.C., Niewiadomski, M., Pang, S.W., Lee, A.W.F., Louie, P.K.K., 1999. Street-level concentrations of nitrogen dioxide and suspended particulate matter in Hong Kong. Atmos. Environ. 33, 1±11. Meszaros, E., Barcza, T., Gelencser, A., Hlavay, J., Kiss, Gy., Krivacsy, Z., Molnar, A., Polyak, K., 1997. Size distributions of inorganic and organic species in the atmospheric aerosol in Hungary. J. Aerosol Sci. 28, 1163±1175. Model 310 Universal Air SamplerTM Instruction Manual (USATM ), 1996. MSP Corporation, 1313 Fifth Street, SE Suite 206 Minneapolis, MN, USA, 55414. Nriagu, J.O., Pacyna, J.M., 1988a. Nature 333, 134±139. Nriagu, J.O., Pacyna, J.M., 1988b. Quantitative assessment of worldwide contamination of air water and soils by race metals. Nature 33, 134±135. Pacyna, J.M., 1986. Atmospheric trace elements from natural and anthropogenic sources. In: Toxic Metals in the Atmosphere. Wiley, New York, pp. 33±52. Preining, O., 1996. The many facets of aerosol science. J. Aerosol Sci. 27, S1±S6. Saskia, C., Van Der Zee, S.C., Hoek, G., Harrssema, H., Brunekreef, B., 1998. Characterization of particulate air pollution in urban and non-urban areas in the Netherlands. Atmos. Environ. 32, 3717±3729.
644
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
Van Malderen, H., Hoornaert, S., Van Grieken, R., 1996. Identi®cation of individual aerosol particles containing Cr, Pb and Zn above the North Sea. Environ. Sci. Technol. 30, 489±498.
WHO (World Health Organization), 1987. Air Quality Guidelines for Europe, WHO Regional Oce for Europe, European Series, vol. 23. WHO Regional Publications, Copenhagen.
The study of ®ne and coarse particles, and metallic elements for the daytime and night-time in a suburban area of central Taiwan, Taichung Guor-Cheng Fang a,*, Cheng-Nan Chang b, Yuh-Shen Wu a, Vicky Wang e, Peter Pi-Cheng Fu c, Ding-Guor Yang b, Shun-Chin Chen b, Chia-Chium Chu d a
Department of Environmental Engineering and Health, Hungkuang Institute of Technology, Sha-Lu, Taichung 433, Taiwan, ROC b Department of Environmental Science, Tunghai University, Taichung 407, Taiwan, ROC c Division of Biochemical Toxicology, National Center for Toxicological Research, Jeerson, AR 72079, USA d Chest Medicine, Chien Yu Hospital, Kaohsiung, Taiwan, ROC e Oce of Secretary, Chief Secretary, Hungkuang Institute of Technology, Sha-Lu, Taichung 433, Taiwan, ROC Received 9 July 1999; accepted 11 October 1999
Abstract Daily average concentrations of ®ne and coarse particulates, and TSP samples have been measured simultaneously at daytime and night-time periods by using Universal and PS-1 sampler in a suburban area of central Taiwan from June to August 1998. The samples were analyzed by atomic absorption spectrometry to determine the ®ne and coarse particulate concentrations of metallic elements (Ca, Fe, Mn, Pb, Cu, Zn and Cr). The concentration of PM2:5 and TSP showed a decreased trend for the daytime period. The ®ne particle concentrations were about two times as that of coarse particulate concentrations. The averaged ®ne particulate concentrations at daytime are higher than at night-time. Ca and Fe were mostly in the coarse particulate mode. The correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night periods. Pb showed a similar distribution ratio with Mn for the ®ne to coarse particle ratios at both day and night period. Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. Cu and Zn were mainly in ®ne particles at both day and night period. Fe and Ca consist of the major parts of all the elements. Elemental Mn is the lowest among the rest of the heavy metals. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Metal analysis; Fine particles; Coarse particle; Air sampling
1. Introduction Atmospheric aerosol particles play an important role in our everyday life and in the control of dierent processes in the air (Preining, 1996). It is well documented that environmental eects of aerosol particles depend on *
Corresponding author. Tel.: +886-46318652, extn: 230; fax: +886-43502102. E-mail address: [email protected] (G.-C. Fang).
