Water Air Soil Pollut (2015) 226:167 DOI 10.1007/s11270-015-2408-4

Arsenic Removal and Transformation by Pseudomonas sp. Strain GE-1-Induced Ferrihydrite: Co-precipitation Versus Adsorption Wei Xiu & Huaming Guo & Qiong Liu & Zeyun Liu & Yan’e Zou & Baogang Zhang

Received: 2 December 2014 / Accepted: 29 March 2015 # Springer International Publishing Switzerland 2015

Abstract Hundreds of millions of people are at risk from drinking arsenic (As)-contaminated groundwater in the world, making As removal from aquatic systems of utmost importance. However, characteristics of As removal by bacteria-induced ferrihydrite and coupled with redox processes are still not clear. Two-line ferrihydrite was formed in the presence of aerobic Fe(II)oxidizing bacterium, Pseudomonas sp. strain GE-1. Arsenic co-precipitation with and adsorption onto ferrihydrite induced by Pseudomonas sp. strain GE-1 and redox processes of As were investigated. Results demonstrated that co-precipitation performed better in As(V) removal than As(III) removal, while adsorption showed higher capacity for As(III) removal. X-ray absorption near-edge spectroscopy (XANES) indicated that As(III) oxidation occurred in solid phases during co-precipitation and adsorption. Detection of As species in solution showed that As(V) was reduced to As(III) during co-precipitation, although no As(V) reduction occurred during adsorption. Arsenic immobilization by Electronic supplementary material The online version of this article (doi:10.1007/s11270-015-2408-4) contains supplementary material, which is available to authorized users. W. Xiu : H. Guo State Key Laboratory of Biogeology and Environmental Geology, China University of Geosciences, Beijing 100083, People’s Republic of China W. Xiu : H. Guo (*) : Q. Liu : Z. Liu : Y. Zou : B. Zhang School of Water Resources and Environment, China University of Geosciences (Beijing), Beijing 100083, People’s Republic of China e-mail: [email protected]

Pseudomonas sp. strain GE-1-induced ferrihydrite in the presence of the strains may be applied as an alternative remediation strategy. Keywords Arsenic removal . As(III) oxidation . Fe(II)-oxidizing bacteria . Ferrihydrite

1 Introduction High arsenic (As) groundwater, causing significant health impacts on mankind, has been regarded as a serious environmental issue worldwide (Karim 2000; Smedley and Kinniburgh 2002). Arsenic concentrations of drinking groundwater above the World Health Organization (WHO) maximum contaminant level have universally been found generally as a result of natural processes, including geothermal sources, weathering of As-bearing minerals, and release of As adsorbed on mineral surfaces due to microbial activity (Harvey et al. 2002; Islam et al. 2004; Morin and Calas 2006; Bhattacharya et al. 2007; Alam et al. 2014). Inorganic forms of As in aqueous environment (groundwater and surface water) are the species arsenate (As(V)) and arsenite (As(III)). The dominant As(III) species is arsenious acid (H3AsO30) at pH <9, while As(V) mainly exists as oxyanions H2AsO4− and HAsO42− at pH 2–11 (Bissen and Frimmel 2003). Arsenic removal from drinking water can be achieved by precipitation-coagulation, membrane separation, ion-exchange, co-precipitation, and adsorption (Thirunavukkarasu et al. 2003; Berg et al. 2006; Guo

167

Page 2 of 14

et al. 2007a; Han et al. 2011; Lin et al. 2012; Pal et al. 2014). Among these available technologies, coprecipitation of As with Fe oxides and adsorption on Fe-containing substances at neutral pH have been of increasing interest due to their high specific surface areas, positive surface charges at neutral pH, and sufficient adsorption sites (Banerjee et al. 2008; Jessen et al. 2005). Ferrihydrite, a poorly ordered hydrous Fe oxide with a high specific surface area (241–272 m2·g−1) and commonly present in low-temperature geochemical environments, is one of the most important minerals in the sequestration of As (Jang et al. 2006; Michel et al. 2007). Previous studies showed that abiotically synthesized ferrihydrite has a high capacity for As removal via both co-precipitation and adsorption (Bhandari et al. 2011; Jia and Demopoulos 2005; Raven et al. 1998; Tokoro et al. 2010; Zhao et al. 2011). Jia and Demopoulos (2005) suggested that almost all As (300 mg/L) was removed by either adsorption or coprecipitation at molar ratio of Fe/As=8 and pH=8. However, these studies are less relevant to As removal by bacteria-induced ferrihydrite in the presence of Asresistant bacteria. It has been well recognized that As speciation in natural environment strongly affects its mobility and bioavailability at water-mineral interface. Arsenic(III) is more soluble and mobile than As(V) due to its weaker affinity to minerals (Pedersen et al. 2006). Accordingly, factors influencing the transformation between As(III) and As(V) inevitably affect the fate and the mobilization of As in the natural environment. Although Fe and Mn minerals provide important abiotic pathways for As transformation from As(III) to As(V) (Amirbahman et al. 2006), microbes can lead to redox transformation of As via biological metabolism or enzyme catalysis (Paez-Espino et al. 2009). Although As adsorption on Fe-(oxyhydr)oxides has been intensively studied under both oxic and anoxic conditions over a wide pH range (Dixit and Hering 2003; Hug and Leupin 2003; Jia and Demopoulos 2005), mechanisms by which As(III) is oxidized homogenously during its removal at neutral pH in aerated waters, and by heterogeneous oxidation of As(III) in presence of abiogenic/biogenic Fe-(oxyhydr)oxides, are controversial issues in the literatures. Greenleaf et al. (2003) have showed that nearly all As(III) in the aqueous phase is converted to As(V) in a ferrihydrite-loaded column, where ferrihydrite acts as the oxidant. However, Ona-Nguema et al. (2010) have recently suggested that As(III) is oxidized upon sorption

Water Air Soil Pollut (2015) 226:167

onto ferrihydrite only after addition of Fe(II), which confirms that Fe(II) is able to catalyze As(III) oxidation on the surface of ferrihydrite under oxic condition. This Fe(II,III) system, characterized by the occurrence of electron transfer, adsorption and/or co-precipitation during the interaction between Fe-(oxyhydr)oxides and As species, may play an important role in As retention and redox transformations at oxic–anoxic boundaries in natural systems (Root et al. 2007). Iron(II)-oxidizing bacteria induce Fe(II) oxidation that typically leads to the formation of amorphous or poorly crystalline Fe(III) phases and the efficient removal of As from the culture solution (Hohmann et al. 2010; Liu et al. 2013). However, it is not clear whether there is a difference in redox transformation of As between adsorption and coprecipitation by Fe-(oxyhydr)oxides in the presence of Fe(II)-oxidizing bacteria and its role in As removal. Based on this knowledge gap, the objectives of this study were to (i) determine characteristics of As removal from aqueous solution by bacteria-induced ferrihydrite, through co-precipitation and adsorption, (ii) investigate effects of As species on As removal by bacteria-induced ferrihydrite and (iii) evaluate As redox processes of As immobilization by adsorption and co-precipitation.

