Article pubs.acs.org/JAFC
Both Phosphorus Fertilizers and Indigenous Bacteria Enhance Arsenic Release into Groundwater in Arsenic-Contaminated Aquifers Tzu-Yu Lin,† Chia-Cheng Wei,† Chi-Wei Huang,† Chun-Han Chang,† Fu-Lan Hsu,§ and Vivian Hsiu-Chuan Liao*,† †
Department of Bioenvironmental Systems Engineering, National Taiwan University, No. 1, Sec. 4, Roosevelt Road, Taipei 106, Taiwan § Forest Chemistry Division, Taiwan Forestry Research Institute, 53 Nanhai Road, Taipei 100, Taiwan ABSTRACT: Arsenic (As) is a human carcinogen, and arsenic contamination in groundwater is a worldwide public health concern. Arsenic-affected areas are found in many places but are reported mostly in agricultural farmlands, yet the interaction of fertilizers, microorganisms, and arsenic mobilization in arsenic-contaminated aquifers remains uncharacterized. This study investigates the effects of fertilizers and bacteria on the mobilization of arsenic in two arsenic-contaminated aquifers. We performed microcosm experiments using arsenic-contaminated sediments and amended with inorganic nitrogenous or phosphorus fertilizers for 1 and 4 months under aerobic and anaerobic conditions. The results show that microcosms amended with 100 mg/L phosphorus fertilizers (dipotassium phosphate), but not nitrogenous fertilizers (ammonium sulfate), significantly increase aqueous As(III) release in arsenic-contaminated sediments under anaerobic condition. We also show that concentrations of iron, manganese, potassium, sodium, calcium, and magnesium are increased in the aqueous phase and that the addition of dipotassium phosphate causes a further increase in aqueous iron, potassium, and sodium, suggesting that multiple metal elements may take part in the arsenic release process. Furthermore, microbial analysis indicates that the dominant microbial phylum is shifted from α-proteobacteria to β- and γ-proteobacteria when the As(III) is increased and phosphate is added in the aquifer. Our results provide evidence that both phosphorus fertilizers and microorganisms can mediate the release of arsenic to groundwater in arsenic-contaminated sediments under anaerobic condition. Our study suggests that agricultural activity such as the use of fertilizers and monitoring phosphate concentration in groundwater should be taken into consideration for the management of arsenic in groundwater. KEYWORDS: arsenic, groundwater, sediment, phosphate fertilizers, microorganisms, agricultural activity solubility, mobility, bioavailability, and toxicity of arsenic.21,22 Microorganisms transform arsenic by oxidizing As(III) or reducing As(V), with reactions including energy generation and detoxification.22,23 Anaerobic bacteria reduce As(V), coupled to the metabolism of environmentally relevant compounds, such as iron and sulfur.22 Aerobic microorganisms convert As(III) to As(V) as a detoxification reaction or for the reducing power to fix CO2.19 However, the interaction between the abiotic and biotic factors that participate in arsenic solubility is still not fully understood. Nevertheless, arsenic transformation and sorption/ desorption processes have been previously described. In abiotic transformation, As(V) is reduced to As(III) as
1. INTRODUCTION Arsenic (As), a known human carcinogen, is widely distributed in food, water, soil, and air.1,2 It is released into water through both biotic and abiotic conditions. In most cases, arseniccontaminated groundwater is derived naturally from arsenicrich aquifer sediments.3 Much evidence has shown that arseniccontaining groundwater for drinking, household, and agricultural uses causes major health problems for humans in many places around the world, such as Bangladesh, India, China, and Taiwan.3−8 Epidemiologic studies have shown that arsenic exposure is associated with skin, liver, bladder, and other cancers.9 The World Health Organization recommends