Optimization of the removal of arsenic from groundwater using ion exchange C.N. Mulligan, A.K.M. Saiduzzaman & J. Hadjinicolaou Department of Building, Civil and Environmental Engineering, Concordia University, Montreal, Quebec, Canada ABSTRACT: Various technologies have been evaluated for removal of arsenic from the ground- water. Ion exchange using anion exchange resins is used primarily for water treatment with low amounts of sulfate. Oxidation of this form to As(V) is required before removal. In addition, the production and disposal of spent regenerants and media has also not been studied extensively. Although residual production needs to be minimized as much as possible, there is inadequate data regarding the amounts and compositions of the residuals generated by ion exchange processes and methods for disposal. This study involves the optimization of the ion exchange process and the conditions for regeneration of the ion exchange media. Removal efficiency increased as the pH decreased from 8.5 to 6.5, concentration increased from 500 to 1600 ppb and velocity decreased from 1.94 cm/min to 0.65 cm/min. Regeneration of the ion exchange media enabled desorption of 97% of the adsorbed arsenic. 1 INTRODUCTION 1.1 Arsenic groundwater contamination The use of groundwater as a source of potable water supply presents serious problems in many coun- tries including Bangladesh, India, Thailand, Japan, China, US, and Canada, amongst others. Until recently, it was believed that in most cases, the groundwater is pure enough for drinking, while the sur- face waters must be treated (purified) before drinking. Therefore, groundwater is preferred. The exist- ence of arsenic and other groundwater issues in these countries have changed things dramatically. All through the past three decades there was a move forward to reduce the serious mortality, predominantly among infants, caused by diseases in surface waters in Bangladesh and other coun- tries. International aid agencies such as UNICEF became involved in funding the drilling of shal- low tubewells to gain access to groundwater for domestic supply, which was, uncontaminated by bacteria and otherwise believed to be clean. More than 10 million wells were constructed and it was greatly successful in reducing child death in Bangladesh. Although arsenic is a widely distributed element in the earth’s crust, but it was not generally found in a water-soluble form and thus does not cause a risk to the safety of drinking water sup- plies (Kinniburgh & Smedley 2001). Arsenic problems have long been recognized to occur in sulphide-rich metaliferous strata (principally, therefore, in particular mining areas) and in some geothermal areas. There were also reports in the literature of arsenic occurring in some arid or semi-arid inland basins, for example in parts of Argentina and the United States but the existence of arsenic in soluble form in the anaerobic groundwater of alluvial and deltaic plains was not by and large recognized until 1995. Arsenic containing minerals include arsenopyrite (FeAs), realgar (AsS), orpiment (As 2 S 3 ), niccolite (NiAs) and cobalite (CoAsS) (Boyle & Jonasson 1973). Arsenic is a contaminant that originates from and is transported to natural waters through erosion and dissolution of arsenic-containing rocks and soil. Climate and redox potential are significant fac- tors in the transport of arsenic. Arsenic sulfides can be oxidized, releasing arsenic to the environment. 237 Natural Arsenic in Groundwater: Occurrence, Remediation and Management – Bundschuh, Bhattacharya and Chandrasekharam (eds) © 2005, Taylor & Francis Group, London, ISBN 04 1536 700 X Copyright © 2005 Taylor & Francis Group plc, London, UK Arsenic occurs in natural waters in both organic and inorganic forms but the inorganic arsenic is the most prevalent and is the most likely to exist in significant concentrations. The valence and species of inorganic arsenic are dependent on the oxidation–reduction conditions and the pH of the water. Environmental arsenic is mainly found in two forms: arsenic V (arsenate), and arsenic III (arsenite). Arsenic V is the common oxidized state found in surface water and some groundwater sources. Arsenic III is un-oxidized and mainly found in deep anaerobic groundwater sources. 