2 Materials and Methods 2.1 Ferrihydrite Induced by Pseudomonas sp. Strain GE-1 All chemicals used in this study were of analytical reagent grade. All volumetric flasks and vessels were cleaned by soaking in 10 % HNO3 for at least 24 h and rinsed several times with deionized water. Biogenic ferrihydrite was formed in the presence of an aerobic Fe(II)oxidizing bacterium, Pseudomonas sp. strain GE-1, which was previously isolated from a rusty iron wirecontaining tap water and routinely cultured in the modified Winogradsky’s medium (Liu et al. 2013). The medium was composed of the following (per liter): 0.5 g KH2PO4·3H2O, 0.5 g NaNO3, 0.2 g CaCl2·2H2O, 0.5 g MgSO4·7H2O, 0.5 g (NH4)2SO4, and 10 g ammonium ferric citrate (FeC6H5O7·NH4OH). The medium had Fe(II) concentration of around 17.0 mg/L due to the partial photochemical reduction of ammonium ferric citrate (O’Neil et al. 2001), and the strong citrate complexation of Fe(II) which keeps it in the reduced state even under aerobic conditions. Culture pH was adjusted to 7.0

Water Air Soil Pollut (2015) 226:167

with 1.25 mol/L NaOH solution. The growth media had total dissolved Fe dosage of 37.5 mM (high Fe dosage series). All sterilization was conducted in an autoclave at 121 °C for 20 min. For all experiments, the volume percent of inoculum was always 5 %. All batches were equilibrated at 25 °C for 168 h in a shaking water bath (150 rpm). Arsenic-free control batches with the culture medium and the GE-1 strain allowed to identify biogenic Fe(III) precipitation.

2.2 Experimental Setup Stock As(V) and As(III) solutions (5000 mg/L) were prepared from sodium hydrogen arsenate (Na2HAsO4·7H2O, Fluka Chemical) and sodium arsenite (NaAsO2, Fluka Chemical), respectively, using deionized acidified water. Stock solutions were filtered with 0.22 μm filter membrane to get rid of any microbes prior to experiments, and added into the culture media to achieve specific As(III) and As(V) concentrations (either13.3 μM or 1.33 mM) in all experiments. For Pseudomonas sp. strain GE-1-mediated coprecipitation experiments, As(III) (or As(V)) was added into sterilized media prior to the introduction of strain GE-1 and the formation of bacteria-induced ferrihydrite. To accomplish this, 5 % (vol.%) of fresh GE-1 culture adapted to concentrations of 13.3 μM and 1.33 mM As(III) (or As (V)) were incubated in 100 mL of the culture media containing initial As(III) (or As(V)) concentrations of 13.3 μM and 1.33 mM, respectively. For Pseudomonas sp. strain GE-1-mediated adsorption experiments, As(III) (or As(V)) was introduced into the culture after biogenic Fe(III) precipitation. To accomplish this, 5 % (vol.%) of fresh GE-1 culture adapted to concentrations of 13.3 μM and 1.33 mM As(III) (or As (V)) were incubated in 100 mL of fresh culture media free of As for low and high As adsorption batches, respectively. After precipitation of biogenic ferrihydrite, As(III) (or As(V)) was added to the suspensions to achieve two different initial As concentrations (13.3 μM and 1.33 mM for low and high As concentrations, respectively). For low Fe dosage experiments, both Pseudomonas sp. strain GE-1-mediated co-precipitation and adsorption experiments were performed in the media with total dissolved Fe dosage of 8.0 mM by decreasing ammonium ferric citrate (FeC6H5O7·NH4OH) from 10 to 2 g/L without changing concentrations of other ingredients.

Page 3 of 14 167

Abiotic control experiments were carried out with the As-containing culture in the absence of strain GE-1 for both co-precipitation experiments and adsorption experiments. Biotic controls were carried out with 1.33 mM As(III)-containing culture media free of NO3− in the presence of strain GE-1, which were set to evaluate the role of NO3− in Fe(II) oxidation. The suspensions were taken at different time intervals and divided into two aliquots. One was used to detect concentration of adenosine triphosphate (ATP). As the Bmolecular unit of currency^ of intracellular energy transfer, concentration of ATP shows a clear correlation with counts of bacteria cells, and therefore is used to indirectly reflect the number of bacterial cells (Okibe and Johnson 2011). The other was centrifuged at 5000 rpm for 10 min, filtered with 0.22 μm filter membrane, and finally analyzed for total dissolved As, As species, Fe species, NO3−, and NO2−. The Fe precipitates were sampled according to the following steps: centrifuging at 1000 rpm for 10 min and rinsing by deionized water, repeating these two steps three times, and then centrifuging at 3000 rpm for 20 min. After that, it was dried in anaerobic glove box (Coy Lab, USA) and preserved in anaerobic amber glass bottles with a headspace of N2/H2 (92.5/7.5). All experiments were conducted in duplicate and reported as a mean value. 2.3 Sample Analysis Total dissolved As and Fe concentrations were determined by inductively coupled plasma-optical emission spectrometry (ICP-OES; iCAP 6300, Thermo). Immediately after sampling, concentrations of As(III) and As(V) were measured using high-performance liquid chromatography-hydride generation-atomic fluorescence spectrophotometer (HPLC-HG-AFS; Jitian Corp., Beijing) with a detection limit of 1.0 μg/L. Concentration of dissolved Fe(II) was analyzed using a spectrophotometer (DR 2800, HACH) with 1,10phenanthroline method (Tamura et al. 1974). Samples were analyzed for NO3− and NO2− by ion chromatography (ICS-1000, Dionex). Concentrations of ATP were determined by ATP fluorescence detector (AF-100, DKK-TOA). X-ray absorption near-edge spectroscopy (XANES) was used to detect the As species in solid phase. The Xray absorption data were recorded at room temperature at beamline BL15U1 of the Shanghai Synchrotron Radiation Facility (SSRF), China. Prior to XAS

