Arch. Environ. Contam. Toxicol. 47, 162–167 (2004) DOI: 10.1007/s00244-004-3054-9
A R C H I V E S O F
Environmental Contamination a n d Toxicology © 2004 Springer-Verlag New York, LLC
Heavy Metals in Airborne Particulate Matter of Urban Coimbatore R. Mohanraj, P. A. Azeez, T. Priscilla Environmental Impact Assessment Division, Salim Ali Centre for Ornithology and Natural History, Anaikatty, Coimbatore-641 108, India
Received: 28 April 2003 /Accepted: 25 January 2004
Abstract. Exposures to airborne metals are known to cause physiological responses in organisms and wide-ranging health effects in humans. Hence determination of metals in particulate matter is important from a toxicological perspective. In the current study heavy metals associated with respirable (RSPM) and nonrespirable (NRSPM) fractions of suspended particulate matter were estimated in air samples from six stations in Coimbatore, India, during March 1999 to February 2001. The mean quantity of heavy metals in RSPM was in the order Zn ⬎ Cu ⬎ Pb ⬎ Ni ⬎ Cr ⬎ Cd. Concentrations of these heavy metals were in the range of BDL (below detectable level) to 2147 ng/m3 in RSPM. The highest level of lead (2147 ng/m3) was recorded at an industrial station. The station also had the highest mean value (481 ⫾ 544.3 ng/m3), suggesting the importance of industrial operations in determining the ambient concentrations of lead. Significant positive correlation among metals excepting lead and copper suggests that they originate mostly from a common source. Air samples of urban and industrial areas showed higher concentrations than residential (Urban) and suburban areas.
Particulate matter (PM) in atmosphere on entering the respiratory system is reported to cause wide-ranging health effects including cancer and heart failures (Wichmann et al. 2000; Pope et al. 2002; Lee and Schwartz 1999; Dockery et al. 1992; Dockery and Pope 1994; Dockery 2001). Heavy metals associated with PM have a definite influence on the biological functions affecting the normal development and growth of body tissues and their proper functioning (Fergusson 1990; Dasilva and Williams 1991). PM of size less than 10 m (PM10 or respirable suspended particulate matter [RSPM]) contains high concentrations of heavy metals of toxicological interest (Rizzio et al. 1999). About 75–90% of metals such as Cu, Cd, Ni, Zn, and Pb are found in the PM 10 fraction. Among the PM10 fraction, the alveolar fraction (0 –1.1 m) contained 50 –70% of these metals, followed by the bronchial fraction (1.1– 4.6 m) and tracheopharyngeal fraction (4.6 –9 m). Although many metals are normal constituents of tissue, metals
Correspondence to: P. A. Azeez; email:
[email protected]
such as arsenic, antimony, lead, cadmium, mercury, and bismuth are known to be toxic even at low levels. For the past few decades, elevated levels of metals and their compounds, both inorganic and organic, have been released to the environment as a result of a variety of anthropogenic activities. Main sources of heavy metals include various industrial sources such as mining activities, foundries and smelters, and diffuse sources such as combustion, traffic, and piping (Irwin 1997). In India due to increasing traffic, unplanned urban and industrial development, growing energy consumption, and the high influx of population to urban areas, alarming levels of particulate matter are reported in urban atmospheres (Agarwal and Narain 1999). The number of urban agglomerations/cities in India with populations over a million increased from 5 in 1951 to 9 in 1971 and 23 in 1991. The vehicle population grew from 0.3 million in 1951 to almost 40 million in 1997–1998, more than a 100-fold increase. Despite the increasing levels of particulate matter, information on PM10-associated metals in Indian atmosphere is meagre. Coimbatore, an important industrial city of India, is also a rapidly growing principal urban agglomeration, with an urban population of 14 lakhs. Currently there are 312 medium- and large-scale textile mills with 5 million spindles employing more than 65,604 workers in and around the city (SIMA 2000). There also are about 37,000 registered small-scale industries employing more than a lakh of workers. The industrial sector of the district is fast-growing and occupies about 7.84% of the land in urban Coimbatore. The total number of vehicles registered in Coimbatore has grown to about 500,000. This increasing trend of automobiles in Coimbatore is suspected to be a major contributing factor to the current elevated levels of particulate matter (136.5–206.5 g/m3) in the city (Mohanraj 2002), compared to earlier reported levels (40.7– 65.7 g/m3 [Agarwal and Narain 1999]). In the present scenario it was felt interesting to study the airborne metals in the Coimbatore atmosphere, which are largely unknown.