1.2 Treatment technologies Due to the problems of arsenic in the groundwater, economic solutions need to be found to ensure the safety of the drinking water. Several common treatment technologies are used for removal of inorganic contaminants, including arsenic, from drinking water supplies. Large-scale treatment facil- ities often use conventional coagulation with alum or iron salts followed by filtration to remove arsenic. Lime softening and iron removal also are common, conventional treatment processes that can potentially remove arsenic from source waters. In small communities small-scale systems often use ion exchange adsorption because of their ease of handling and sludge-free operations (Clifford 1999). Treatment options identified by EPA include ion exchange, reverse osmosis, activated alumina, nanofiltration, electrodialysis reversal, coagulation/filtration, lime softening, greensand filtration and other iron/manganese removal processes, and emerging technologies not yet identi- fied (USEPA 2003). Pre-oxidation technology includes chlorination, potassium permanganate, and ozone, although aeration over a significant time period is also possible (Wang et al. 2000). Both chlorine and potas- sium permanganate can oxidize arsenite to arsenate within a minute (pH 6.3 to 8.3). Chlorine is low cost but can lead to formation of disinfection products, whereas permanganate is not a disin- fectant does not form byproducts, can be regenerated but is more costly. Ion Exchange (IX) can effectively remove arsenic using anion exchange resins. It is recom- mended as a BAT (best available technology) primarily for sites with low sulfate because sulfate is preferred over arsenic. Sulfate will compete for binding sites resulting in shorter run lengths (USEPA 2003). Nitrate and nitrite may cause some problems since they will also adsorb onto the fresh media and then are desorbed by arsenic. These increased levels could be hazardous. Therefore, ion exchange is recommended for sulfate levels less than 50mg/L and nitrite and nitrate levels less than 5 mg/L. Total dissolved solids should also be less than 500 mg/L. Activated Alumina (AA) is an effective arsenic removal technology; however, the capacity of acti- vated alumina to remove arsenic is very pH sensitive. It is limited to a pH range 5.5 to 6. High removals can be achieved over a broad range of pH, but shorter run lengths will be observed at higher pH. Because arsenic is strongly adsorbed to the media, only about 50–70% of the adsorbed arsenic is removed. The brine stream produced by the regeneration process then requires disposal (USEPA 2003). Reverse Osmosis (RO) can provide arsenic removal efficiencies of greater than 95% when the operating pressure is ideal. Water rejection (on the order of 20–25%) may be an issue in water- scarce regions and may prompt systems employing RO to seek greater levels of water recovery. Water recovery is the volume of drinking water produced by the process divided by the influent stream (product water/influent stream). Increased water recovery is often more expensive, since it can involve recycling of water through treatment units to allow more efficient separation of solids from water. This can also produce more concentrated solid wastes. However, the waste stream will generally not be as concentrated as anion exchange brines, so it should be easier to dispose of. Although reverse osmosis is listed as a BAT as it removes over 95% arsenic, it was not used to develop national costs because other options are more cost effective and have much smaller waste streams (USEPA 2003). It could however, be cost effective if it is utilized for multipurposes such as removal of total dissolved solids (TDS) and arsenic. Modified Coagulation/Filtration (C/F) is an effective existing treatment process for removal of As(V) according to laboratory, pilot-plant, and full-scale tests. The type of coagulant and dosage used affects the efficiency of the process. Below a pH of approximately 7, removals with alum or ferric sulfate/chloride are similar. Above a pH of 7, removals with alum decrease dramatically 238 Copyright © 2005 Taylor & Francis Group plc, London, UK (at a pH of 7.8, alum removal efficiency is about 40%). Other coagulants are also less effective than ferric sulfate/chloride. Systems may need to lower the pH or add more coagulant to achieve higher removals over 90% (USEPA 2003). Enhanced Lime Softening (ELS), operates optimally at a greater than pH 10.5 is likely to pro- vide a high percentage of arsenic removal. Systems operating lime softening at lower pH values will need to increase the pH to achieve higher removals of arsenic (USEPA 2003). Magnesium addition may also be required. Sludge production is increased compared to lime softening if arsenic removal is required. Electrodialysis Reversal (EDR) can produce effluent water quality comparable to reverse osmo- sis. EDR systems are fully automated, require little operator attention, and do not require chem- ical addition. EDR systems, however, are typically more expensive than nanofiltration and reverse osmosis systems. It should be noted that while electrodialysis reversal is listed as a BAT, it was not used to develop national costs because other options are more cost effective and have much smaller waste streams (USEPA 2003). Oxidation/Filtration (including greensand filtration) has an advantage in that there is not as much competition with other ions. Arsenic is co-precipitated with the iron during iron removal. Sufficient iron needs to be present to achieve high arsenic removals. One study recommended a 20:1 iron to arsenic ratio. Removals of approximately 80% were achieved when iron to arsenic ratio was 20:1. When the iron to arsenic ratio was lowered (7:1), removals