Environ. Sci. Technol. 2008, 42, 6318–6323
Bangladesh and Vietnam: Different Groundwater Compositions Require Different Approaches to Arsenic Mitigation STEPHAN J. HUG, OLIVIER X. LEUPIN, AND MICHAEL BERG Swiss Federal Institute of Aquatic Science and Technology (Eawag)
THIS IMAGE AND TOC GRAPHIC BY SAMUEL LUZI, EAWAG
To be successful, the mitigation strategy must take into account the geological differences in groundwater, the economic resources of the population, and the availability of infrastructure for water treatment.
For more than 10 years it has been known that 30-50 million people in West Bengal and Bangladesh are dependent on drinking water with high levels of arsenic. In Vietnam, high arsenic concentrations in groundwater were discovered nine years ago. How do the two countries cope with this common health threat? Where are similarities, where are differences?
Geology and groundwater: similar but not the same At first glance (Figure 1), the geographical settings of the Bengal, Red River, and Mekong deltas are similar. However, there are major local differences that have important consequences for arsenic mitigation. Bangladesh is largely located on the world’s largest delta plain, on sediments originating from the Himalayan mountain range and deposited over millions of years. From 22,000-12,000 years ago, when sea levels fell to 135 m below the present level, meandering rivers formed a landscape of extended terraced valleys. With rising sea 6318
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 17, 2008
levels, younger sediments consisting of sands, silts, and clays have filled the valleys and lower terraces and now cover most of the delta. The depth of these younger (Holocene) layers varies between 0 and 250 m, but generally slopes down from north to south. The deeper and older (Pleistocene) sediments are typically brown-to-orange, with arsenic tightly bound to iron(hydr)oxides. In contrast, the Holocene sediments are gray-to-black and contain groundwater with high arsenic concentrations. Dissolved arsenic concentrations often show a distinct peak of several hundred µg/L at 25-50 m depth. While there is agreement that arsenic release from naturally present As-bearing iron(hydr)oxides under reductive conditions is the main source of dissolved arsenic (1, 2), open questions remain about the exact mechanisms and dynamics of arsenic mobilization and the influence of recent human activities such as large-scale pumping of groundwater for irrigation (3). Due to the underlying geology, the risk for the population in urban centers is distinctly different from that in rural regions (Table 1). Dhaka, with a population of over 10 million, is located on a Pleistocene terrace with largely arsenic-free groundwater. The affected population lives predominantly in rural areas, villages, and in smaller cities. In Vietnam elevated arsenic is found in aquifers of the Red River and the Mekong River deltas. Where Pleistocene sediments reach the surface in the upper Red River delta, groundwater arsenic concentrations are largely below 10 µg/L (WHO limit). The Holocene sediments are partly delimited from the Pleistocene by a layer of clay and/or silt. Lenses of organic material (peat) are embedded in the sediments in the south of Hanoi, causing NH4 concentrations in the groundwater to reach up to 50 mg/L (4). The Vietnamese part of the Mekong delta consists of up to eight aquifers (5). The often brackish or saline upper aquifer consisting of quaternary sediments has typically a high content of organic matter (up to 23% (6)) and is strongly reducing. Public drinking water is pumped from the lower and older aquifer (Neogene) at depths of 150-250 m. No arsenic has been detected in this deep water so far (5). The Cambodian floodplain associated with higher arsenic levels in the groundwater has been described (7) as unconsolidated Holocene and Plio-Pleistocene sediments overlaying a basaltic bedrock. Regions at risk in Cambodia are the lowland alluvial areas. The strongly reducing groundwaters in the upper aquifers of the Red River and Mekong delta floodplain have high concentrations of As, Fe, and Mn. The moderately elevated levels of iron in the Mekong region are comparable to those found in Bangladesh, while the average iron concentrations in groundwaters of the Red River delta are considerably higher (Table 1).