167

Page 4 of 14

analysis, dried samples were ground and enclosed by Kapton tape in an anaerobic chamber with O2 levels <1.0 ppm. In order to avoid oxygen from contacting the samples and restrain the beam-induced oxidation during measurement, samples were tightly sealed by the Kapton tape. Repeated measurement of samples showed that As(III) oxidation by the X-ray beam was limited in this case. Beamline was equipped with a double-crystal Si(111) monochromator. Spectra were collected in both transmission mode using ion chambers and fluorescent mode with silicon drift fluorescence detector, from −50 to 150 eV relative to the As K-edge of 11,867 eV. During the measurement, the synchrotron was operated at energy of 3.5 GeV and a current between 150 and 210 mA (Guo et al. 2013a). In addition to sodium arsenite and sodium arsenate, arsenate-adsorbed ferrihydrite (HFO-As(V)) and arsenite-adsorbed ferrihydrite (HFO-As(III)) were used as model compounds. The last two compounds were synthesized in batches with ferrihydrite as adsorbent, a dosage of 50 g/L, and initial concentration of 1.33 mM As(V) and As(III), respectively, under anaerobic conditions. Data processing was performed using the program ATHENA (Ravel and Newville 2005). Fourier transform infrared spectroscopy (FTIR) spectra of the samples were obtained on a Bruker TENSOR 27 FTIR. The KBr/sample discs were prepared by mixing 0.5 % of finely ground dry samples in KBr. The measurement resolution was set at 4 cm−1, and the spectra were collected in the range of 400–4000 cm−1 with 200 scans.

3 Results 3.1 Arsenic Removal by Biogenic Ferrihydrite Variations in As concentration in aqueous phase as a function of incubation time are shown in Fig. 1a, b for high total dissolved Fe dosage series. Initial As concentrations were around 13.3 μM and 1.33 mM in culture media for low and high As(III) (or As (V)) concentrations series with Fe/As molar ratio of 2800 and 28, respectively. 3.1.1 Arsenic Removal by Co-precipitation Results of co-precipitation experiments showed that total As in experimental batches slightly decreased

Water Air Soil Pollut (2015) 226:167

between 0 and 36 h, and sharply decreased between 36 and 120 h, and then achieved equilibrium after 120 h. Finally, 99±0.5 % of total As was removed for both low and high As concentrations series (Fig. 1a, b). In controls, total As generally kept constant during experiments. The removal efficiency of As(V) was higher than that of As(III) in both low and high As concentration series within 84 h, although the final removal was not significantly different after equilibrium (Fig. 1a, b). Longer equilibrium time (120 h) was observed for high As(III) concentration series. It indicated that higher As concentration may lead to longer equilibrium time for As removal (Fig. 1a, b). 3.1.2 Arsenic Removal by Adsorption In the case of adsorption experiments, total As drastically decreased between 72 and 84 h, and then achieved equilibrium at 120 h after incubation or between 0 and 12 h and at 48 h after introduction of As(III) or As(V) (Fig. 1c, d). Finally, higher removal efficiency was achieved (99±0.3 %) in low As concentration series than that in high As concentration series (97±0.5 %). Total As concentrations always kept stable at around 13.3 μM and 1.33 mM in controls for low and high As concentrations series, respectively (Fig. 1c, d). It was suggested that higher As concentration would result in lower removal efficiency. Although initial As concentration did not significantly affect equilibrium time for As removal, As(III) was removed more rapidly than As(V). In low As concentration series, 99±0.5 % of As(III) was scavenged within 1 h after introduced at 72 h, while 96±0.5 % of As(V) within 24 h (Fig. 1c). Besides, more As was adsorbed from As(III) batches (99±0.5 % with residual As concentration of 10 μg/L) than As(V) batches (97±0.5 % with residual As concentration of 25 μg/L) after equilibrium. In high As concentration series, As concentration decreased from 101 to 3.8 mg/L in As(V) batches, and from 100 to 2.5 mg/L in As(III) batches (Fig. 1d). The results suggested that both adsorption rate and efficiency for As(III) were slightly higher than those for As(V). 3.1.3 Effect of Fe Dosage on As Removal Precipitation of total dissolved iron was accomplished within 60 h in batches with Fe dosage of 8.0 mM (low Fe dosage series, Fig. S1a in the Supplementary Material), while within 72 h in batches with Fe dosage

Water Air Soil Pollut (2015) 226:167

Page 5 of 14 167

Fig. 1 Variations in total As concentration in co-precipitation experiments (a 13.3 μM As(V) or As(III) batches with or without Fe(II)-oxidizing bacteria; b 1.33 mM As(V) or As(III) batches with or without Fe(II)-oxidizing bacteria) and adsorption experiments

(c 13.3 μM As(V) or As(III) batches with or without Fe(II)oxidizing bacteria; d 1.33 mM As(V) or As(III) batches with or without Fe(II)-oxidizing bacteria)

of 37.5 mM (high Fe dosage series, Fig. 1a). Lower Fe dosage achieved lower As removal efficiencies for both As(III) and As(V) in co-precipitation experiments, although kinetic rates for As removal for low Fe dosage were generally identical to those for high Fe dosage. Residual As concentrations in both As(V) and As(III) batches slightly decreased between 0 and 36 h, and sharply decreased between 36 and 60 h (Fig. S1a in the Supplementary Material). However, As removal was in equilibrium after 72 h in As(V) batches, being earlier than 96 h in As(III) batches, with removal efficiencies of 97 ± 0.5 and 95 ± 0.5 %, respectively (Fig. S1a in the Supplementary Material). Relatively higher removal efficiency was observed for As(V) batches, in comparison with that for As(III) batches,

which indicates that co-precipitation removed As(V) more efficiently than As(III). Similar to high Fe dosage series, both As(V) and As(III) were rapidly removed after As introduction in adsorption experiments with low Fe dosage series (Fig. S1b in the Supplementary Material). However, the overall removal efficiency (95±0.4 and 92±0.5 % for As(III) and As(V) batches, respectively) in low Fe dosage series was lower than those in high Fe dosage series (99±0.5 and 97±0.5 % for As(III) batches and As(V) batches, respectively). Although the equilibrium was observed at about 84 h for low Fe dosage series, which was slightly earlier than 96 h for high Fe dosage series, less As (50~60 %) was removed within 1 h after As introduction. It was observed that more As(III) was