arsenic guideline values for drinking water of ≤10 μg/L.10 However, arsenic concentrations in drinking water in many countries often exceed this limit.11 Arsenic can be solubilized in groundwater through natural geological reactions,3 and various arsenic mobilization mechanisms involved in biotic or abiotic processes have been proposed, including oxidation of arsenic-rich pyrite,12 reductive dissolution of iron oxyhydroxide phases,13,14 competition of solutes for sorption sites on iron oxides,15 arsenic-rich mineral weathering,16−18 and microbial reduction of sorbed arsenate (As(V)) to the potentially more mobile arsenite (As(III)) under anoxic conditions.19,20 A number of studies have suggested that microorganisms play a significant role in arsenic species transformation, leading to significant differences of © 2016 American Chemical Society
H3AsO3 + H 2O → HAsO4 2 − + 4H+ + 2e−
In biotransformation, a strictly anaerobic bacterium, Chrysiogenes arsenatis, has been found to reduce As(V) to As(III) by using acetate as the electron donor and carbon source:24 Received: Revised: Accepted: Published: 2214
January 21, 2016 February 27, 2016 March 3, 2016 March 3, 2016 DOI: 10.1021/acs.jafc.6b00253 J. Agric. Food Chem. 2016, 64, 2214−2222
Article
Journal of Agricultural and Food Chemistry acetate− + 2HAsO4 2 − + 2H 2AsO4 − + 5H+ → 4H3AsO3 + 2HCO3−
Manning and Goldberg25 used a one-site model to describe arsenic adsorption in oxide minerals (SOH): 2SOH + H3AsO4 ↔ (SO)2 HAsO4 + 2H 2O
In Taiwan, arsenic-contaminated aquifers typically exist in the southwestern coast area, where it is historically associated with endemic cases of blackfoot disease (BFD). BFD is a peripheral vascular disease caused by long-term ingestion of arsenic-contaminated groundwater.26 There have been only limited studies that examine factors influencing arsenic’s speciation and mobility in the aquifers of the BFD endemic area. Recently, the potential mechanisms in controlling the release of arsenic species using sediments from arseniccontaminated shallow alluvial aquifers in the BFD region have been described.27 The sediments were collected at 20 m depth (dark gray fine sandy loam) at a location in the southern Zhuoshui River alluvial fan of Taiwan (23°34′07″ N, 120°10′04″ E).27 It was indicated that bacteria can release arsenic into groundwater from sediments without the addition of electron donors, suggesting that arsenic mobilization was not limited by the availability of electron donors in the arseniccontaminated sediments.27 Moreover, it has been suggested that trace elements, such as arsenic, cadmium, and lead, are present in phosphorus fertilizers that might pose potential environmental risks in agricultural areas.28 However, no trace element release data from the fertilizers were reported. Arsenic-affected areas in many places were reported mostly in agricultural farmlands, yet the interaction of fertilizers, microorganisms, and arsenic mobilization in arsenic-contaminated aquifers remains uncharacterized. The BFD endemic region of Taiwan is generally characterized by agricultural fields that have heavy application of fertilizers; therefore, to gain further insight into the factors influencing sedimentary arsenic mobilization in the BFD region, we used the BFD endemic area as an example to investigate the influence of addition of inorganic nitrogenous or phosphorus fertilizers in arsenic mobilization in an arsenic-contaminated aquifer. In addition, we examined multiple elements (Fe, Mn, K, Na, Ca, As, Cd, Mg, Cr, Cu, Ni, Zn, P, and S) in the microcosms associated with arsenic mobilization as the association of high levels of arsenic and multiple metal elements in groundwater has been previously described.29−31 Finally, bacterial communities in different sediment incubations were analyzed by clone library, 16S rRNA PCR methods, and phylogenetic analysis.
