You are on page 1of 6

Letters in Applied Microbiology 2002, 34, 258262

Arsenic (III) oxidizing Microbacterium lacticum and its use in the treatment of arsenic contaminated groundwater
S.A. Mokashi and K.M. Paknikar
Division of Microbial Sciences, Agharkar Research Institute, Pune, India
2001/271: received 13 September 2001 and accepted 7 January 2002

S . A . M O K A S H I A N D K . M . P A K N I K A R . 2002.

Aims: To develop a microbially-assisted process for the removal of arsenic from contaminated groundwater. Methods and Results: A culture of Microbacterium lacticum oxidizing up to 50 mmol l1 arsenic (III) was isolated from municipal sewage by an enrichment culture technique. Using culture immobilized on brick pieces and packed in a glass column, complete oxidation of As (III) from groundwater could be quickly achieved at neutral pH and ambient temperature with methanol as substrate. The oxidized As species were removed from groundwater using three different methods: zero valent iron, activated charcoal and ferric chloride. Conclusions: The oxidation of groundwater As (III) by a M. lacticum-immobilized column, followed by its removal using activated carbon, could be an efcient method for the treatment of As (III)-contaminated groundwater. Signicance and Impact of the Study: The study will be useful in developing a combined microbiologicalchemical process for treating arsenic-contaminated groundwater.

INTRODUCTION Groundwater has been one of the safest sources of drinking water, especially in rural areas where piped water is not available. However, it is now recognized that in West Bengal (India) and Bangladesh, more than 40 million people are drinking arsenic (As)-contaminated water with levels of As above 005 mg l1, and around 3 million people are suffering from As toxicity (Das et al. 1995; Karim 2000; Yokota et al. 2001). It is reported by the US Environmental Protection Agency (US EPA) that As is a persistent bioaccumulative carcinogen and is among the top 20 toxic substances identied by the US Agency of Toxic Substances and Disease Registry. So far, no treatment is available for chronic As toxicity. Therefore, the provision of As-free water is essential. The known treatment processes for As removal include coagulation, ltration, lime softening, activated alumina adsorption, ion exchange, reverse osmosis, electrodialysis reversal and nanoltration (US EPA 2001). Although

Correspondence to: Dr K.M. Paknikar, Division of Microbial Sciences, Agharkar Research Institute, G.G. Agarkar Road, Pune 411 004, India (e-mail: paknikar@vsnl.com).

these processes are effective in removing about 8095% As from solutions, they have high operating costs. Furthermore, oxidation of As (III) to As (V) is a prerequisite for all the conventional treatment processes (Khoe et al. 1997). The rate of oxidation of As (III) to As (V) by oxygen is extremely slow and hence, stronger and costly oxidants such as chlorine, hydrogen peroxide, ozone etc. are needed as part of the As removal process (Kim and Nriagu 1999). Other complications include the need for special equipment due to operating parameters, such as high pressure, elevated temperature, use of electrolysis techniques etc. Therefore, it would be advantageous if a method could be developed that is cheaper and easy to operate. In the above context, microbiological methods for oxidation of As offer considerable promise. The oxidized As species can then be removed by any of the conventional methods. The present paper describes a Microbacterium lacticum strain, isolated from municipal sewage, capable of oxidizing As (III) at an exceptionally high rate. The paper also describes a laboratory-scale process for the oxidation of As (III)-contaminated groundwater in an immobilized cell reactor containing M. lacticum, followed by the removal of oxidized As using zero valent iron, activated charcoal or ferric chloride.
2002 The Society for Applied Microbiology