167

Page 6 of 14

removed by adsorption than As(V). Furthermore, the difference in As removal between As(III) and As(V) batches for low Fe dosage series was more apparent than that for high Fe dosage series. 3.2 Redox Transformation of As During Co-precipitation and Adsorption 3.2.1 Arsenic Species in Solid Phase The XANES spectra of solid samples exhibited a wellresolved edge structure with an absorption maximum at 11,875 and 11,871.3 eV, which correspond to As(V) and As(III) coordinated to oxygen, respectively (Fig. 2). Due to the high As content, solid samples from experiments with low Fe/As molar ratio (28) were chosen to detect the As redox state. XANES spectra analysis was performed using linear combination fitting. In both co-precipitation and adsorption, As(III) was gradually oxidized to As(V) in the biogenic ferrihydrite in As(III) batches. Extent of As(III) oxidation was higher in co-precipitation batches than adsorption ones. For 1.33 mM As(III)-treated series, 30 % of As(III) was oxidized to As(V) after 48 h and 80 % after 168 h in coprecipitation experiments (Fig. 2a; Table S1 in the Supplementary Material). However, in adsorption experiments, 62 % of As(III) was oxidized to As(V) at 168 h (Fig. 2c; Table S1 in the Supplementary Material). Arsenic(V) reduction was observed during co-precipitation, but not during adsorption. For 1.33 mM As(V)treated series, As(III) was detected in the As-containing ferrihydrite from co-precipitation experiments and its proportion generally kept stable (between 34 and 38 % of total As) with increasing reaction time, except that the proportion of As(III) decreased to 29 % in batches with incubation time of 168 h. It generally indicated that As(III) was oxidized to some extent with increasing reaction time (Fig. 2b; Table S1 in the Supplementary Material). However, As(III) was not observed in Asadsorbing ferrihydrite from adsorption experiments (Fig. 2d). 3.2.2 Arsenic Species in Solution Interestingly, As(III) with concentration of 0.1 mM was detected in solutions after 48 h incubation in coprecipitation experiments with initial As(V) of 1.33 mM (Fig. 3). After that, both As(III) and As(V) concentrations gradually decreased due to co-precipitation and/or

Water Air Soil Pollut (2015) 226:167

adsorption. It indicated that As(V) reduction occurred in the solutions of the co-precipitation system in the presence of Fe(II), Fe(III), and Pseudomonas sp. strain GE1. The presence of As(III) in both solutions and solids implied that As(V) reduction occurred in solutions (and/ or solids). However, no As(III) was detected in solutions from adsorption experiments with initial As(V) concentration of 1.33 mM (Table S1 in the Supplementary Material). The absence of As(III) in both solutions and solids indicated that As(V) was not reduced in the system of As-adsorbing ferrihydrite. In As(III) batches, although As(III) oxidation occurred in the solid phases, no As(V) was detectable in aqueous solutions during co-precipitation and adsorption (Table S1 in the Supplementary Material). It indicated that As(III) oxidation would occur in the biogenic ferrihydrite during co-precipitation and adsorption.

4 Discussion 4.1 Formation of Biogenic Ferrihydrite Iron(II) was oxidized and precipitated in the presence of Pseudomonas sp. strain GE-1 (Fig. 4). Iron(II) concentration decreased from 0.304 mM to <0.5 μM during cultivation. However, in controls, pH and concentrations of Fe(II) were stable throughout the experiments at around 7.0 and 0.304 mM, respectively. After oxidation, Fe(III) was precipitated during cultivation, with the formation of the red brown Fe(III) precipitation (Fig. S2). The pH of culture media (initial pH, 7.0) increased, and finally achieved equilibrium at 9.0 in 120 h incubation. In the presence of Pseudomonas sp. strain GE-1, the increase in solution pH due to bacterial respiration led to Fe(II) oxidation and Fe oxide precipitation. However, no any precipitation was observed in controls without Pseudomonas sp. strain GE-1 after increasing in solution pH from 7.0 to 9.0 by simply adding NaOH, possibly due to the absence of Fe(II) oxidation. Therefore, Pseudomonas sp. strain GE-1 was expected to induce Fe(II) oxidation and therefore Fe oxide precipitation. The red brown Fe(III) precipitation, produced in the presence of Pseudomonas sp. strain GE-1, was two line ferrihydrite (Liu et al. 2013). FTIR spectra further confirmed the observation of Liu et al. (2013). Biogenic Fe(III) precipitation showed the representative FTIR

Water Air Soil Pollut (2015) 226:167

Page 7 of 14 167

Fig. 2 Arsenic K-edge XANES spectra of As-co-precipitating ferrihydrite: a batches with 1.33 mM As(III) and Fe(II)-oxidizing bacteria; b batches with 1.33 mM As(V) and Fe(II)-oxidizing

bacteria. Arsenic K-edge XANES spectra of As-adsorbing ferrihydrite: c batches with 1.33 mM As(III) and Fe(II)-oxidizing bacteria; d batches with 1.33 mM As(V) and Fe(II)-oxidizing bacteria

bands at 1616, 1389, and 453 cm−1 in fingerprint district (Fig. 5a), which were identical to typical bands at 1616,

1384, and 453 cm−1 in characteristic bands of synthesized ferrihydrite (Jia et al. 2007; Katsoyiannis and

167

Page 8 of 14

Fig. 3 Changes in As species in solutions with cultivation time in co-precipitation experiment with 1.33 mM As(V) and Pseudomonas sp. strain GE-1

Fig. 4 Variations in residual Fe(II), concentration of ATP, NO2− and pH in 1.33 mM As(III)containing growth media with/without NO3−

Water Air Soil Pollut (2015) 226:167

Water Air Soil Pollut (2015) 226:167

Zouboulis 2004). Additionally, the bands at 1039 cm−1 might be due to the presence of sulfate in the medium (Myneni et al. 1998). Similarly, oxidation of dissolved Fe(II) by Rhodobacter ferrooxidans strain SW2 induced Fe oxide precipitation, which had a X-ray diffraction (XRD) pattern similar to the reference XRD spectrum of synthetic ferrihydrite (Kappler and Newman 2004). The microaerophilic bacteria Leptothrix ochracea and Gallionella ferruginea also induced ferrihydrite precipitation on its sheath and formed a stalk under certain conditions (Konhauser et al. 2011). The biological Fe(II) oxidation by Pseudomonas sp. strain GE-1 might be NO3− dependent or accelerated by NO 3 − reduction during microbial metabolism. Concentrations of NO 3 − decreased from 365 to 169 mg/L during the cultivation, which led to the presence of around 115 mg/L NO2− in growth media. However, NO3− concentration was kept relatively constant, and no NO2− was detectable in controls without the strains. Additionally, apparent oxidation of Fe(II) was about 12 h earlier in batches with growth media with 365 mg/L NO3− than the batches without NO3− (Fig. 4). It indicated that the presence of NO3−, which would be the electron acceptor, promoted Fe(II) oxidation. Similarly, Picardal et al. (2011) suggested that some species of the genus Dechlorospirillum, primarily known as perchlorate and nitrate reducers, would be members of the microbial communities involved in Fe redox cycling at the oxic–anoxic transition zones in freshwater sediments. Microbial metabolism would induce the abovementioned redox processes. Although the peak of ATP was observed at 36 h in batches with growth media, which is 12 h earlier than the batches without NO3−, the peak of ATP corresponded to the abrupt decrease in Fe(II) concentration and the sharp increase in NO2− concentration in both batches. At the low ATP concentrations, oxidation of Fe(II) was limited in the final stage of incubation (between 96 and 168 h) (Fig. 4). In comparison with batches free of As, the decreases in dissolved Fe and Fe(II) always lagged in As-treated batches, especially in As(III)-treated series (Fig. S2 in the Supplementary Material). It might be due to the toxicity of high As concentration on microbial metabolism. With biogenic ferrihydrite, over 95 % of As(V) and As(III) were removed from the media either through co-precipitation with and/or adsorption onto Fe(III) oxides (above stated). Therefore, both Fe(II) oxidization and Fe(III) precipitation led to As removal. Similar