Figure 1. Map showing sampling sites ASYL718 (▲) and ASYC160 (■). been previously described.35,36 The Chianan Plain has been reported to be one of the most highly arsenic-contaminated areas in Taiwan.35,37 To collect the sediment samples, a borehole was drilled. A drill rig and a split-tube sampler with PVC liner (50 mm outside diameter) were used to collect sediment cores. Cores were extruded and segmented in an anaerobic, sterilized glovebox containing N2 gas. After sampling, sediment samples were stored in sterile polyethylene bags and stored at 4 °C in the dark to reduce microbial activities before use. Further sediment-related experimental procedures were performed only under strict anoxic conditions. 2.2. Microcosm Incubations. A number of microcosm incubations under both aerobic and anaerobic conditions were conducted. About 1.0 g of sediments was mixed with 10 mL of sterilized artificial groundwater (AGW) that was based on the constituents of a groundwater sample near the study site except for omitting nitrates and phosphates (MgCl2, 1.968 mM; MgSO4·7H2O, 0.325 mM; NaHCO3, 0.51 mM; NaNO3, 0.009 mM; and CaCO3, 5.495 mM). Microcosms were incubated at room temperature (24−26 °C) in the dark under both aerobic and anaerobic conditions. The incubation conditions were (1) sterilized soil sample (abotic control); (2) unsterilized soil sample; (3) sterilized soil sample amended with 100 mg/L dipotassium phosphate (K2HPO4); (4) unsterilized soil sample amended with 100 mg/L dipotassium phosphate; (5) sterilized soil sample amended with 500 mg/L ammonium sulfate ((NH4)2SO4); and (6) unsterilized soil sample amended with 500 mg/L ammonium sulfate. The concentrations of inorganic nitrogenous and phosphorus fertilizers in the microcosms were selected according to the fertilizer requirement recommendations for rice from the Miaoli District Agricultural Research and Extension Station, Council of Agriculture, Executive Yuan, Taiwan. For anaerobic conditions, all of the experimental procedures and materials including medium and soil were placed in the anaerobic chamber with mix gas (N2, 80%; H2, 10%; and CO2, 10%) as inflow and catalyst to reach anaerobic condition for a few days before the beginning of anaerobic assays, which caused the elimination of O2 from materials and soil samples. In addition, the anaerobic indicator was used to monitor the anaerobic condition. Microcosms were sampled at the first and fourth months. Microcosms were subsampled carefully to prevent microbial contamination. The aqueous phase arsenic species were analyzed within 12 h. Each microcosm was prepared in triplicate for anaerobic incubation and in duplicate for aerobic incubation due to the limited amount of sediment samples. 2.3. Chemical Analysis. Prior to microcosm incubation, the levels of multiple chemical elements, including Fe, Mn, K, Na, Ca, As, Cd, Mg, Cr, Cu, Ni, Zn, P, and S, in the sediments were analyzed by X-ray fluorescence (XRF) (SPECTRO XEPOS, Kleve, Germany). For aqueous phase analyses, about 2 mL of sample was drawn from each microcosm at the fourth month in an anaerobic chamber. Samples were passed through a 0.45 μm filter for arsenic species analysis using
2. MATERIALS AND METHODS 2.1. Description of Sampling Sites. The sampling site ASYL718 (from 18 m in depth) is located in the southern Zhuoshui River alluvial fan of Taiwan (23°37′ N, 120°09′ E) (Figure 1), which is near the southwestern coast and in the BFD endemic area. The southern Zhuoshui River alluvial fan has an area approximately 24 km2 with 24 km from north to south and 1 km from east to west.32 A hydrogeological study indicated that the Zhuoshui River alluvial fan was formed in the late Quaternary and is partitioned into the proximal-, mid-, and distal-fan areas.33,34 High concentrations of arsenic were found in both aquifers and aquitards in the area, and arsenic in the groundwater originated from the aquitard of the marine sequence.34 Sampling site ASYC160 (from 160 m in depth) (23°19′ N, 120°13′ E) (Figure 1) is located along the southwestern coast of Taiwan, in the Chianan Plain. Detailed hydrogeology and lithology of the area have 2215