The water supply situation in Bangladesh In the large cities, groundwater is treated in central water treatment plants. In rural areas, the water supply is highly decentralized, and more than 12 million tubewells provide 10.1021/es7028284
2008 American Chemical Society
Published on Web 08/28/2008
FIGURE 1. Three arsenic-affected alluvial delta regions: B: Bengal delta (Ganges, Brahmaputra, and Meghna rivers); R: Red River delta (Vietnam); and M: Mekong delta (Vietnam and Cambodia). Bangladesh with 144,000 km2 has a population of 147 million; Vietnam with 330,000 km2 has a population of 84 million. (inset) The district map of Bangladesh with average As concentrations in shallow tubewells (10-90 m deep) shows that the South-Central districts are the most affected. In the yellow and red colored districts combined (with a population of 24.5 million), 94% (red) and 79% (red and yellow) of the 10-90 m deep wells have more than 50 µg/L As. (Calculated from the BGS-DPHE database (2)). drinking water from a depth of 15-60 m (2). The Bangladeshi government, NGOs, and international organizations have made great efforts to reduce arsenic exposure. The most successful program so far has been the testing of more than 5 million wells, which has led 29% of the affected population to switch from unsafe to safe wells (8). The second measure, reaching 12% of the affected population, has been the construction of deep tubewells (8). In spite of these successes, 57% of the population is still at risk. Five other mitigation approaches (arsenic removal, dug wells, pond sand filtration, rainwater collection, and piped water systems) are so far each reaching less than 1% of the population (8).
The situation in Vietnam In 2001, an article in ES&T reported arsenic contamination of groundwater in Vietnam (4). The Red River delta of Vietnam has a population of some 11 million people. In rural areas, people changed the source of drinking water 11-14 years ago, and nowadays they consume groundwater that is pumped through individual tubewells (9). In the city of Hanoi, groundwater has been used for >100 years for public water supply. Hanoi has a strongly increasing water demand due to rapid population and economic
growth. Water treatment facilities exclusively exploit the lower (Pleistocene) aquifer containing variable levels of dissolved arsenic, iron, and ammonium (10) (Table 1). Today, 10 major well fields supply water to city treatment facilities, which in 2005 processed 610,000 m3 of water per day. Arsenic at concentrations that can exceed 100 µg/L is partly removed in the public treatment plants to lower, but not always fully acceptable, levels (4). The Mekong delta has become one of the most densely populated and productive areas in Asia. Around 10,000 km2 of the delta area is Cambodian and 52,000 km2 is Vietnamese territory. The number of people pumping contaminated groundwater (Table 1) for drinking has been estimated at 0.5-1 million (5). According to the Vietnamese authorities and UNICEF, 20% of the tubewells in the Mekong delta have As levels exceeding 10 µg/L.
Options for arsenic removal from drinking water Bangladesh: low natural iron and high phosphate concentrations require additional iron for arsenic removal. International research efforts have resulted in a much improved understanding of the reactions that are important for arsenic removal and in the development of promising VOL. 42, NO. 17, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6319
TABLE 1. Comparison of the Average Groundwater Compositions and Conditions in the Three Deltas in Bangladesh, Vietnam, and Cambodia groundwater composition
Bangladesh marked red
pHinitial HCO3- (mM) Ca (mM) Mg (mM) Si (mg/L) Fe (mg/L) P (mg/L) As (µg/L) As > 50 µg/L As > 10 µg/L Mn (mg/L) NH4-N (mg/L)
7.0 ( 0.2 8.0 ( 2.3 1.4 ( 1.2 0.9 ( 0.8 21 ( 6 3.7 ( 5.4 0.7 ( 1.2 62 ( 127 27% 44% 0.6 ( 0.8 2.0 ( 1.0
people at risk potential risk area large cities affected affected aquifers mitigation
30-50 Mio 10.2 Mio 70% none upper (Holocene) aquifer well switching, deep wells
2.0 ( 1.5 1.5 ( 0.8 19 ( 5 4.2 ( 3.8 1.9 ( 1.8 301 ( 200 94% >98% 0.5 ( 0.6 -
Red River delta b
Mekong delta Vietnam d
Mekong delta Cambodia e
7.0 ( 0.4 8.4 ( 2.9 1.9 ( 0.9 0.9 ( 0.4 17 ( 5 13.7 ( 10.6 0.8 ( 0.7 159 ( 418 c 48% c 72% c 0.6 ( 0.6 c 13.5 ( 17.6
6.8 ( 0.6 3.8 ( 2.5 1.7 ( 2.0 2.6 ( 2.9 20 ( 9 2.6 ( 7.4 0.3 ( 0.9 39 ( 128 12% 27% 3.4 ( 5.9 5.0 ( 7.1
6.9 ( 0.4 5.5 ( 2.5 1.1 ( 0.9 1.0 ( 0.9 20 ( 7 2.2 ( 3.3 0.5 ( 0.7 150 ( 276 40% 49% 0.7 ( 0.7 4.9 ( 9.1
11 Mio 50% Hanoi two shallow aquifers sand filter
0.5-1 Mio 30% none three shallow aquifers surface water
a Concentrations in shallow (defined here as 10-90 m deep) tubewells (2988 of 3534) from the BGS-DPHE database (2). From ref 9. c From ref 4. d From ref 5. e From ref 7.