REMOVAL OF ARSENIC FROM GROUNDWATER

259

MATERIALS AND METHODS Culture used and preparation of inoculum A bacterial culture capable of oxidizing As (III) was isolated from a municipal sewage sample by the enrichment culture technique. The enrichment was set up in a 4 l capacity glass reactor lled with sewage sample supplemented with 05 mg l1 As (III). The culture was identied as Microbacterium lacticum using cultural and biochemical tests (Jones and Collins 1986) and conrmed by 16 S rDNA sequencing as described by Weeger et al. (1999). For the preparation of inoculum, the culture was grown in 150 ml Erlenmeyer asks containing 50 ml buffered Enrichment and Growth (EG) medium (composition: 003 g l1 NH4Cl, 001 g l1 NaCl, 001 g l1 MgSO4, 015 g l1 yeast extract and 05 g l1 peptone, pH 75) supplemented with 10 mmol l1 sodium arsenite. The asks were incubated at 30C in an orbital shaker incubator (Gallenkamp, Loughborough, UK) for 24 h at 100 rev min1. The medium was centrifuged at 23 800 g for 10 min, the cell pellet was washed twice with Tris-HCl buffer (pH 75) and resuspended in physiological saline to give a protein concentration of 216 mg l1, which equates to a titre of 109 cfu ml1. Inocula for subsequent experiments were taken from this suspension. Minimum inhibitory concentrations (MIC) The MIC values for As (III) and Fe (III) for M. lacticum were checked using a 1% inoculum into EG medium containing either sodium arsenite (01100 mmol l1) or ferric chloride (0008008 mmol l1). The experimental sets were incubated at 30C for 24 h. Groundwater sample A groundwater sample was collected from a known As (III)contaminated tube well from a village in Bangladesh. The sample was analysed for total As content and other physical chemical parameters (see below). A simulated sample of groundwater was freshly prepared based on the results of this analysis and used in further experiments. Factors affecting As oxidation Factors affecting As (III) oxidation were investigated in a series of batch culture experiments using EG medium asks as described above. The effects of various factors, such as pH (55105), temperature (2045C), inoculum size in terms of protein concentration (0170216 mg l1) and carbon sources (acetate, lactate, citrate, methanol, sucrose and glucose, 15 mmol l1 each), were checked by monitoring As oxidation over a period of 24 h.

Immobilization of cells and oxidation of As (III)-contaminated water in continuous mode The cells were immobilized on brick pieces by mixing 130 ml log phase culture (109 cfu ml1) grown in EG medium with 200 g brick pieces (average size 23 cm diameter) in a glass column (125 cm height 4 cm diameter) for 24 h. The glass column was sparged with air. This cycle was repeated 15 times to obtain a visible biolm of the culture on the brick pieces. Simulated groundwater containing As (III) was continuously passed through the glass column in an upow mode at ambient temperature. During the total 6 weeks operation of the reactor, four groundwater samples containing incremental concentrations of As (III), i.e. 05, 1, 5 and 10 mg l1, were used. The ow rate of the groundwater was regulated by a programmable peristaltic pump (Ismatec, Glattbrugg, Zurich, Switzerland) to ensure complete oxidation of As (III). Corresponding hydraulic retention times, i.e. volume of the reactor (130 ml)/ow rate ml min1, were calculated. The efuent samples from the reactor were collected daily to determine the pH and for qualitative estimation of As species by microplate silver nitrate (MSN) assay. The bacterial count of the treated efuent was determined in terms of total viable count using nutrient agar. As removal using zero valent iron, activated charcoal and ferric chloride Aliquots (50 ml) of the treated (efuent) groundwater from the immobilized cell reactor were contacted with the three reagents, zero valent iron, activated charcoal and ferric chloride, in separate batch experiments. Experiments with zero valent iron (Sisco Laboratories, Mumbai, India) were performed as described by Su and Puls (2001). Activated charcoal experiments were carried out using a glass column packed with 10 g activated charcoal (phosphorous free, 300 mesh, 2% ash content; s.d ne chemicals, Mumbai, India), and ferric chloride (Sarabhai Chemicals, Baroda, India) experiments were performed by mixing the efuent with varying concentrations (ranging between 2 and 500 mg l1) of the reagent. The efciency of the three methods was compared with respect to time required for total removal of As (V). Other physical and chemical characteristics, such as colour, turbidity, pH, sulphates, chlorides, sodium, Total Dissolved Solids (TDS), hardness and bacterial count of the treated groundwater (TVC), were also considered. Analyses For quantitative estimation of As species, 20 ll aliquots were spotted on Whatman paper no. 3. The two As species were separated by running a chromatograph in an