DOI: 10.1021/acs.jafc.6b00253 J. Agric. Food Chem. 2016, 64, 2214−2222
Article
Journal of Agricultural and Food Chemistry Table 1. Multiple Element Measurement for Sampling Site ASYL718a original sedimentb (mg/kg) Fe Mn K Na Ca As Cd Mg Cr Cu Ni Zn P S
28515 131 26140 − 13415 14 6 31350 27 8 23 58 − 391
artificial groundwater (AGW) (μg/L)
anaerobic sterilized sedimentc (μg/L)
anaerobic nonsterilized sedimentc (μg/L)
anaerobic sterilized sediment/Pd (μg/L)
anaerobic nonsterilized sediment/Pd (μg/L)
37 −f 445 16397 18237 − − 42196 9 38 7 −
33 143 80 922 11970 2e − 5240 − − − −
19073 595 3320 17778 105363 66e − 62172 − − − −
233 137 2553 1052 6795 17e − 5183 − − − −
24468 393 35308 19602 60603 110e − 59322 − − − −
a All data are means of three replicates of measurements. bSoil concentration analyzed by XRF. cAqueous concentration analyzed by ICP-AES after 4 months of incubation under anaerobic condition without 100 mg/L phosphorus fertilizer. dAqueous concentration analyzed by ICP-AES after 4 months of incubation under anaerobic condition with 100 mg/L phosphorus fertilizer. eAqueous concentration of total As analyzed by HPLC-ICPMS after 4 months of incubation under anaerobic condition. f−, below the detection limit of the instrument.
using ClustalX software.40 Distance analysis and phylogenetic tree construction were analyzed using the neighbor-joining (NJ) method (MEGA vers. 4).41 A total of 1000 bootstrap replications were calculated. Bootstrapping is a method to test the reliability of the multiple sequence alignment. The tree was unrooted. An unrooted tree gives information about the relationships between taxa without making assumptions about ancestry.
high-performance liquid chromatography/inductive coupled plasma/ mass spectrometry (HPLC-ICP-MS) (Agilent 1260 Infinity Quaternary LC System; Agilent 7700x, Santa Clara, CA, USA). A column containing a mixture of 2.0 mM PBS/0.2 mM EDTA (pH6.0) solution was used as the mobile phase. Parameters for ICP-MS included RF power, 1400 W; plasma gas, 15 L/min; auxiliary gas, 1.0 L/min; carrier gas, 1.1 L/min; and sampling depth, 7.5 mm. Condition settings for LC included mobile phase flow rate, 1.0 mL/min; column temperature, ambient; injection volume, 50 μL; run time, 10 min; and number of injections, 1. Typical detection limits for four species, As(III), As(V), MMAA (monomethylarsonic acid), and DMAA (dimethylarsinic acid), are around 0.1 ppb. However, the detection limits regarding actual samples may be somewhat higher due to the sample preparation steps and interferences from matrix. The filtered aqueous samples were also analyzed for multiple metal elements by inductively coupled plasma/atomic emission spectroscopy (ICP-AES) (SPECTRO CIROS 120). 2.4. 