and inexpensive water treatment units. The simplest method takes advantage of the naturally present dissolved Fe(II), by coprecipitation of arsenic with the brown hydrous ferric oxides (HFO) that form within 15-60 min when pumped groundwater comes in contact with air. HFO with sorbed As can be removed by settling or by filtration through simple sand filters (Figure 2). In regions with sufficient natural iron, this method works well. However, several studies have shown (11-13) that the natural iron concentrations in the affected regions in Bangladesh are generally too low to remove more than 50% of the arsenic. High As(III) concentrations combined with low iron and high phosphate and silicate concentrations are a difficult mix for arsenic removal. As(III) sorbs relatively weakly on precipitating HFO at circumneutral pH and is out-competed by the strongly sorbing phosphate and by more weakly sorbing but abundant silicate and carbonate. Generally, efficient arsenic removal requires oxidation of As(III) to the strongly sorbing As(V). To optimally use naturally present Fe(II) and to avoid the need for chemical oxidants, alternative methods have been evaluated. For example, photochemical As(III) oxidation with sunlight leads to significantly higher arsenic removal (40-90%) than coprecipitation alone (14). However, even after complete oxidation of As(III) with hypochlorite, addition of iron (e.g., FeCl3) to an Fe/As ratio of >40 (mg/mg) or 53 (M/M) is typically needed (11) to lower As to below 50 µg/L (current limit in Bangladesh), because the ratio of iron and phosphate is unfavorable in the most affected districts of Bangladesh (Figure 3). Arsenic removal is minimal as long as there is free phosphate. At neutral pH, a molar ratio of 1.5-2.0 Fe per P is needed to remove phosphate and only Fe in excess of this ratio is available to remove As(III) and As(V). Silicate and carbonate also exert a negative effect by competing with arsenic sorption, while calcium has a positive effect by increasing phosphate sorption and precipitation together with iron. However, the effects of calcium, silicate, and carbonate are more difficult to quantify, and since their concentrations do not differ as much in different regions, As/Fe and P/Fe ratios are the most important factors for arsenic removal. In the worst-affected districts of Bangladesh (Figure 1) arsenic removal with natural Fe(II) is clearly insufficient (Figure 3). Unfortunately, with the high percentage of affected 6320
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 17, 2008
M. BERG
b
Bangladesh average a
FIGURE 2. Arsenic removal with natural or added Fe(II): in aerated groundwater, dissolved Fe(II) and partly dissolved As(III) are oxidized and Fe(III) forms sparsely soluble brown Fe(III)(hydr)oxides with sorbed arsenic. In contrast to the weakly sorbing As(III) (H3AsO3), As(V) and phosphate anions (H2AsO4-/ HAsO42- and H2PO4-/HPO42-, respectively) sorb strongly to Fe(III)(hydr)oxide particles, which can be removed by settling or by filtration. Photos in the background show a passive precipitation/ settling tank (top) and a sand filter unit (bottom), both applied in Vietnam. tubewells, the possibility for well switching is limited (Table S1), and the installation of deep wells is also more difficult than in the less affected districts. In the southern districts, depths of typically more than 180 m are required to reach brown sediments, and there is often a narrow depth range where arsenic concentrations are low while manganese concentrations and salinity are not too high. In regions with