Methodology Suspended particulate matter (SPM) was collected using a respirable dust sampler (Model APM 451; Envirotech, India) at six sampling stations (A–F, Figure 1) during March 1999 to February 2001 at monthly intervals. Sampling stations were selected to represent urban– commercial (Stations A and B) activities, urban–residential (C), in-
Heavy Metals in Airborne Particulate Matter
163
Fig. 1. Map showing sampling locations
dustrial (Station D), suburban highway (Station E), and suburban–rural mixed (Station F) activities. The respirable fraction (RSPM or PM10) of the SPM was collected on preweighed Whatman glass-microfiber filter paper (GF/A; 20.3 ⫻ 25.4 cm). SPM bigger than 10 m (nonrespirable SPM; NRSPM) was collected in separate sampling bottles. The sampler was run for 24 h at an average airflow rate of 1.15 m3/min. The glass-fiber filter carrying RSPM was cut into small pieces of 5.3 cm2 and transferred into Teflon-jacketed digestion vials. To this, 8 ml of nitric acid, 2 ml of hydrofluoric acid, and 1 ml of perchloric acid were added, and the mixture was loaded onto a microwave digestion system (Milestone Model MLS 1200 and exhaust module EM 45) and digested at 150°C for 20 min. After cooling, the digested solutions were filtered using Whatman filter paper (Grade 1), transferred to plastic measuring cylinders, made up to 15 ml with double-distilled water, and analyzed for lead, copper, chromium, cadmium, zinc, and nickel using an atomic absorption spectrometer (Perkin Elmer Model 3300). In the case of NRSPM, known weights of dust samples were digested in 3 ml of concentrated nitric acid and 1 ml of perchloric acid and left overnight. The contents were then heated to about 100°C for 1 h, cooled, and filtered after rinsing with 3 ml 1:1 nitric acid. This mixture was made up to volume with double-distilled water and the contents were transferred to clean polyethylene bottles and analyzed for Cu, Zn, Pb, and Cd using the atomic
absorption spectrometer. The detection limits of the instrument in a solution of comparable chemical constituents were 0.01 ppm Pb, 0.002 ppm Cu and Cr, 0.009 ppm Zn, and 0.0006 ppm Cd. No recovery test had been performed using a comparable standard, although analysis with the addition of known metal concentrations to select samples gave conforming results.
Results and Discussion Annual averages of RSPM and NRSPM at various stations during the study period are shown in Table 1. The mean quantum of heavy metals in RSPM examined in the present study was in the order Zn ⬎ Cu ⬎ Pb ⬎ Ni ⬎Cr ⬎ Cd, ranging from below detectable limits (BDL) to 1131 ng/m3 (Table 2). RSPM-bound metal levels in the current findings are comparable to the values in a similar study by Chelani et al. (2001) in Mumbai (Table 2). The significant positive correlation among some of the metals suggests that they originate from a common source (Table 3). The abundances of metals associated with NRSPM were in the following order: Cu ⬎ Zn ⬎ Pb ⬎ Cd, with the following ranges: Cu, 9 –550.6 ng/m3; Pb, 2.2–254.8 ng/m3; Zn, 5.7–760 ng/m3; and Cd, BDL–2.8 ng/m3.