2002 The Society for Applied Microbiology, Letters in Applied Microbiology, 34, 258262

260 S . A . M O K A S H I A N D K . M . P A K N I K A R

isopropanol:water (7:3) solvent system. The two forms of As, As (III) and As (V), were distinguished by 01 mol l1 silver nitrate reagent (violet spot for arsenate and yellow spot for arsenite). The two spots were cut and eluted in a 10:4:1 mixture of concentrated HCl, KI (15%) and SnCl2 (40%). As (III) and As (V) concentrations in the eluants were determined by silver diethyl dithiocarbamate assay (Greenberg et al. 1992). Qualitative estimation of As species in the experimental samples was carried out by a microplate silver nitrate (MSN) assay. In a standard 96-well microplate, 150 ll of the sample were mixed with 150 ll 01 mol l1 silver nitrate. The microplate was observed periodically for development of a coloured precipitate. Arsenite, upon reaction with silver nitrate, is known to give a yellow precipitate of silverorthoarsenite, while arsenate generates a dark red or violet coloured precipitate of silver-orthoarsenate. The assay was validated by comparing the test results with those obtained by HPLC separation of As species, followed by analysis on an Inductively Coupled Plasma Atomic Emission Spectrophotometer (ICP-AES, Jobin-Yvon, Langjumeau, France) as described by (Weeger et al. 1999). Total protein estimation was carried out according to the method of Lowry et al. (1951). Chlorides, sulphates and sodium were estimated by argentometry, turbidimetry and atomic emission spectrophotometry, respectively (Greenberg et al. 1992). RESULTS The M. lacticum culture used in the present study exhibited MIC values of 50 mmol l1 and 002 mmol l1 for As (III) and Fe (III), respectively. The optimum pH for oxidation of As (III) by M. lacticum was 75.
12

Oxidation of As (III) by M. lacticum was signicantly inuenced by various factors. Optimum pH and temperature for As (III) oxidation was found to be 75 and 30C, respectively. Under optimal pH and temperature conditions, within the range of inoculum density tested, there was a direct proportional relationship between inoculum density and oxidation efciency. Complete oxidation of As occurred in 5 h with an inoculum density of 216 mg protein, which corresponded to 109 cfu ml1. With regard to the carbon source, the culture could utilize different carbon sources such as citrate, acetate and, to some extent, sucrose. However, effective utilization of methanol and lactate was observed, with the latter giving maximum oxidation efciency (87%) at a concentration of 15 mmol l1. Therefore, methanol was selected as a source of carbon in further studies. The culture could completely oxidize As (III) concentrations in the range 05 to 3500 mg l1 with methanol as substrate. In the immobilized cell reactor operated in continuous mode, the culture could completely oxidize 05, 1, 5 and 10 mg l1 As (III) in 25, 60, 330 and 600 s under optimum conditions, as determined by MSN assay. The performance of the reactor over a period of 40 days (10 days per concentration) with respect to inuent As (III) concentration and hydraulic retention time (HRT) is shown in Fig. 1. The Total Viable Count (TVC) of the treated water was 103 cfu ml1. The data presented in Table 1 show that >999% As (V) could be removed using the zero valent iron, activated charcoal and ferric chloride methods. Of the three methods, ferric chloride precipitation was the fastest. With the activated charcoal and ferric chloride precipitation methods, the pH of the treated groundwater remained unaltered, whereas in the zero valent iron method, the pH of the
700 600 500 Hydraulic retention time (s)