Page 9 of 14 167

studies had shown that aerobic Fe(II)-oxidizing bacteria, e.g., Gallionella and Leptothrix species, and anaerobic nitrate-reducing Fe(II)-oxidizing bacteria, Acidovorax sp. strain BoFeN1, effectively immobilized As(III) from contaminated water through co-precipitation and/or sorption during microbial Fe(II) oxidation (Hohmann et al. 2010; Yang et al. 2014).

4.2 Difference in As Removal Between Adsorption and Co-precipitation It is generally noted that different removal processes including co-precipitation with or adsorption on Fe-(oxyhydr)oxides resulted in differences in removal efficiency, kinetics, and redox transformation of As species (Amirbahman et al. 2006; Greenleaf et al. 2003; Tufano and Fendorf 2008). The present study showed that co-precipitation had greater As removal efficiency than that of adsorption. In high As concentrations series (initial As/Fe molar ratio of 0.0357 based on total dissolved Fe), it was shown that molar ratios of Asremved/Fe during co-precipitation achieved 0.03511±0.0005 and 0.03526±0.0005 for As(III) and As(V), which were higher than 0.03481± 0.0005 and 0.03433±0.0003 for As(III) and As(V) during adsorption, respectively. Abiotically synthesized ferrihydrite showed approximately twice As(V) sorption density during co-precipitation (0.175 mol As(V)/mol of Fe(III)) as much as adsorption (0.090 mol As(V)/mol of Fe(III)) with initial As/Fe molar ratio of 0.20 after 24 h of reaction at pH 8 (Fuller et al. 1993). Additionally, Jia and Demopoulos (2005) observed that abiotically synthesized ferrihydrite had Asremved/Fe molar ratio of 0.244 by co-precipitation, being higher than adsorption (0.221) with the initial As/Fe of 0.25 and pH 8 (Jia and Demopoulos 2005). The discrepancy between abiotically synthesized and Pseudomonas sp. strain GE1-induced ferrihydrite in molar ratios of Asremved/Fe was probably due to the initial molar ratio of As/Fe. In this study, the initial molar ratio of 0.0357 is much lower than those applied in Fuller et al. (1993) and Jia and Demopoulos (2005). The other cause for the low sorption density attributed to the combined effect of coexisting anions (e.g., phosphate) and various microbial metabolites (e.g., proteins, carbohydrates, or even fragments of cells), which would compete for the coordination active sites, especially in the case of As(V) (Guo et al. 2007b; Kleinert et al. 2011).

167

Page 10 of 14

Both biogenic and chemical synthesized ferrihydrite showed that co-precipitation had greater As removal efficiency than adsorption. During co-precipitation, biogenic ferrihydrite was gradually formed in the presence of adsorbates, which affected the rates of crystal growth and led to large specific surface area, therefore having a high As binding ability to their surface (Hohmann et al. 2010). Another possible reason was that co-precipitation exposed the maximum of the coordination active sites and provided enough active sites for As(V) removal. Jia and Demopoulos (2005) suggested that the higher As(V) removal efficiency by co-precipitation than adsorption was probably due to the maximization of the coordination sites during the neutralization of the acidic As(V)–Fe(III) solution, where there was a better chance of contacting between As(V) and Fe(III). Last but not least, possible reason was that redox transrmation of As would also to some extent contribute to the higher As(V) removal efficiency by co-precipitation. For example, in co-precipitation series, As(V) reduction into As(III) was coupled with Fe(II) oxidation and subsequently increased in Fe(III) oxide precipitation, which may play an important role in its higher removal efficiency and kinetics rate for As(V) removal during coprecipitation. It indicated that experimental conditions had more significant effect on As(V) removal than As(III) removal. In the low Fe dosage, As(III) removal efficiency was around 95±0.5 %, which is lower than As(V) removal (around 97±0.5 %) during co-precipitation. However, that higher As(III) removal (95±0.4 %) was observed than As(V) (92±0.5 %) in adsorption experiments. It showed that adsorption was more efficient for As(III) removal than As(V) removal. Chemical synthesized ferrihydrite also exhibited a relatively greater adsorption for As(III) at higher pH values, whereas As(V) was more strongly adsorbed at lower pH values (Dixit and Hering 2003; Raven et al. 1998). Raven et al. (1998) suggested that adsorption of As(III) species, neutral H3AsO30 (pKa) in the pH range of 4.6–9.2, would be less strongly influenced by anion repulsion forces than adsorption of As(V) species on ferrihydrite surface. In this study with solution pH between 7.0 and 9.0, As(III) dominantly existed as neutral H3AsO30. Therefore, there was less effect of anion repulsion forces and negatively charged ferrihydrite surface on neutral H3AsO30 adsorption than negatively charged As(V) species adsorption. Therefore, similar to chemical synthesized ferrihydrite, biogenic ferrihydrite would adsorb As(III) more