164
R. Mohanraj et al.
Table 1. Mean RSPM and NRSPM during March 1999 to February 2001 at sampling stations Sampling station
RSPM (g/m3)
NRSPM (g/m3)
SBI Road—A Gandhipuram—B Saibabacolony—C SIDCO—D Peelamaedu—E Kuniamuthur—F
84.24 ⫾ 7.98 84.4 ⫾ 16.4 59.7 ⫾ 13.1 77.37 ⫾ 23.6 71.3 ⫾ 22.07 51.73 ⫾ 17.46
93.6 ⫾ 14.09 122 ⫾ 39 119.45 ⫾ 101.4 123.6 ⫾ 72.24 116.6 ⫾ 76.05 84.82 ⫾ 69.7
Lead Overall lead concentrations in RSPM during the study period across the sampling stations ranged between 6.7 and 2147 ng/m3 (Figure 2). Among the stations, Station E (industrial area) recorded the highest concentration, 2147 ng/m3. The station also had the highest mean value (481 ⫾ 544.3 ng/m3). An average of 81% lead is found in the RSPM fraction of the total SPM, though it varied between 56 and 91.3% among stations, with Station D having a high fraction of lead in RSPM. Lead levels showed significant variation among the stations (p ⬍ 0.001), while within stations no significant variation was seen between the first and the second year of the study period or among various seasons. At station A, lead levels had a significant positive relationship with windspeed (r ⫽ 0.5, p ⬍ 0.05) and NRSPM levels (r ⫽ 0.51, p ⬍ 0.01). This probably denotes resuspension of road dust as a major source of lead. At Station B, which had a mean lead concentration of 149.2 ⫾ 76 ng/m3, no significant relationship was found with meteorological factors. At Station C, the lead concentration ranged between 16.2 and 133.2 ng/m3, with an average of 43.2 ng/m3. No association of lead was found with any of the meteorological parameters at Station C. The highest lead concentration (2147 ng/m3) was recorded at SIDCO (Station D). Station D also had the highest mean value (481 ⫾ 544.3 ng/m3). Lead levels in RSPM appear to exceed the USEPA standards for lead (1500 ng/m3) during 3 months in the study period at Station D. Wide monthly variation in lead levels at Station D indicates the importance of industrial operations in determining the ambient concentrations of lead. At SIDCO, except for the temperature (minimum), which had a negative correlation with lead levels (r ⫽ ⫺0.45, p ⬍ 0.01), no other meteorological factors were found to be influential. An average of 81.4 ⫾ 56.9 ng/m3 of lead is found at Station E, with a significant correlation with RSPM values (r ⫽ 0.6, P ⬍ 0.01). Station F recorded the lowest mean lead level compared other stations and also had relatively less urban activity and vehicular movement. At Station F, lead values showed a significant positive relationship with RSPM (r ⫽ 0.66, p ⬍ 0.01) and negative relationship with windspeed (r ⫽ ⫺0.42, p ⬍ 0.05). The overall pattern of lead distribution shows industrial operation as a major source at SIDCO, while in other places vehicular emission may have a major role. Leaded petrol causes about 90% of airborne lead pollution in cities, the rest of which comes from factories, power plants, lead pipes, and lead-based paints (Down to Earth 1997). A study in five megacities in India showed that among 1852 children tested, 51.4% had blood lead levels above 10 g/dl. The percentage of children having 10 g/dl or higher blood
lead levels ranged from 39.9% in Bangalore to 61.8% in Mumbai (CPCB 2000).
Copper Copper concentration in RSPM samples of stations ranged between 43.8 and 973 ng/m3 (Figure 3). The mean station wise concentrations were found in the order: Station A ⬎ Station B ⬎ Station D ⬎ Station E ⬎ Station C ⬎ Station F. The highest level was recorded at Station E (973 ng/m3) during November 1999, while the lowest value was at Station F during November 2001. ANOVA of copper showed significant variations among stations and seasons (p ⬍ 0.001). In the case of copper on average 79% of the metal occurred in the RSPM fraction of the total SPM. The urban samples of Stations A and B had comparatively higher copper in the RSPM fraction (about 85%), indicating vehicular traffic as an important source for copper. As the volume of traffic was less at Stations C and F, the copper concentrations also declined. Overall, copper levels, irrespective of station, showed increasing values with increased RSPM concentrations (r ⫽ 0.58, p ⬍ 0.01), probably indicating RSPM as a major repository for copper. Copper had positive correlations with other metals such as Cr and Cd (r ⫽ 0.29, p ⬍ 0.05, and r ⫽ 0.36, p ⬍ 0.05, respectively).