10 As (III) concentration (mg l1)

8 400 6 300 4 200 2 0 1 5 9 13 17 21 25 29 33 Days 37 100 0

Fig. 1 Performance of an immobilized cell reactor with respect to As (III) concentration () and hydraulic retention time ()

2002 The Society for Applied Microbiology, Letters in Applied Microbiology, 34, 258262

REMOVAL OF ARSENIC FROM GROUNDWATER

261

Table 1 Comparison of results of As (V) removal by conventional methods after treating immobilized cell reactor Reactor efuent after treatment with Parameters As (V) % As (V) removal Treatment time pH Colour Turbidity Sulphates Chlorides Sodium TDS Hardness as CaCO3 Bacterial count Permissible limits 005 mg l)1 6585 15 TCU 5 NTU 400 mg l)1 250 mg l)1 200 mg l)1 1000 mg l)1 500 mg l)1 Absent in 100 ml Reactor efuent before treatment 05 mg l)1 72 Colourless Turbid 271 mg l)1 110 mg l)1 69 mg l)1 506 mg l)1 237 mg l)1 103 cells ml)1 Zero valent iron <005 mg l)1 >999% Immediate 713 Colourless Clear 93 mg l)1 126 mg l)1 85 mg l)1 508 mg l)1 69 mg l)1 Nil Activated charcoal <005 mg l)1 >999% 96 h 80 Colourless Clear 121 mg l)1 93 mg l)1 65 mg l)1 487 mg l)1 102 mg l)1 Nil Ferric chloride <005 mg l)1 >999% 25 h 72 Colourless Clear 105 mg l)1 79 mg l)1 52 mg l)1 34 mg l)1 58 mg l)1 Nil

TCU: True Colour Units; NTU: Nephelometric Turbidity Units; TDS: Total Dissolved Solids.

treated groundwater increased from 72 to 80. None of the methods imparted any colour, turbidity or hardness to the treated groundwater. Other parameters, i.e. TDS, sulphates, chlorides and sodium, were also found to be well within the permissible limits prescribed by Greenberg et al. (1992). In the case of the ferric chloride precipitation method, the levels of Fe (III) in the treated groundwater were found to be approximately 15 mg l1. Bacterial counts of the treated groundwater samples were zero using all the three methods. DISCUSSION An MIC value of 500 mg l1 has been reported for a Zooglea strain isolated from As-contaminated water (Weeger et al. 1999). Trevors et al. (1985) have reported bacterial strains resistant in the range of 10952660 mmol l1. However, the strains were unable to oxidize As at these high concentrations. Microbacterium lacticum described in this paper not only has the highest MIC value for As (III) amongst the bacterial species reported so far, it is also able to retain the As (III) oxidizing capacity at this concentration (50 mmol l1). This property of the culture could be signicant in the treatment of As-contaminated waters which have high concentrations of As, such as those that result from mining operations. The pH of the completely treated (As-free) groundwater was 72. Thus, readjustment of the water pH would not be required prior to consumption. Most of the studies using microbial As oxidation have used pH values ranging from 68 to 73. As reported by Phillips and Taylor (1976), after complete oxidation of As (III) by Alcaligenes faecalis, the pH of the medium dropped from 70 to <50, which, being acidic, would need readjustment. In the case of conventional