16S rRNA Analysis. Given the potential roles of bacteria attributed to arsenic mobilization, the microbial communities present in the microcosms after 4 months of anaerobic incubation from sampling site ASYL718 were examined, using the 16S rRNA clone libraries. In addition, bacterial communities in the original sediment (without 4 months of incubation) were also characterized. Gemonic DNA from sediments was isolated using the MO BIO PowerSoil DNA Isolation Kit (MO BIO Laboratories, Carlsbad, CA, USA) following the manufacturer’s instructions. Subsequently, 16S rRNA fragment (645 bp) was amplified by a polymerase chain reaction (PCR) from extracted DNA samples with the universal bacterial primers.38 The amplified PCR products were further purified using a BioMan kit (GeneTeks, Taipei, Taiwan). The purified PCR product was then cloned into the yT&A vector using a TA cloning kit and competent Escherichia coli cells (Yeastern Biotech, Taipei, Taiwan) following the manufacturer’s instructions. About 30 randomly selected recombinant clones from each cloned library were selected for DNA sequence analysis. The presence of chimera sequence was checked using the Pintail program.39 16S rRNA nucleotide sequences were analyzed using the BLAST program against the NCBI database. The 16S rRNA clones were then assigned to a genus, a family, or an order according to the identities of the BLAST results. 16S rRNA sequences identified in this study were deposited in the GenBank database with the accession numbers KF826198− KF826280. 2.5. Phylogenetic Analysis. 16S rRNA sequences of the recombinant clones and related reference sequences were analyzed
3. RESULTS AND DISCUSSION 3.1. Sediment Description. The sediments collected at 18 m depth from ASYL718 (Figure 1) were dark gray fine sandy loam. The sediments contained 14 mg/kg As, high levels of Fe (28515 mg/kg), K (26140 mg/kg), Ca (13415 mg/kg), Mg (31350 mg/kg), and lower levels of Mn (131 mg/kg) and S (391 mg/kg) (Table 1). For sampling site ASYC160 (Figure 1), the sediments collected at 160 m depth were brown and silty clay containing As (21 mg/kg), Fe (39555 mg/kg), K (27865 mg/kg), Ca (12670 mg/kg), Mg (7900 mg/kg), Mn (457 mg/kg), and S (3997 mg/kg). 3.2. Effects of Inorganic Nitrogenous and Phosphorus Fertilizers on Arsenic Mobilization under Aerobic Incubation. In aerobic conditions, all of the microcosm incubations from both sampling sites ASYL718 and ASYC160 showed negligible monomethylarsonic acid (MMA) and dimethylarsinic acid (DMA) release from arsenic-containing sediments (data not shown). In addition, aqueous As(III) was insignificant (Figure 2). For sites ASYL718, an addition of 100 mg/L dipotassium phosphate (inorganic phosphorus fertilizers) enhanced As(V) release from both sterilized and nonsterilized sediments with mean values of 13 and 20 μg/L As(V), respectively, for 1 month of aerobic incubation and 81.1 and 136 μg/L As(V), respectively, for 4 months of aerobic incubation (Figure 2A). In contrast, the addition of 500 mg/L ammonium sulfate (inorganic nitrogenous fertilizers) had no effect on As(V) release in either sterilized or nonsterilized sediments after 1 and 4 months of incubation (Figure 2A). For site ASYC160, a small amount of As(V) (about 6 μg/L) was released from both the original sterilized and nonsterilized sediments after 1 month of aerobic incubation, whereas As(V) 2216
DOI: 10.1021/acs.jafc.6b00253 J. Agric. Food Chem. 2016, 64, 2214−2222
Article
Journal of Agricultural and Food Chemistry
In contrast, for site ASYC160, the effect of phosphate on As(V) mobilization was not as significant as that of site ASYL718 and the presence of microorganisms did not play a significant role in mediating As(V) mobilization (Figure 2B). The discrepancy of As(V) mobilization between the two sampling sites might be due to the geochemical properties and bacteria present in the sediments. 3.3. Effects of Inorganic Phosphorus Fertilizers on Arsenic Mobilization under Anaerobic Incubation. In anaerobic conditions, all of the microcosm incubations from both sampling sites ASYL718 and ASYC160 did not exhibit arsenic mobilization, including As(V), MMA, and DMA (data not shown). We therefore presented only As(III) concentrations in Figures 3 and 4.