Water Air Soil Pollut (2015) 226:167 Fig. 5 FTIR spectra of Pseudomonas sp. strain GE-1-induced b ferrihydrite (a), As-co-precipitating solids (b), and As-adsorbing solids (c) at different contact time in experimental batches with the presence of Fe-oxidizing bacteria and high Fe dosage

efficiently than As(V) in the presence of Pseudomonas sp. strain GE-1. 4.3 Redox Processes During As Removal As mentioned above, As(III) oxidation appeared in the solid during co-precipitation and adsorption. It was confirmed by both EXANES and FTIR results. The bands of 806 and 852 cm−1 were observed in the solids from both co-precipitation and adsorption series (Fig. 5b, c). The infrared band at 852 cm−1 was associated with arsenate adsorbed on amorphous Fe oxides (Myneni et al. 1998). The band was more apparent in As coprecipitation solids than As adsorption solids. Therefore, this provided another evidence that solution As(III) was firstly co-precipitated with and/or adsorbed onto biogenic ferrihydrite, and then converted to As(V). It was noted that catalytic oxidation of As(III) by ferrihydrite involved an initial step of As(III) adsorption on ferrihydrite, followed by an As(III)/As(V) conversion step (Zhao et al. 2011). No As related bands were clearly found in low As series (data not shown), which might result from the low As contents in biogenic ferrihydrite and/or the concealing effects of strong bands of biogenic ferrihydrite between 400 and 800 cm−1. Infrared spectra of As(III), which usually had the bands around 783– 794 cm−1 (Voegelin and Hug 2003), were not apparently observed in this study (Fig. 5). One reason might be due to the difficulty in detecting the presence of adsorbed As(III) on the surface of Fe oxides (Goldberg and Johnston 2001). Additionally, organic groups in the biogenic ferrihydrite, having a strong band at 450– 780 cm−1, concealed the spectra of As(III). Similarly, Goldberg and Johnston (2001) also found that observation of the bands in the region of 600–630 cm−1 was precluded by Fe oxides with a very intense IR band in the 600 cm−1 region. Another reason might be the low contents of As(III) in the biogenic ferrihydrite due to the oxidation of As(III) to As(V). Although As(III) oxidation was observed on Ascontaining ferrihydrite regardless of co-precipitation and adsorption, no As(V) was detected in aqueous solutions during co-precipitation and adsorption in As(III) batches (Table S1 in the Supplementary Material). It

Water Air Soil Pollut (2015) 226:167

Page 11 of 14 167

167

Page 12 of 14

indicated that As(III) oxidation would only take place in the solid, instead of the solution, during co-precipitation and adsorption. Similarly, As(III) being adsorbed on Fe oxides, including amorphous Fe oxides (ferrihydrite) and goethite, was oxidized to As(V), although different roles were played in oxidation process (Hug and Leupin 2003; Jang and Dempsey 2008; Zhao et al. 2011). For ferrihydrite, it was noted that catalyst or oxidant generally played a role in the oxidation of adsorbed As(III). Greenleaf et al. (2003) found that nearly all As(III) was converted to As(V) in a ferrihydrite column experiment, and ferrihydrite was thought as the oxidant in this reaction. Co-adsorption of As(III) and As(V) and oxidation of As(III) on ferrihydrite were observed, with ferrihydrite as the catalyst and dissolved O2 as the electron accepter (Jang and Dempsey 2008). In the present study, the presence of dissolved Fe(II) during formation of biogenic ferrihydrite indicated that the biogenic ferrihydrite was an Fe(II)-Fe(III) coexistence system. In this Fe(II)–Fe(III) system, intermediates (possibly an Fe(IV) species) or strongly oxidizing radical species, produced during the oxidation of Fe(II) by O2, were believed to be the major contributors for As(III) oxidation under oxic conditions (Hug and Leupin 2003; Ona-Nguema et al. 2010). It indicated that adsorbed As(III) was oxidized in the biogenic ferrihydrite as the catalyst. Reduction of As(V) would occur during coprecipitation in the presence of the strains, while no As(V) was reduced during adsorption experiments. Microbial activity, including specific gene expression or various microbial metabolites during the incubation of Pseudomonas sp. strain GE-1, likely played a role in As(V) reduction (Silver and Phung 2005) during coprecipitation. When As(III) oxidation in solids and As(V) reduction in solutions were taken together, the net As redox process appeared to be As(V) reduction in As(V) coprecipitation series. In 1.33 mM As(V) co-precipitation series, As(V) reduction occurred after 36 h, which was later than NO3− reduction (starting from 24 h). It indicated that NO3− was preferentially used as the electron acceptor, in comparison with As(V) in the presence of Pseudomonas sp. strain GE-1. This is consistent with the normal sequence of the redox processes (Guo et al. 2013a, b). Under this condition, the amount of electrons accepted from reduction of NO3− and As(V) was 2.8 and 1.1 mmol/L, respectively, which was far more than electrons donated by Fe(II) oxidation (about 0.6 mmol/ L). Due to no Fe(II) oxidation in the absence of the

Water Air Soil Pollut (2015) 226:167

strains in controls, oxygen would be ruled out from the redox processes. Accordingly, Fe(II) and organic compounds were the electron donors, and As(V) and NO3− were the electron acceptors in this system. Since organic component and inorganic carbon components were not monitored during the experiments, the exact sequence of oxidation processes could not be characterized. Three hypotheses could be made for oxidation processes. One is that Fe(II) oxidation occurred prior to organic carbon oxidation, one is that organic carbon oxidation occurred prior to Fe(II) oxidation, and the last is that organic carbon oxidation and Fe(II) oxidation happened simultaneously. In each case, the electrons provided by organic compounds would be 3.3 mmol/L, since the number of electrons donated by Fe(II) oxidation was around 0.6 mmol/L. In the first case, Fe(II) oxidation was coupled with a part of NO3− reduction (0.3 mmol/L), and 1.1 mmol/L NO3− and 0.55 mmol/L As(V) were reduced by organic components. In the second case, NO3− reduction was coupled with a part of organic component oxidation (0.7 mmol/L organic C to CO2), and As(V) was reduced by both organic components (0.125 mmol/L organic C to CO2) and Fe(II) (0.6 mmol/ L Fe(II) to Fe(III)). In the last case, the exact processes of electron transfers could not be figured out. Whatever, the exact redox sequences and electron transfers should be investigated in more details in future studies.

5 Conclusions In the present work, As was rapidly removed from solution in both co-precipitation and adsorption experiments. Co-precipitation with biogenic ferrihydrite had higher removal efficiency for As(V) than As(III). However, As(III) was more efficiently removed than As(V) through adsorption on biogenic ferrihydrite. Arsenic(V) reduction was observed during coprecipitation with ferrihydrite. Arsenic(III) oxidation occurred in both As(III)-co-precipitating and As(III)adsorbing ferrihydrite. Biogenic ferrihydrite may be applied as an alternative remediation strategy to immobilize As. Acknowledgments The study is financially supported by the National Natural Science Foundation of China (Nos. 41222020 and 41172224), the National Key Basic Research Development Program (973 Program, No. 2010CB428804), the Fundamental Research Funds for the Central Universities (No. 2652013028), and the Fok Ying-Tung Education Foundation, China (Grant No.