treatment processes using chlorine, hydrogen peroxide or ozone for As (III) oxidation prior to its removal by adsorption or precipitation, the pH of the treated water requires readjustment before its use or disposal. Considering these drawbacks of the conventional methods, oxidation of As (III) using M. lacticum would appear to be highly advantageous. Microbacterium lacticum could bring about complete oxidation of As (III) at 30C. Therefore, it would not be necessary to maintain the reactor temperature, as 30C could be the average ambient temperature encountered in tropical countries such as India and Bangladesh. Microbacterium lacticum has the potential to efciently oxidize As. The rate of transformation was directly dependent on cell density. When 109 cfu ml1 were used in ask experiments, complete oxidation of 133 mmol l1 (100 mg l1) As (III) was obtained in 40 min. Weeger et al. (1999) reported that a species of Zooloea could completely oxidize 133 mmol l1 (100 mg l1) As (III) in 25 h at a cell density of 109 cfu ml1. Thus, M. lacticum has the highest As (III) oxidizing efciency among the strains reported so far. The culture is also able to utilize a variety of carbon sources such as lactate, citrate and methanol. Considering all these properties, M. lacticum could be an appropriate choice for oxidation of As (III) from aqueous solutions. Results of the As (V) removal experiments using zero valent iron, activated charcoal and ferric chloride indicated that all three methods could effectively remove As from groundwater. However, treatment of As (V)-contaminated groundwater with zero valent iron removed As at a very slow rate. Removal of As (V) using ferric chloride was faster than with activated charcoal, but ferric iron reacts with As (V) to

2002 The Society for Applied Microbiology, Letters in Applied Microbiology, 34, 258262

262 S . A . M O K A S H I A N D K . M . P A K N I K A R

form a precipitate of FeAsO4. This secondary sludge containing As could pose a disposal problem, and would preclude the use of this method in any household treatment process, such as lters. Experiments with activated charcoal indicated that As (V) removal is achieved at a relatively faster rate than with zero valent iron, and the pH of the treated water remained almost unaltered. The advantage of this method over the ferric chloride method is that activated charcoal could be packed in a suitable column and used as a lter for household purposes. After saturation, the column could be replaced, or adsorbed As could be desorbed using the phosphate displacement reaction (Su and Puls 2001) and the same column re-used. The particle size of the activated charcoal used in the batch-type experiments was 300 mesh, which would be unsuitable for use on a continuous scale due to the problem of clogging of the column. This problem could, however, be overcome by using activated charcoal of a larger particle size. Treatment of As-containing wastewater by activated carbon has been reported to be an effective technique for the removal of As species. Huang and Fu (1984) and Huang and Vane (1989) found that As removal could be best accomplished in the As (V) state. Based on As (III) oxidation and activated charcoal adsorption studies, a prototype lter for As removal from groundwater is suggested. The lter could consist of three columns, the rst column containing an immobilized M. lacticum culture capable of oxidizing As (III), followed by a second column of activated charcoal which could adsorb As (V). The half-life of the immoblized M. lacticum column is expected to be at least 40 days. If necessary, such a column could be revived by providing easily assimilable substrates such as methanol. In order to remove bacterial cells from the treated groundwater, an additional column could be provided for ultraviolet/membrane lter treatment of the water. Extensive, expanded, laboratory-scale studies of this model are necessary prior to actual use at the eld site. ACKNOWLEDGEMENTS The authors wish to thank Prof. Marie-Claire Lett, Universite Louis Pasteur, Strasbourg for help in 16 S rDNA sequencing of the culture. Part of this work was carried out under research project no. 60 (0027)/98/EMR-II sponsored by the Council of Scientic and Industrial Research, New Delhi. KMP is thankful to Prof. Dipankar Chakrabarty, School of Environmental Sciences, Jadavpur

University for providing an opportunity for collection of groundwater samples. REFERENCES