Chromium Overall chromium values in RSPM samples ranged from BDL to 87 ng/m3 (Figure 4). The mean concentration in the study area was 14.2 ⫾ 14.1 ng/m3. The highest Cr content was noted in the industrial area (during October 2001), followed by urban– commercial, suburban highway, urban, urban–residential, and suburban areas (Station D ⬎ B ⬎ E ⬎ A ⬎ C ⬎ F). Higher concentrations of Cr in the industrial area are probably attributable to chrome plating in industries. Cr was below detectable limits at Stations C and F. ANOVA showed significantly high variation among stations and the 2 consecutive study years (p ⬍ 0.001) but not among the seasons.
Cadmium In TSPM an average of 86% of cadmium is present in RSPM, without much variation between samples. The distribution of cadmium in RSPM samples was in the range of BDL–9.1 ng/m3 (Figure 5). The stationwise mean concentrations of Cd, in decreasing order, are as follows: D ⬎ A ⬎ B ⬎ E ⬎ C ⬎ F (industrial ⬎ urban ⬎ suburban Highway, Urban—residential ⬎ suburban). Cd concentrations were below detectable limits at Stations C, E, and F. The highest value (9.1 ng/m3) of Cd was recorded at Station D (industrial) during February 2002. ANOVA of cadmium concentrations showed that variation was significant among stations (p ⬍ 0.001), while it was not among seasons or the 2 study years. An earlier report (Krishnamurti and Viswanathan 1991), found a Cd concentration in Tamilnadu state (Coimbatore is the second-largest district of Tamilnadu) varying between 0.004 and 0.02 g/m3,
Heavy Metals in Airborne Particulate Matter
165
Table 2. Mean metal concentrations (ng/m3) at sampling stations Metal
Mean
SD
Range in present study
Range in Mumbai (Chelani et al. 2001)
Lead Copper Chromium Cadmium Zinc Nickel
143.5 388.6 14.2 2.8 519.9 31.37
271.7 218.3 14.1 1.5 179.76 15.29
6–2147 43.8–973 0–87.3 0–9.1 122–1131 6.5–86.3
80–1170 — 8–180 1–180 — —
Table 3. Pearson’s correlation coefficient between metals
Lead Copper Cadmium Chromium Zinc Nickel
Lead
Copper
Cadmium
Chromium
Zinc
Nickel
1.00 0.15 0.46** 0.58** 0.03 0.18
1.00 0.36** 0.29** 0.23 0.21
1.00 0.47** 0.29* 0.22*
1.00 0.08 0.05
1.00 0.35
1.00
Note. Significance levels: ** p ⬍ 0.01; * p ⬍ 0.05.
Fig. 2. Lead distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximum values, respectively
with a mean concentration of 0.006 g/m3 in TSPM. Cd values of the current study (BDL– 0.009 g/m3) also fall in this range.
Zinc Zinc in RSPM ranged between 122.2 and 1131 ng/m3, with a mean concentration of 519.19 ⫾ 79.76 ng/m3 (Figure 6). The stationwise mean concentrations in decreasing order is as follows: A ⬎ B ⬎ E ⬎ D ⬎ F ⬎ C. The highest (1131 ng/m3) and
Fig. 3. Copper distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximum values, respectively
the lowest values were recorded at Station B and Station F during the months of July and March, respectively. ANOVA showed significant variation in zinc values among stations (p ⬍ 0.001), while the variation was not significant between the 2 study years or the seasons. A positive correlation was observed between RSPM and zinc associated with RSPM, probably indicating RSPM as a major carrier of zinc (r ⫽ 0.32, p ⬍ 0.01). No significant association was found between meteorological parameters and zinc values in RSPM. In an earlier report (Krishnamurti and Viswanathan
166
Fig. 4. Chromium distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximum values, respectively
Fig. 5. Cadmium distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximum values, respectively
1991), in SPM samples from Tamilnadu state, zinc ranged between 0.016 and 0.838 g/m3. The current levels of zinc in RSPM of Coimbatore (0.122–1.131 g/m3) are comparatively higher.