Water Air Soil Pollut (2015) 226:167 131017). The authors would like to thank the Shanghai Synchrotron Radiation Facility (Beamline BL15U) and its staff (X.H. Yu and A.G. Li) for allowing us to perform the XANES analysis. Dr. G.H. Shi is much acknowledged for his help in FTIR analysis. Dr. Michael Kersten is much acknowledged for his constructive comments which significantly improved the quality of the manuscript.

References Alam, M. S., Wu, Y., & Cheng, T. (2014). Silicate minerals as a source of arsenic contamination in groundwater. Water, Air, & Soil Pollution, 225(11), 1–15. Amirbahman, A., Kent, D. B., Curtis, G. P., & Davis, J. A. (2006). Kinetics of sorption and abiotic oxidation of arsenic(III) by aquifer materials. Geochimica Et Cosmochimica Acta, 70(3), 533–547. Banerjee, K., Amy, G. L., Prevost, M., Nour, S., Jekel, M., Gallagher, P. M., & Blumenschein, C. D. (2008). Kinetic and thermodynamic aspects of adsorption of arsenic onto granular ferric hydroxide (GFH). Water Research, 42(13), 3371–3378. Berg, M., Luzi, S., Trang, P. T. K., Viet, P. H., Giger, W., & Stuben, D. (2006). Arsenic removal from groundwater by household sand filters: comparative field study, model calculations, and health benefits. Environmental Science & Technology, 40(17), 5567–5573. Bhandari, N., Reeder, R. J., & Strongin, D. R. (2011). Photoinduced oxidation of arsenite to arsenate on ferrihydrite. Environmental Science & Technology, 45(7), 2783– 2789. Bhattacharya, P., Welch, A. H., Stollenwerk, K. G., McLaughlin, M. J., Bundschuh, J., & Panaullah, G. (2007). Arsenic in the environment: biology and chemistry. Science of the Total Environment, 379(2–3), 109–120. Bissen, M., & Frimmel, F. H. (2003). Arsenic—a review. Part 1: occurrence, toxicity, speciation, mobility. Acta Hydrochimica Et Hydrobiologica, 31(1), 9–18. Dixit, S., & Hering, J. G. (2003). Comparison of arsenic(V) and arsenic(III) sorption onto iron oxide minerals: Implications for arsenic mobility. Environmental Science & Technology, 37(18), 4182–4189. Fuller, C. C., Davis, J. A., & Waychunas, G. A. (1993). Surface chemistry of ferrihydrite. Part 2. Kinetics of arsenate adsorption and coprecipitation. Geochimica Et Cosmochimica Acta, 57(10), 2271–2282. Goldberg, S., & Johnston, C. T. (2001). Mechanisms of arsenic adsorption on amorphous oxides evaluated using macroscopic measurements, vibrational spectroscopy, and surface complexation modeling. Journal of Colloid and Interface Science, 234(1), 204–216. Greenleaf, J. E., Cumbal, L., Staina, I., & SenGupta, A. K. (2003). Abiotic As(III) oxidation by hydrated Fe(III) oxide (HFO) microparticles in a plug flow columnar configuration. Process Safety and Environmental Protection, 81(B2), 87–98. Guo, H., Stueben, D., & Berner, Z. (2007a). Adsorption of arsenic(III) and arsenic(V) from groundwater using natural siderite as the adsorbent. Journal of Colloid and Interface Science, 315(1), 47–53.

Page 13 of 14 167 Guo, H., Stüben, D., & Berner, Z. (2007b). Removal of arsenic from aqueous solution by natural siderite and hematite. Applied Geochemistry, 22(5), 1039–1051. Guo, H., Ren, Y., Liu, Q., Zhao, K., & Li, Y. (2013a). Enhancement of arsenic adsorption during mineral transformation from siderite to goethite: mechanism and application. Environmental Science & Technology, 47(2), 1009–1016. Guo, H. M., Liu, C., Lu, H., Wanty, R., Wang, J., & Zhou, Y. Z. (2013b). Pathways of coupled arsenic and iron cycling in high arsenic groundwater of the Hetao basin, Inner Mongolia, China: an iron isotope approach. Geochimica Et Cosmochimica Acta, 112, 130–145. Han, X., Li, Y.-L., & Gu, J.-D. (2011). Oxidation of As(III) by MnO2 in the absence and presence of Fe(II) under acidic conditions. Geochimica Et Cosmochimica Acta, 75(2), 368–379. Harvey, C. F., Swartz, C. H., Badruzzaman, A. B. M., Keon-Blute, 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. (2002). Arsenic mobility and groundwater extraction in Bangladesh. Science, 298(5598), 1602–1606. Hohmann, C., Winkler, E., Morin, G., & Kappler, A. (2010). Anaerobic Fe(II)-oxidizing bacteria show as resistance and immobilize as during Fe(III) mineral precipitation. Environmental Science & Technology, 44(1), 94–101. Hug, S. J., & Leupin, O. (2003). Iron-catalyzed oxidation of arsenic(III) by oxygen and by hydrogen peroxide: pHdependent formation of oxidants in the Fenton reaction. Environmental Science & Technology, 37(12), 2734–2742. Islam, F. S., Gault, A. G., Boothman, C., Polya, D. A., Charnock, J. M., Chatterjee, D., & Lloyd, J. R. (2004). Role of metalreducing bacteria in arsenic release from Bengal delta sediments. Nature, 430(6995), 68–71. Jang, J.-H., & Dempsey, B. A. (2008). Coadsorption of arsenic(III) and arsenic(V) onto hydrous ferric oxide: effects on abiotic oxidation of arsenic(III), extraction efficiency, and model accuracy. Environmental Science & Technology, 42(8), 2893–2898. Jang, M., Min, S. H., Kim, T. H., & Park, J. K. (2006). Removal of arsenite and arsenate using hydrous ferric oxide incorporated into naturally occurring porous diatomite. Environmental Science & Technology, 40(5), 1636–1643. Jessen, S., Larsen, F., Koch, C. B., & Arvin, E. (2005). Sorption and desorption of arsenic to ferrihydrite in a sand filter. Environmental Science & Technology, 39(20), 8045–8051. Jia, Y. F., & Demopoulos, G. P. (2005). Adsorption of arsenate onto ferrihydrite from aqueous solution: influence of media (sulfate vs nitrate), added gypsum, and pH alteration. Environmental Science & Technology, 39(24), 9523–9527. Jia, Y., Xu, L., Wang, X., & Demopoulos, G. P. (2007). Infrared spectroscopic and X-ray diffraction characterization of the nature of adsorbed arsenate on ferrihydrite. Geochimica Et Cosmochimica Acta, 71(7), 1643–1654. Kappler, A., & Newman, D. K. (2004). Formation of Fe(III)minerals by Fe(II)-oxidizing photoautotrophic bacteria. Geochimica Et Cosmochimica Acta, 68(6), 1217–1226. Karim, M. (2000). Arsenic in groundwater and health problems in Bangladesh. Water Research, 34(1), 304–310. Katsoyiannis, I. A., & Zouboulis, A. I. (2004). Application of biological processes for the removal of arsenic from groundwaters. Water Research, 38(1), 17–26.