Das, D., Chatterjee, A., Mandal, B.K., Samanta, G. and Chakraborti, D. (1995) Arsenic in ground water in six districts of West Bengal, India: the biggest Arsenic calamity in the world. Part 2. Arsenic concentration in drinking water, hair, nails, urine, skin-scale and liver tissue (Biopsy) of the affected people. Analyst 120, 917924. Greenberg, A.E., Clesceri, L.S. and Eaton, A.D. (1992) Standard Methods for Examination of Water and Wastewater 18th edn. Washington D.C.: American Public Health Association. Huang, C.P. and Fu, P.L.K. (1984) Treatment of arsenic (V)containing water by activated carbon. Journal of Water Pollution Control Federation 56, 233. Huang, C.P. and Vane, L.M. (1989) Enhancing As (V) removal by Fe (II)-treated activated carbon. Journal of Water Pollution Control Federation 61, 1596. Jones, D. and Collins, M.D. (1986) Irregular, non-sporing, Grampositive rods. In Bergeys Manual of Systematic Bacteriology Vol. II, ed. Sneath, P.H.A., Mair, N.S., Sharp, M.E. and Holt, J.G. pp. 12611418. Baltimore: Williams & Wilkins. Karim, M. (2000) Arsenic in groundwater and health problems in Bangladesh. Water Research 34, 304310. Khoe, G.H., Emett, M.T. and Robins, R.G. (1997) Photoassisted oxidation of species in solution. United States Patent no. 5 688 378. Kim, M.J. and Nriagu, J. (1999) Oxidation of arsenite in groundwater using ozone and oxygen. Science of the Total Environment 247, 7179. Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) Protein measurement with Folin phenol reagent. Journal of Biological Chemistry 193, 265275. Mallick, S. and Rajgopal, N.R. (1996) Groundwater development in the arsenic-affected alluvial belt of West Bengal Some questions. Current science 70, 956958. Phillips, S.E. and Taylor, M.L. (1976) Oxidation of arsenite to arsenate by Alcaligenes faecalis. Applied and Environmental Microbiology 32, 392399. Su, C. and Puls, R.W. (2001) Arsenate and arsenite removal by zerovalant iron: kinetics, redox transformation and implications for in Situ groundwater remediation. Environmental Science and Technolnology 35, 14871492. Trevors, J.T., Oddie, K.M. and Belliveau, B.H. (1985) Metal resistance in Bacteria. FEMS Microbiological Review 32, 3954. Weeger, W., Lievremont, D., Perret, M. Lagarde, F., Hubert, J.C., Leroy, M. and Lett, M.C. (1999) Oxidation of arsenite to arsenate by a bacterium isolated from an aquatic environment. Biometals 12, 141149.

2002 The Society for Applied Microbiology, Letters in Applied Microbiology, 34, 258262

Letters in Applied Microbiology 2002, 35, 171

Erratum
Mokashi, S.A. & Paknikar, K.M. (2002) Arsenic (III) oxidizing Microbacterium lacticum and its use in the treatment of arsenic contaminated groundwater. Letters in Applied Microbiology 34, 258262.

The publishers regret an error which has appeared on page 261 of the above paper. Table 1 is incorrect. The correct Table 1 is given below.
Table 1 Comparison of results of as (V) removal by conventional methods after treating immobilized cell reactor Reactor efuent before treatment 05 mg l)1 72 Colourless Turbid 271 mg l)1 110 mg l)1 69 mg l)1 506 mg l)1 237 mg l)1 103 cells ml)1 Reactor efuent after treatment with Zero valent iron < 005 mg l)1 > 999% 96 h 80 Colourless Clear 121 mg l)1 93 mg l)1 65 mg l)1 487 mg l)1 102 mg l)1 Nil Activated charcoal < 005 mg l)1 > 999% 25 h 72 Colourless Clear 105 mg l)1 79 mg l)1 52 mg l)1 34 mg l)1 58 mg l)1 Nil Ferric chloride < 005 mg l)1 > 999% Immediate 713 Colourless Clear 93 mg l)1 126 mg l)1 85 mg l)1 508 mg l)1 69 mg l)1 Nil

Parameters As (V) % As (V) removal Treatment time pH Colour Turbidity Sulphates Chlorides Sodium TDS Hardness as CaCO3 Bacterial count

Permissible limits 005 mg l)1 6585 15 TCU 5 NTU 400 mg l)1 250 mg l)1 200 mg l)1 1000 mg l)1 500 mg l)1 Absent in 100 ml

TCU: True Colour Units; NTU: Nephelometric Turbidity Units; TDS: Total Dissolved Solids.

2002 The Society for Applied Microbiology