their size and chemical composition. Thus, solar radiation transfer in the air, cloud-aerosol interactions and biospheric impacts are all determined by the size distribution of particles of dierent composition (Meszaros et al., 1997). Particulate matter with a 50% cut-o diameter of 10 lm (PM10 ) has been associated in epidemiological studies with increased mortality, morbidity and decreased lung function (Saskia et al., 1998). Lam et al. (1999) reported that PM2:5 concentrations constituted about 70% of the total PM10 . Investigators have sug-
0045-6535/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 9 9 ) 0 0 5 0 7 - X
640
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
gested that trace metals distributed widely throughout the lung on ®ne particles could catalyze the formation of oxidants within the lung, which in turn produce tissue damage (Ghio et al., 1996; Dreher et al., 1997). Industrial processes, such as metal re®ning and fossil fuel combustion, have lead to a substantial increase of trace elements in the atmosphere (WHO, 1987). Heavy metals such as Pb, Cu and Zn mainly found in the particulate phase (Buat-Menard, 1993). They are mainly emitted into the atmosphere by heavy industry, coal burning, metallurgical smelters and automobile trac (Pacyna, 1986; Nriagu and Pacyna, 1988b). The ever increasing dispersion of heavy metals through the atmosphere, water, and soil is a major concern due to their hazardous eect on human health, the possible changes they initiate in natural biochemical processes in all ecosystems, and their inevitable accumulation in the food chain (Van Malderen et al., 1996). The annual total toxicity of all heavy metals mobilized was reported to exceed the combination toxicity of all the radioactive and organic wastes generated each year (Nriagu and Pacyna, 1988a). To understand the particulate and metal element concentrations after transporting long distances from urban area, the study is intended to characterize ®ne (PM2:5 ), coarse (PM2:5±10 ), and inhalable (PM10 ) concentrations and the metal element particulate size fraction in suburban environment during daytime and night-time period. 2. Experimental 2.1. Sampling site Samples of atmospheric aerosol were collected from June to August in 1998 at a four-story building (12 m height) at Thunghai University (THU). Sampling time was divided into daytime and night-time. The daytime period was de®ned from 8:00 a.m. to 8:00 p.m. and the night-time period was de®ned from 8:00 p.m. to 8:00 a.m. THU, which is located between the urban and the country region, is 12 km from downtown, near to the trac road and surrounded by trees and the campus residents (Fig. 1). Daily dichotomous samples (<2.5 and 2.5±10 lm) were collected on Te¯on ®lters with Universal sampler. TSP (total suspended particulate) samples were concurrently collected on Te¯on ®lters by PS-1 samplers. The Model 310 Universal Air samplerTM (UASTM ) has a design inlet sampling ¯ow rate of 300 l per min (lpm). The sampler is provided with an omnidirectional inlet, a PM10 (10 lm cut) virtual impactor classi®er, either a PM2:5 or PM10 virtual impactor classi®er, a ®ne particle ®lter and a PUF sampler. Air is sampled at 300 lpm (10.6 acfm) from the ambient atmosphere through an omni-directional, cylindrical inlet. Particles greater than 10 lm aerodynamic equivalent
Fig. 1. Map of Taichung suburban sampling site (THU).