Figure 2. Effects of phosphorus and nitrogenous fertilizers on arsenic concentrations under aerobic incubation: (A) ASYL718 and (B) ASYC160 sediments amended with 100 mg/L phosphorus fertilizers and 500 mg/L nitrogenous fertilizers after 1 and 4 months of aerobic incubation. Data are the mean of two replicate experiments.
became insignificant after 4 months of aerobic incubation (Figure 2B). An addition of 100 mg/L dipotassium phosphate increased As(V) release from both sterilized and nonsterilized sediments with a mean value of 26 μg/L As(V) for 1 month of aerobic incubation and 12 and 2 μg/L As(V), respectively, for 4 months of aerobic incubation (Figure 2B). The addition of 500 mg/L ammonium sulfate (inorganic nitrogenous fertilizers) only slightly increased As(V) release (about 5−7 μg/L) in both sterilized or nonsterilized sediments after 1 month of incubation, whereas As(V) became insignificant after 4 months of aerobic incubation (Figure 2B). A number of studies have indicated that because it has a chemical characterization similar to As(V), phosphate has a high potential to desorb As(V) from soil through ion exchange reactions, doing so by competing with As(V) for adsorption sites and dissolving arsenic-rich metal oxides present in soils.42−45 In addition, it has been shown that the bacterium GFAJ-1 can use As(V) instead of phosphate as an energy source, which has profound geochemical significance.46 Recently, Basturea et al.47 reported that both E. coli and GFAJ-1 could use arsenic instead of phosphorus for their growth because As(V) induces massive ribosome breakdown, which provides a source of phosphate. Therefore, Figure 2A showed that in aerobic conditions the addition of phosphate increases As(V) mobilization and that the presence of microorganisms even further enhances As(V) release. This might be due to the competition of phosphate with As(V) for adsorption sites, thereby dissolving arsenic-rich metal oxides present in sediment along with the presence of bacteria in sediment.
Figure 3. Effects of phosphorus fertilizers on arsenic concentrations under anaerobic incubation: (A) ASYL718 and (B) ASYC160 sediments amended with 100 mg/L phosphorus fertilizer after 1 and 4 months of anaerobic incubation. The error bar represents the mean ± standard error of mean (SEM) of three replicate experiments.
For site ASYL718, similar to the previous finding in the BFD region,27 the anaerobic nonsterile microcosm released significant amounts of As(III) (260.7 ± 22.5 and 66.4 ± 4.6 μg/L for 1 and 4 months, respectively) (Figure 3A), indicating there was a microbially mediated As(III) release process in the arseniccontaminated sediments. This finding similar to a previous report27 was not surprising because both sampling sites are located in the BFD endemic area and have similar soil depths and textures (dark gray sandy loam). For site ASYL718, the sterilized microcosm with 100 mg/L phosphate released As(III) about 10 and 15 μg/L after 1 and 4 month, respectively, incubations (Figure 3A). This suggests that As(III) might be desorbed from soil surface sites by phosphate. Under anaerobic incubation, the nonsterilized 2217
DOI: 10.1021/acs.jafc.6b00253 J. Agric. Food Chem. 2016, 64, 2214−2222
Article
Journal of Agricultural and Food Chemistry
about 7 and 13 μg/L after 1 and 4 months of incubations, respectively) (Figure 3B). Although the amount of As(III) release was not as much as that of sampling site ASYL718, the sedimentary As(III) release enhanced by microorganisms and phosphate in the sediment was evident. Taken together, both microorganisms and phosphate play roles in the aqueous As(III) release in arsenic-contaminated sediments under anaerobic condition. 3.4. Effects of Inorganic Nitrogenous Fertilizers on Arsenic Mobilization under Anaerobic Incubation. Unlike the phosphorus fertilizers, the addition of nitrogenous fertilizers with 500 mg/L ammonium sulfate under anaerobic conditions did not show a significant effect on the increase of arsenic solubility in either sterilized or nonsterilized sediments (Figure 4). This suggests that nitrogenous fertilizers do not have a significant role in the release of arsenic into groundwater. It is interesting that one study reported that the application of nitrogen fertilizers increased the release of arsenic from sediment to groundwater in Bangladesh,51 which differs from our present finding. It is likely that there are different microorganisms present in the sediments in different regions (Bangladesh vs BFD region), resulting in different arsenic release mechanisms. 