FIGURE 3. (a) Average molar arsenic (also listed in µg/L), phosphate, and iron concentrations in Bangladesh, Vietnam, and Cambodia (standard deviations are indicated by black lines). (b) As, Fe, and P ratios. In groundwater with high phosphate and silicate concentrations, Fe/As molar ratios (blue bars) over ∼54 are typically needed for efficient arsenic removal. Phosphate binds strongly to precipitating hydrous ferric oxides, and only the iron remaining after phosphate removal (beige bars show the remaining Fe/As ratios) is available for arsenic removal. limited options for alternative water sources, optimization of water treatment remains an urgent task. It has been found that addition of Fe(II) leads to better As(III) removal than addition of Fe(III), as oxidation of Fe(II) in aerated water produces reactive intermediates that lead to partial oxidation of As(III) to As(V) (13). Repetitive or continuous Fe(II) addition can lead to complete As(III) oxidation and removal by sorption on HFO, without added oxidants (15). A convenient and inexpensive source of Fe(II) is metallic iron (16, 17). In filters containing zerovalent iron, Fe(II) and HFO are continuously formed by corrosion of iron in contact with aerated water (15). Simple filters with iron in various forms, for example, filings, turnings, or nails, have shown great promise for some years in Bangladesh (18, 19) and Nepal (20). The SONO filters with composite iron matrix, built and promoted by Prof. Hussam and Dr. Munir, have recently won the prestigious $1 million Gold Award of the Grainger Prize as the best of 15 tested arsenic removal technologies (21). More than 30,000 SONO filters have been distributed in Bangladesh. Using several kilograms of iron and sand, the filters can treat 120 L of water per day to below
10 µg/L for up to several years. The filters cost around $35 per unit and are practically maintenance-free. Further studies will be needed to understand the influence of different water compositions, as phosphate and other ions influence arsenic removal with metallic iron (16, 22). The Grainger Challenge (www.nae.edu/nae/grainger.nsf) was conducted with water containing 150 µg/L As(III), 150 µg/L As(V), and 2 mg/L Fe(II) without competing anions. Arsenic can also be removed by adsorption on prefabricated columns or by addition of oxidants and coagulants. The Grainger Silver Prize of $200,000 was awarded to a group of researchers led by Prof. A. K. SenGupta in conjunction with the nonprofit organization Water For People (23). Their technology is based on alumina columns, which combine the advantage of coprecipitation of arsenic with HFO by promoting oxidation of dissolved iron at the top of the column, followed by adsorption onto activated alumina, which also acts as a barrier for the arsenicladen HFO. The columns can be regenerated after 8-12 months with caustic soda and sulfuric acid (24). The Bronze Award went to Gregory S. Allgood of the Children’s Safe Drinking Water program at Procter & Gamble, for small VOL. 42, NO. 17, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6321
packets with calcium hypochlorite and iron sulfate designed to treat 10 L of water (23). Vietnam: High natural iron concentrations facilitate arsenic removal. High natural iron concentrations in the Red River delta facilitate arsenic removal. People in the Vietnamese countryside started to use sand filters (Figure 3) to remove the high levels of dissolved iron (10 mg/L tastes metallic or “fishy”. The sand filters also significantly lower arsenic concentrations during the precipitation of the iron on the sand surface. In a study conducted by Berg et al. (9), more than 90% percent of the investigated sand filters reduced the arsenic concentration below 50 µg/L and 40% even reduced the concentration below the WHO guideline of 10 µ/L. Phosphate concentrations >2.5 mg/L slightly reduce the efficiency of sand filters, but the natural iron concentration in most wells is sufficiently high to remove both phosphate and arsenic (Figure 3). Less dissolved iron is present in groundwater of the Mekong delta (see Table 1), which restricts the potential use of arsenic removal by sand filters in these regions.