R. Mohanraj et al.
Fig. 6. Zinc distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximum values, respectively
Fig. 7. Nickel distribution at Stations A–F. The bottom boundary of each box indicates the 25th percentile, lines within the box mark the median and mean, and the top boundary of the box indicates the 75th percentile. Whiskers (error bars) above and below the box indicate the 90th and 10th percentile mean and outlying points. Black circles closest to and farthest from zero indicate the minimum and maximumvalues, respectively
Nickel Nickel in RSPM varied between 6.5 and 86.3 ng/m3, with an average concentration of 31.37 ⫾ 15.29 ng/m3 (Figure 7). The stationwise mean concentration was in the order: A ⬎ B ⬎ D ⬎ E ⬎ C ⬎ F. The highest value (86.3 ng/m3) was recorded at Station A during December 1999, while the lowest was recorded at Station F during April 1999, clearly indicating the
Heavy Metals in Airborne Particulate Matter
urban–rural difference. ANOVA of nickel levels showed significant variation among stations and between the 2 years (p ⬍ 0.001). Nickel bound to RSPM is positively related to RSPM (r ⫽ 0.36, p ⬍ 0.01). In India, according to Krishnamurti and Viswanathan (1991) the highest concentration of Ni (6.7 g/ m3) was recorded at Uttar Pradesh during winter. Some other significant figures were 3.2 g/m3 from Chandigargh, 2.34 g/m3 from Andhra Pradesh, and around 1 g/m3 from Tamilnadu, Bihar, and Punjab. In all other states the highest concentration observed was less than 1 g/m3.
Conclusion Heavy metals in the respirable fraction of particulate matter in urban— commercial and industrial areas of Coimbatore showed higher concentrations than in residential (Urban) and suburban areas. The significant positive correlation among metals in RSPM, excepting lead and copper, suggests that mostly they originate from a common source. In general it may be concluded that the atmospheric concentrations of heavy metals recorded in the current study fall in the normal range reported elsewhere in urban areas. However, with the current pace of vehicular growth and urban development, metals such as lead, zinc, and copper may rise to higher levels. Certain metals like chromium, cadmium, and zinc showed an increasing trend compared to the earlier study in Coimbatore.
References Agarwal A, Narain S (1999) The citizen’s fifth report. Part II. Statistical database. Centre for Science and Environment, New Delhi Chelani AB, Gajghate DG, Hasan MZ (2001) Airborne toxic metals in air of Mumbai City, India. Bull Environ Contam Toxicol 66:196 – 205 CPCB (2000) Air quality status and trends in India, National Ambient
167
Air Quality Monitoring Series: NAAQMS/14/1999 –2000. Central Pollution Control Board, New Delhi Dasilva FJJR, Williams RJP (1991) The biochemistry of elements. Clarendon Press, Oxford Dockery DW (2001) Epidemiological evidence of cardiovascular effects of particulate air pollution. Environ Health Persp 109:483– 486 Dockery DW, Pope CA III (1994) Acute respiratory effects of particulate air pollution. Annu Rev Public Health 15:107–132 Dockery DW, Schwartz J, Spengler JD (1992) Air pollution and mortality: Associations with particulates and acid aerosols. Environ Res 59:362–373 Down to Earth (1997) Leading to pollution. Centre for Science and Environment, New Delhi Fergusson JE (1990) The heavy elements: Chemistry, environmental impact and health effects. Pergamon Press, Oxford Krishnamurti CR, Viswanathan P (1991) Toxic metals in Indian environment. Tata McGraw–Hill, New Delhi Lee JT, Schwartz J (1999) Reanalysis of the effects of air pollution on daily mortality in Seoul, Korea: A case-crossover design. Environ Health Persp 107:633– 636 Mohanraj R (2002) Air quality of Coimbatore with emphasis on respirable suspended particulate matter. PhD thesis, Bharathiar University, Coimbatore, India Pope CA III, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, Thurston GD (2002) Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution. JAMA 287: 1132–1141 Rizzio E, Giaveri G, Arginelli D, Gini L, Profumo A, Galorini M (1999) Trace elements total content and particle sizes distribution in the air particulate matter of a rural residential area in north Italy investigated by neutron activation analysis. Sci Total Environ 226:47–56 SIMA (2000) Directory of South Indian Mill Owners Association (SIMA) Wichmann HE, Spix C, Tuch T, Wolke G, Peters A, Heinrich J, Kreyling WJ, Heyder J (2000) Daily mortality and fine and ultrafine particles in Erfurt, Germany. Part I. Role of particle number and particle mass. Health Effects Institute Report No. 98, pp 5–93