diameter are removed from the sampled air stream by the PM10 classi®er and discarded. Particle less than 10 lm ¯ow to the PM2:5 classi®er are located downstream. Particles in the 2.5±10 lm range are collected on a 62 mm ´ 165 mm ®lter and those smaller than 2.5 mm are collected on a 200 mm ´ 250 mm ®nal ®lter. The ®ltered air stream is then directed through the PUF sampler to collect the volatile organic compounds in the ®ltered air stream (Model 310 Universal Air SamplerTM Instruction Manual (USATM), 1996). The PS-1 (GPS1 PUF Sampler, General Metal Work) consists of nine basic assembles: dual chamber, sampling module, ¯ow vent, magnetic gage, voltage variator, elapsed time indicator, pump, seven-day skip timer, exhaust hose and aluminum shelter. The PS-1 is a completed air sampling system designed to simultaneously collect suspended airborne particles at ¯ow rate up to 280 l per min and the ¯ow rate was adjusted to 200 l per min in this study. The Te¯on ®lter paper (diameter 10.2 lm) is used to ®lter the suspended particle in this study. 2.2. Mass measurement In this study, the Te¯on ®lters (62 mm ´ 165 mm and 200 mm ´ 250 mm) were used to collect the coarse (2.5±10 lm) and ®ne (<2.5 lm) particulate concentration, respectively, in central Taiwan. The Te¯on ®lters were previously dried in a desiccator for a 48 h period and weighed to a precision of 10 lg in an analytical balance. After collection, the used ®lters were placed in the desiccator dried for another 48 h period and weighed again in the same analytical balance. The procedure is the same as the previous study (Infante and Acosta, 1991). 2.3. Chemical analysis After sampling, the 62 mm ´ 165 mm ®lters and the 200 mm ´ 250 mm ®lters were cut into appropriate size
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
641
Table 1 Fine, coarse particulate, PM10 and TSP concentrations; PM10 /TSP and ®ne/coarse ratios during daytime, 1998 (lg/m3 ) Date 6/18 6/27 7/2 7/15 7/31 8/3 8/6 8/11 Average
Fine 25.48 50.16 34.62 44.61 26.40 26.54 8.82 26.84 30.43
Coarse 9.94 14.54 9.66 17.29 12.10 8.09 9.70 14.09 11.93
PM10 35.43 64.70 44.28 61.89 38.50 34.63 18.52 40.93 42.36
then digested in 20 ml concentrated nitric acid at 150± 200°C for 2 h. The ®lters were diluted to 25 ml with distilled±deionized water. After ®ltration the solutions were analyzed by A Hitachi Z-5000 series polarized zeeman atomic absorption spectrophotometer. The analysis procedure is the same as the previous study (Fang et al., 1997). Background contamination was routinely monitored by using operational blanks (unexposed ®lters) which were processed simultaneously with ®eld samples. In this study, the background contamination is insigni®cant and can be ignored. Recovery eciencies were calculated and analyzed with at least 10% of the samples spiking with a known amount of metal. The results indicated that the ranges of recovery eciencies among every 10 samples were varied between 95% and 102% and the reproducibility tests were varied between 98% and 105% for all the chemical species.
TSP 47.21 72.61 75.11 64.22 59.20 67.80 52.30 53.91 61.54
PM10 /TSP 0.75 0.89 0.59 0.96 0.65 0.51 0.35 0.76 0.68
Fine/coarse 2.56 3.45 3.58 2.58 2.18 3.28 0.91 1.90 2.56
daytime, and the only exception is also on the sampling date of 6 August, which showed that the TSP concentrations were about three times as that of PM10 concentrations. The ratios of ®ne particle concentrations to coarse particle concentrations displayed that the ®ne particle concentrations are almost greater than that of coarse particle concentrations. Table 2 showed that averaged ®ne particle concentrations were mostly higher than coarse particle concentrations except on the date of 15 July, 3 August, and 6 August. In general, the averaged ratio of PM10 /TSP in the night-time was the same as daytime except the date of 3 August, 6 August and 11 August. The ratios of ®ne particle/coarse particle were averaged 1.51. The averaged ®ne particulate and PM10 concentrations at daytime are higher than at night-time (one-sided P-value <0.05, T-test). However, for coarse particulate and TSP concentrations, there were no signi®cant dierences between day and night period (two-sided P-value <0.05, T-test).