Conclusions Vietnam and Bangladesh are both confronted with high arsenic concentrations, but distinct water compositions require different solutions. Arsenic mitigation depends for the most part on natural factors, such as the availability of alternative water sources and the feasibility of water treatment. If several options are available, socioeconomic factors determine which mitigation option is implemented most successfully. The socially accepted and already widespread sand filters in the Red River delta have advantages for their simplicity and low cost of operation. The removal of iron from the pumped water is immediately apparent even to people who are not aware of the arsenic problem. Thus, sand filters are a good option in Vietnam and in other affected regions with high concentrations of dissolved iron. Arsenic removal in the worst-affected districts of Bangladesh is considerably more difficult. Since there are currently no selective sorbents, both arsenic and phosphate have to be removed and fixed-bed columns will require frequent regeneration or replacement. Activated alumina columns that can be regenerated have shown very good results. Filter columns with zerovalent iron are very promising, as metallic iron is inexpensive, widely available, and capable of forming precipitates with very high sorption site densities. An improved understanding of the reactions over long periods of operation can lead to further optimization and wider applicability. An issue that is often discussed is the sludge produced in water treatment units. Sludge with elevated arsenic concentrations needs to be collected and handled properly. Containment under oxic conditions or in closed disposal sites are good solutions (25). However, the quantities of arsenic in water used for drinking are small compared to the amounts of arsenic pumped into rice fields by irrigation and probably partly remobilized during monsoon flooding. In the long term, controlled transport and release of treatment sludge into large rivers during high water levels, ensuring rapid dilution and transport into the ocean, could be studied as an alternative to containment. Several mitigation options are now available and should be implemented to avoid further exposure to arsenic-tainted drinking water.
Supporting Information Available Table S1, Vital statistics of Bangladesh and Vietnam; Table S2, Arsenic, iron, and phosphate concentrations and ratios in all districts of Bangladesh; Figure S3, Map of districts with 6322
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 42, NO. 17, 2008
arsenic concentrations. This information is available free of charge via the Internet at http://pubs.acs.org. Stephan J. Hug and Michael Berg are senior research scientists at Eawag. Olivier X. Leupin was a temporary research scientist at Eawag at the time this manuscript was written, and he is now Project Manager for Safety Analysis at Nagra, Wettingen, Switzerland. Address correspondence about this article to Hug at
[email protected].
Literature Cited (1) Nickson, R. T.; McArthur, J. M.; Ravenscroft, P.; Burgess, W. G.; Ahmed, K. M. Mechanism of arsenic release to groundwater, Bangladesh and West Bengal. Appl. Geochem. 2000, 15, 403–413. (2) British Geological Survey, DPHE. Arsenic contamination of groundwater in Bangladesh. In British Geological Survey Technical Report WC/00/19; 4 Volumes; Kinniburgh, D. G., Smedley, P. L., Eds.; British Geological Survey: Keyworth, U.K., 2001. (3) Harvey, C. F.; Swartz, C. H.; Badruzzaman, A. B. M.; KeonBlute, N.; Yu, W.; Ali, M. A.; Jay, J.; Beckie, R.; Niedan, V.; Brabander, D.; Oates, P. M.; Ashfaque, K. N.; Islam, S.; Hemond, H. F.; Ahmed, M. F. Arsenic mobility and groundwater extraction in Bangladesh. Science 2002, 298, 1602–1606. (4) Berg, M.; Tran, H. C.; Nguyen, T. C.; Pham, H. V.; Schertenleib, R.; Giger, W. Arsenic contamination of groundwater and drinking water in Vietnam: A human health threat. Environ. Sci. Technol. 2001, 35, 2621–2626. (5) Berg, M.; Stengel, C.; Trang, P. T. K.; Viet, P. H.; Sampson, M. L.; Leng, M.; Samreth, S.; Fredericks, D. Magnitude of arsenic pollution in the Mekong and Red River Deltas-Cambodia and Vietnam. Sci. Total Environ. 