167

Page 14 of 14

Kleinert, S., Muehe, E. M., Posth, N. R., Dippon, U., Daus, B., & Kappler, A. (2011). Biogenic Fe(III) minerals lower the efficiency of iron-mineral-based commercial filter systems for As removal. Environmental Science & Technology, 45(17), 7533–7541. Konhauser, K. O., Kappler, A., & Roden, E. E. (2011). Iron in microbial metabolisms. Elements, 7(2), 89–93. Lin, S., Lu, D., & Liu, Z. (2012). Removal of As contaminants with magnetic γ-Fe2O3 nanoparticles. Chemical Engineering Journal, 211–212, 46–52. Liu, Q., Guo, H., Li, Y., & Xiang, H. (2013). Acclimation of arsenic-resistant Fe(II)-oxidizing bacteria in aqueous environment. International Biodeterioration & Biodegradation, 76, 86–91. Michel, F. M., Ehm, L., Antao, S. M., Lee, P. L., Chupas, P. J., Liu, G., Strongin, D. R., Schoonen, M. A. A., Phillips, B. L., & Parise, J. B. (2007). The structure of ferrihydrite, a nanocrystalline material. Science, 316(5832), 1726–1729. Morin, G., & Calas, G. (2006). Arsenic in soils, mine tailings, and former industrial sites. Elements, 2(2), 97–101. Myneni, S. C. B., Traina, S. J., Waychunas, G. A., & Logan, T. J. (1998). Experimental and theoretical vibrational spectroscopic evaluation of arsenate coordination in aqueous solutions, solids, and at mineral-water interfaces. Geochimica Et Cosmochimica Acta, 62(19-20), 3285–3300. Okibe, N., & Johnson, D. B. (2011). A rapid ATP-based method for determining active microbial populations in mineral leach liquors. Hydrometallurgy, 108(3), 195–198. Ona-Nguema, G., Morin, G., Wang, Y., Foster, A. L., Juillot, F., Calas, G., & Brown, G. E., Jr. (2010). XANES evidence for rapid arsenic(III) oxidation at magnetite and ferrihydrite surfaces by dissolved O-2 via Fe2+-mediated reactions. Environmental Science & Technology, 44(14), 5416–5422. O'Neil, M., Smith, A., Heckelman, P., & Budavari, S. (2001). The Merck index—an encyclopedia of chemicals, drugs, and biologicals (13th ed.). Whitehouse Station: Merck and Co, Inc. Paez-Espino, D., Tamames, J., de Lorenzo, V., & Canovas, D. (2009). Microbial responses to environmental arsenic. Biometals, 22(1), 117–130. Pal, P., Chakrabortty, S., & Linnanen, L. (2014). A nanofiltration– coagulation integrated system for separation and stabilization of arsenic from groundwater. Science of the Total Environment, 476–477, 601–610. Pedersen, H. D., Postma, D., & Jakobsen, R. (2006). Release of arsenic associated with the reduction and transformation of iron oxides. Geochimica Et Cosmochimica Acta, 70(16), 4116–4129.

Water Air Soil Pollut (2015) 226:167 Picardal, F. W., Zaybak, Z., Chakraborty, A., Schieber, J., & Szewzyk, U. (2011). Microaerophilic, Fe(II)-dependent growth and Fe(II) oxidation by a Dechlorospirillum species. Fems Microbiology Letters, 319(1), 51–57. Ravel, B., & Newville, M. (2005). ATHENA, ARTEMIS, HEPH AESTUS: data analysis for X-ray absorption spectroscopy using IFEFFIT. Journal of Synchrotron Radiation, 12, 537– 541. Raven, K. P., Jain, A., & Loeppert, R. H. (1998). Arsenite and arsenate adsorption on ferrihydrite: kinetics, equilibrium, and adsorption envelopes. Environmental Science & Technology, 32(3), 344–349. Root, R. A., Dixit, S., Campbell, K. M., Jew, A. D., Hering, J. G., & O'Day, P. A. (2007). Arsenic sequestration by sorption processes in high-iron sediments. Geochimica Et Cosmochimica Acta, 71(23), 5782–5803. Silver, S., & Phung, L. T. (2005). Genes and enzymes involved in bacterial oxidation and reduction of inorganic arsenic. Applied and Environmental Microbiology, 71(2), 599–608. Smedley, P. L., & Kinniburgh, D. G. (2002). A review of the source, behaviour and distribution of arsenic in natural waters. Applied Geochemistry, 17(5), 517–568. Tamura, H., Goto, K., Yotsuyanagi, T., & Nagayama, M. (1974). Spectrophotometric determination of iron(II) with 1,10phenanthroline in the presence of large amounts of iron(III). Talanta, 21(4), 314–318. Thirunavukkarasu, O. S., Viraraghavan, T., & Subramanian, K. S. (2003). Arsenic removal from drinking water using iron oxide-coated sand. Water, Air, and Soil Pollution, 142(1-4), 95–111. Tokoro, C., Yatsugi, Y., Koga, H., & Owada, S. (2010). Sorption mechanisms of arsenate during coprecipitation with ferrihydrite in aqueous solution. Environmental Science & Technology, 44(2), 638–643. Tufano, K. J., & Fendorf, S. (2008). Confounding impacts of iron reduction on arsenic retention. Environmental Science & Technology, 42(13), 4777–4783. Voegelin, A., & Hug, S. J. (2003). Catalyzed oxidation of arsenic(III) by hydrogen peroxide on the surface of ferrihydrite: An in situ ATR-FTIR study. Environmental Science & Technology, 37(5), 972–978. Yang, L., Li, X., Chu, Z., Ren, Y., & Zhang, J. (2014). Distribution and genetic diversity of the microorganisms in the biofilter for the simultaneous removal of arsenic, iron and manganese from simulated groundwater. Bioresource Technology, 156, 384–388. Zhao, Z., Jia, Y., Xu, L., & Zhao, S. (2011). Adsorption and heterogeneous oxidation of As(III) on ferrihydrite. Water Research, 45(19), 6496–6504.