3. Results and discussion 3.2. Mass element composition at day and night period 3.1. Ambient air particle mass concentration in suburban Table 1 indicated that the averaged daytime ®ne particulate concentrations were about two times of coarse particulate concentrations except sampling group on the date of 6 August, which exhibit almost equal amount of ®ne and coarse particulate concentrations. The averaged ratios of PM10 /TSP were about 0.68 at
Figs. 2 and 3 show the average metal elements composition in the ®ne and coarse particle mode during the daytime. The results indicated that the elements Fe and Ca contribute the majority of metals composition for both particle modes. The composition of Cr was highest among the rest of the other heavy metal elements (Pb, Mn, Cu, Zn and Cr). Like daytime period, the
Table 2 Fine, coarse particulate, PM10 and TSP concentrations; PM10 /TSP and ®ne/coarse ratios during night-time, 1998 (lg/m3 ) Date 6/18 6/27 7/2 7/15 7/31 8/3 8/6 8/11 Average
Fine 16.47 29.41 48.51 9.82 30.90 6.04 14.22 15.87 21.41
Coarse 14.33 14.62 16.40 14.43 12.84 13.17 15.40 10.4 13.95
PM10 30.81 44.03 64.91 24.25 43.74 19.21 29.62 26.27 35.36
TSP 42.69 53.48 86.89 48.24 50.07 48.12 63.08 56.3 56.11
PM10 /TSP 0.72 0.82 0.75 0.50 0.87 0.40 0.47 0.47 0.63
Fine/coarse 1.15 2.01 2.96 0.68 2.41 0.46 0.92 1.53 1.51
642
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
Fig. 2. Averaged metallic element composition in coarse particle mode at daytime.
Fig. 5. Averaged metallic element composition in coarse particle mode at night-time.
time (Table 1), which suggests the same anthropogenic sources. However, these phenomena had not occurred in the dayetime. The reasons leading to these phenomena need to be further studied. It indicated that they are from the same origin. Table 2 showed the elements Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. These results indicated that these two elements originated from the same emission sources in the suburban area. The coarse particulates of Cu and Zn were positively correlated (R 0.54) at daytime and positively correlated (R 0.42) at night-time, respectively (Table 4). Fig. 3. Average metallic element composition in ®ne particle mode at daytime.
Fig. 4. Averaged metallic element composition in ®ne particle mode at night-time.
average metal element composition in the ®ne and coarse particle mode during the night-time period was also studied (Figs. 4 and 5). The results also indicated that the crustal elements Fe and Ca consist of the major parts of all the elements. Element Mn is the lowest composition among the rest of the heavy metals. Correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night period, respectively (Table 3). The correlation coecient of ®ne particulate Zn and Mn was 0.53 at night-
4. Conclusions Results of measurements presented in this paper give the following conclusions: 1. The averaged ®ne particle concentrations were mostly higher than coarse particle concentrations for both daytime and night-time in the suburban sampling site. 2. The averaged ratio of PM10 /TSP in the night-time ( 0.68) was about the same as daytime ( 0.63). The averaged ®ne particulate and PM10 concentrations at daytime are higher than at night-time (onesided P-value < 0.05, T-test). 3. As for coarse particulate and TSP concentrations, there were no signi®cant dierences between day and night period (two-sided P-value < 0.05, T-test). 4. Ca and Fe were mostly in the coarse particulate mode and correlation coecients were 0.63 and 0.69 for elements Ca and Fe in the coarse particle mode for day and night period, respectively. It indicated that they are from the same origin. 5. Pb showed similar distribution ratios with Mn for the ®ne/coarse particle ratios at both day and night period. Pb and Mn are highly correlated for the day (R 0.78) and night period (R 0.61) at particle size <2.5 lm. These results indicated that these two elements originated from the same emission sources in the suburban area.
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
643
Table 3 Correlation matrix of coarse particulate metallic elements in central Taiwan during daytime and night-time period Daytime Ca Fe Mn Pb Cu Zn Cr
Night-time
Ca
Fe
Mn
Pb
Cu
Zn
Cr
Ca
Fe
Mn
Pb
Cu
Zn
Cr
1
0.63 1
)0.47 0.10 1
0.05 0.34 0.38 1
0.14 0.09 0.42 0.32 1
)0.06 )0.65 0.10 0.11 0.54 1
)0.05 0.14 0.07 0.07 0.12 0.02 1
1
0.69 1
0.22 0.16 1
0.36 )0.20 0.78 1
0.02 )0.21 )0.36 )0.27 1
0.04 )0.46 0.30 0.50 0.42 1
0.14 )0.22 0.21 0.23 )0.04 )0.32 1
Table 4 Correlation matrix of ®ne particulate metallic elements in central Taiwan during daytime and night-time period Daytime Ca Fe Mn Pb Cu Zn Cr
Night-time
Ca
Fe
Mn
Pb
Cu
Zn
Cr
Ca
Fe
Mn
Pb
Cu
Zn
Cr
1
0.09 1
)0.37 )0.23 1
)0.4 )0.38 0.78 1
)0.83 )0.20 0.39 0.39 1
0.09 0.27 )0.13 )0.37 )0.78 1
0.03 0.08 )0.38 )0.31 )0.15 0.22 1
1
)0.26 1
)0.20 0.14 1
)0.51 )0.46 0.61 1
0.09 0.19 )0.03 )0.50 1
)0.20 0.39 0.53 )0.01 0.45 1
0.17 0.02 )0.25 )0.53 0.05 0.16 1
6. Cu and Zn were mainly in ®ne particle mode for both day and night period. The coarse particulates of Cu and Zn were positively correlated (R 0.54) at daytime and positively correlated (R 0.42) at nighttime, respectively. 7. The composition of Cr was highest among the rest of the other heavy metal elements (Pb, Mn, Cu, Zn and Cr). Element Mn is the lowest composition among the rest of the heavy metals.