2007, 372, 413–425. (6) Husson, O.; Verburg, P. H.; Phung, M. T.; Van Mensvoort, M. E. F. Spatial variability of acid sulphate soils in the Plain of Reeds, Mekong delta, Vietnam. Geoderma 2000, 97, 1–19. (7) Buschmann, J.; Berg, M.; Stengel, C.; Sampson, M. L. Arsenic and manganese contamination of drinking water resources in Cambodia: Coincidence of risk areas with low relief topography. Environ. Sci. Technol. 2007, 41, 2146–2152. (8) Ahmed, M. F.; Ahuja, S.; Alauddin, M.; Hug, S. J.; Lloyd, J. R.; Pfaff, A.; Pichler, T.; Saltikov, C.; Stute, M.; van Geen, A. Epidemiology-Ensuring safe drinking water in Bangladesh. Science 2006, 314, 1687–1688. (9) Berg, M.; Luzi, S.; Trang, P. T. K.; Viet, P. H.; Giger, W.; Stuben, D. Arsenic removal from groundwater by household sand filters: Comparative field study, model calculations, and health benefits. Environ. Sci. Technol. 2006, 40, 5567–5573. (10) Dodd, M. C.; Vu, N. D.; Ammann, A.; Le, V. C.; Kissner, R.; Pham, H. V.; Cao, T. H.; Berg, M.; Von Gunten, U. Kinetics and mechanistic aspects of As(III) oxidation by aqueous chlorine, chloramines, and ozone: Relevance to drinking water treatment. Environ. Sci. Technol. 2006, 40, 3285–3292. (11) Meng, Z. G.; Korfiatis, G. P.; Christodoulatos, C.; Bang, S. Treatment of arsenic in Bangladesh well water using a household coprecipitation and filtration system. Water Res. 2001, 35, 2805– 2810. (12) Meng, X. G.; Korfiatis, G. P.; Bang, S. B.; Bang, K. W. Combined effects of anions on arsenic removal by iron hydroxides. Toxicol. Lett. 2002, 133, 103–111. (13) Roberts, L. C.; Hug, S. J.; Ruettimann, T.; Billah, M.; Khan, A. W.; Rahman, M. T. Arsenic removal with iron(II) and iron(III) waters with high silicate and phosphate concentrations. Environ. Sci. Technol. 2004, 38, 307–315. (14) Hug, S. J.; Canonica, L.; Wegelin, M.; Gechter, D.; Von Gunten, U. Solar oxidation and removal of arsenic at circumneutral pH in iron containing waters. Environ. Sci. Technol. 2001, 35, 2114– 2121. (15) Leupin, O. X.; Hug, S. J. Oxidation and removal of arsenic (III) from aerated groundwater by filtration through sand and zerovalent iron. Water Res. 2005, 39, 1729–1740. (16) Su, C. M.; Puls, R. W. Arsenate and arsenite removal by zerovalent iron: Effects of phosphate, silicate, carbonate, borate, sulfate, chromate, molybdate, and nitrate, relative to chloride. Environ. Sci. Technol. 2001, 35, 4562–4568. (17) Nikolaidis, N. P.; Dobbs, G. M.; Lackovic, J. A. Arsenic removal by zero-valent iron: field, laboratory and modeling studies. Water Res. 2003, 37, 1417–1425. (18) Hussam, A.; Munir, A. K. M. A simple and effective arsenic filter based on composite iron matrix: Development and deployment studies for groundwater of Bangladesh. J. Environ. Sci. Health, Part A 2007, 42, 1869–1878.
(19) Leupin, O. X.; Hug, S. J.; Badruzzaman, A. B. M. Arsenic removal from Bangladesh tube well water with filter columns containing zerovalent iron filings and sand. Environ. Sci. Technol. 2005, 39, 8032–8037. (20) Neku, A.; Tandukar, N. An overview of arsenic contamination in groundwater of Nepal and its removal at household level. J. Phys. IV 2003, 107, 941–944. (21) Chatterjee, R. Chemist wins Grainger Challenge for Sustainability. Environ. Sci. Technol. 2007, 41, 2660. (22) Tyrovola, K.; Nikolaidis, N. P.; Veranis, N.; Kallithrakas-Kontos, N.; Koulouridakis, P. E. Arsenic removal from geothermal waters with zero-valent iron-Effect of temperature, phosphate and nitrate. Water Res. 2006, 40, 2375–2386.
(23) Ritter, S. K. Chemist Wins Gold In Million-Dollar Arsenic Challenge. Chem. Eng. News 2007, 85, 19; Feb 12. (24) Sarkar, S.; Gupta, A.; Biswas, R. K.; Deb, A. K.; Greenleaf, J. E.; SenGupta, A. K. Well-head arsenic removal units in remote villages of Indian subcontinent: Field results and performance evaluation. Water Res. 2005, 39, 2196–2206. (25) Sarkar, S.; Blaney, L. M.; Gupta, A.; Ghosh, D.; SenGupta, A. K. Arsenic Removal from Groundwater and Its Safe Containment in a Rural Environment: Validation of a Sustainable Approach. Environ. Sci. Technol. 2008, 42, 4268–4273.
ES7028284
VOL. 42, NO. 17, 2008 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
6323