3.5. Effects of Phosphorus Fertilizers on Metal Element Release in Groundwater. The association of high levels of arsenic and multiple metal elements in groundwater has been described, although the relationship remains undefined.29−31 At the end of 4 months of anaerobic incubation, the aqueous phases of the microcosms from sampling site ASYL718 were analyzed for multiple metal elements by ICP-AES as the microcosm released a higher amount of As(III). In the absence of phosphorus fertilizers, by comparison with anaerobic sterilized sediments, arsenic release co-occurred with high levels of increase in Fe, K, Na, Ca, and Mg and a moderate increase of Mn in anaerobic unsterilized sediments (Table 1). Moreover, the addition of phosphorus fertilizers further enhanced the levels of As, Fe, K, and Na, whereas Mn and Ca showed the opposite trend in anaerobic unsterilized sediments (Table 1). Therefore, in anaerobic unsterilized sediments in the absence of phosphate, the bacterial arsenic mobilization might occur simultaneously with the dissolution of Fe, K, Na, Ca, and Mn, suggesting that the major bacterial mechanism might be via the dissimilatory reduction of As(V) and Fe(III) and possibly with minor bacterial Mn(IV) reduction. Furthermore, in the presence of 100 mg/L phosphorus fertilizers, further elevated arsenic level coincided with elevated levels of Fe, K, and Na, suggesting that a bacterial dissimilatory reduction of As(V) and Fe(III) was the major factor. This finding is in agreement with a previous report for the BFD endemic area27 and studies in other regions, such as West Bengal52,53 and Cambodia.54 A study reported that arsenic in groundwater of Chalkidiki, northern Greece, is highly correlated with K, Na, Mn, and Fe.55 It has been reported that microorganisms were involved in multiple metal element release, including the reduction of iron and arsenic by respiration under anaerobic incubations in arsenic-contaminated sediments.56 In the present study, high levels of As, Fe, K, Na, and Ca and moderate Mn were released from unsterilized sediments under anaerobic condition. This suggests that there is a common geogenic origin of these metal elements and conditions that microbially enhance their mobility.
Figure 4. Effects of nitrogenous fertilizers on arsenic concentrations under anaerobic incubation: (A) ASYL718 and (B) ASYC160 sediments amended with 500 mg/L nitrogenous fertilizer after 1 and 4 months of anaerobic incubation. The error bar represents the mean ± SEM of three replicate experiments.
microcosm with 100 mg/L phosphate released a more significant amount of As(III) than that of sterilized microcosm additional released As(III) about 60 and 45 μg/L after 1 and 4 month, respectively, incubations) (Figure 3A). This suggests that besides desorption, the sedimentary As(III) release was also enhanced by microorganisms in the microcosm with 100 mg/L phosphate addition under anaerobic conditions. It has been reported that the addition of phosphorus fertilizers (NaH2PO4·2H2O) results in higher soluble total As with As(III) as the main arsenic species under anaerobic conditions in a West Bengal (India) soil,48 which was in agreement with the present study, yet the interaction of fertilizers, microorganisms, and arsenic mobilization in aquifer remains uncharacterized. It is noted that the aqueous As(III) concentrations at 4 months of incubation was significantly decreased by comparison with that of 1 month (Figure 3A). This might be due to the coprecipitation of arsenic by the formation of insoluble, hydrated iron oxides, which then became immobilized.48−50 For site ASYC160, a lesser extent of As(III) release occurred in nonsterilized microcosm with 9.3 ± 5.2 and 5.7 ± 3.5 μg/L As(III) after 1 and 4 months of anaerobic incubation, respectively (Figure 3B). In sampling site ASYC160, As(III) displacement from soil surface sites (desorption) by PO43− was not significant. Under anaerobic incubation, the sterilized microcosm with 100 mg/L phosphate only further released As(III)