Acknowledgements The ®nancial support provided by the Hungkuang Institute of Technology of Humanities and Science Council, through a research contract (HKHSC-88-01) is gratefully appreciated.
References Buat-Menard, P., 1993. Global change in atmospheric metal cycles. In: Global Atmospheric Chemical Change. Elsevier, Amsterdam, pp. 271±311. Dreher, K.L., Jaskot, R.H., Lehmann, J.R., Richards, J.H., McGee, J.K., Ghio, A.J., Costa, D.L.J., 1997. Toxicol. Environ. Health 50, 285±305. Fang, G.-C., Cheng, M.-T., Chang, C.-N., 1997. Monitoring and modeling the mass, heavy metal and ion species dry
deposition in central Taiwan. J. Environ. Sci. Health A 32 (8), 2183±2199. Ghio, A.J., Stonehuerner, J., Prithchard, R.J., Piantadosi, C.A., Quigley, D.R., Dreher, K.L., Costa, D.L., 1996. Inhalation Toxicol. 8, 479±494. Infante, R., Acosta, I.L., 1991. Size distribution of trace metals in Ponce, Puerto Rico air particulate matter. Atmos. Environ. B 25, 121±131. Lam, C.K., Leung, D.Y.C., Niewiadomski, M., Pang, S.W., Lee, A.W.F., Louie, P.K.K., 1999. Street-level concentrations of nitrogen dioxide and suspended particulate matter in Hong Kong. Atmos. Environ. 33, 1±11. Meszaros, E., Barcza, T., Gelencser, A., Hlavay, J., Kiss, Gy., Krivacsy, Z., Molnar, A., Polyak, K., 1997. Size distributions of inorganic and organic species in the atmospheric aerosol in Hungary. J. Aerosol Sci. 28, 1163±1175. Model 310 Universal Air SamplerTM Instruction Manual (USATM ), 1996. MSP Corporation, 1313 Fifth Street, SE Suite 206 Minneapolis, MN, USA, 55414. Nriagu, J.O., Pacyna, J.M., 1988a. Nature 333, 134±139. Nriagu, J.O., Pacyna, J.M., 1988b. Quantitative assessment of worldwide contamination of air water and soils by race metals. Nature 33, 134±135. Pacyna, J.M., 1986. Atmospheric trace elements from natural and anthropogenic sources. In: Toxic Metals in the Atmosphere. Wiley, New York, pp. 33±52. Preining, O., 1996. The many facets of aerosol science. J. Aerosol Sci. 27, S1±S6. Saskia, C., Van Der Zee, S.C., Hoek, G., Harrssema, H., Brunekreef, B., 1998. Characterization of particulate air pollution in urban and non-urban areas in the Netherlands. Atmos. Environ. 32, 3717±3729.
644
G.-C. Fang et al. / Chemosphere 41 (2000) 639±644
Van Malderen, H., Hoornaert, S., Van Grieken, R., 1996. Identi®cation of individual aerosol particles containing Cr, Pb and Zn above the North Sea. Environ. Sci. Technol. 30, 489±498.
WHO (World Health Organization), 1987. Air Quality Guidelines for Europe, WHO Regional Oce for Europe, European Series, vol. 23. WHO Regional Publications, Copenhagen.