decreased below 50%. The presence of iron in the source water is critical for arsenic removal. If the source water does not contain iron, oxidizing and filtering the water will not remove arsenic. When the arsenic is pres- ent as As(III), sufficient contact time needs to be provided to convert the As(III) to As(V) for removal by the oxidation/filtration process (USEPA 2003). The removal of arsenic from groundwater has been studied but the inadequate information regarding mechanisms of arsenic removal exists. Ion exchange is not efficient for As(III) removal. Oxidation of this form to As(V) is required before removal. In addition, the production and dis- posal of spent regenerants and media has also not been studied extensively. Although regenerant production needs to be minimized as much as possible, there is inadequate data regarding the amounts and compositions of the residuals generated by ion exchange processes and methods for disposal. This study involves the determination of the mechanisms of arsenic removal by ion exchange and the characterization of the residuals produced by ion exchange to optimize the conditions for regeneration of the ion exchange media. Factors affecting ion exchange capacity are examined. 2MATERIALS AND METHODS 2.1 Experimental setup The treatment process consists of an oxidizing filter followed by an ion exchange column. The effect of pH, arsenate concentration and filtration velocity were examined using a simulated Bangladeshi groundwater. The investigation procedure was as follows. The investigation was carried out in the laboratory on a fixed unit incorporating dynamic columns with Filox-R and the anionic-exchange media Purolite A-300 (chloride ion form) serving as oxidizing and ion-exchanging filters respectively. These materials were purchased from Magnor Inc., Boucherville, Quebec, Canada. Purolite A-300 was a Type II, strongly basic gel anion exchange resin. Whole bead counts are a minimum of 92% clear beads with mechanical strengths ranging over 300 grams. Particle size ranged from ϩ16 mesh Ͻ5% to 50 mesh Ͻ1%. It is unaffected by dilute acids, alkalies, and most solvents. Thus an oxidizing filter was used followed by an anion filtration system. Figure 1A is a schematic diagram of the treatment process used, which consists of the following major elements: Intake: Raw water was pumped from a three-liter flask and flows through the oxidizing filter. A MasterFlex L/S pump with a controller was used. Flow rates were adjusted with the speed control potentiometer as well as an additional flow meter. 239 Copyright © 2005 Taylor & Francis Group plc, London, UK Oxidizing filter: A plastic syringe was installed, which serves as the oxidizing filter column, to oxidize As(III) to As(V). The oxidizing filter column is a ten milliliter and 1.4 cm diameter plas- tic column. A MnO 2 based material (Filox-R) was used as the oxidizing medium and was regen- erated by potassium permanganate solution. FILOX-R™ is the raw, unrefined ore used in the manufacture of FILOX ® filtration media. Chemically, FILOX-R (Raw) is naturally occurring ore. Ion exchange system: After passing through the oxidizing filter, water flows into a plastic col- umn filled with the anion-exchange resin Purolite A-300. The ion-exchange column has the same size as the oxidizing filter and the resin bed is 6 mL. Purolite A-300 is a Type II, strongly basic gel anion-exchange resin. 2.2 Regeneration procedure The oxidizing filter was regenerated with potassium permanganate (133 mg/L) for 30 min at a flow rate of 3 ml/min. This was followed by a slow rinse of 54 ml for 18 min at 3 ml/min and then a fast rinse of 90 ml for 10 min at 9 ml/min. This regeneration is not usually required unless the Filox performance is reduced. The ion-exchange filter (Fig. 1B) was regenerated with 100 ml of potassium chloride solution (2 M) for 33 min at 3 ml/min followed by a slow rinse at the same flow rate for 18 min. Finally a fast rinse was used for 10min at 9ml/min. A small amount of caustic (1%) is used in combination with salt during the regeneration in order to enhance the resin opera- tion. This addition gives higher operating capacity. 2.3 Water composition Simulated water served as the input water. It was made to resemble the real composition of Bangladeshi groundwater (Kinniburgh & Smedley 2001). The following chemicals were used: • H 3 AsO 3 (as specified in the results) • FeCl 3 и6H 2 O (0.15 ppm Fe 3ϩ ) 240 (A) (B) Figure 1. Schematic diagram of the (A) ion exchange setup and (B) regeneration stage. Copyright © 2005 Taylor & Francis Group plc, London, UK • Al(SO 4 ) 3 и14H 2 O (0.0014 ppm Al 3ϩ ) • MnSO 4 иH 2 O(0.001 ppm Mn 2ϩ ) • NH 4 NO 3 (0.001 ppm NH 4 ϩ ) • Na 2 SO 4 (0.26 ppm Na ϩ ) • NaF (0.22 ppm F Ϫ ) • Distilled water (pH 6.8). 2.4 Experimental procedure This experiment called for running three factors; namely, input arsenate concentration, pH and fil- tration velocity, each at two settings, on the ion-exchange process to determine which had the greatest effect on exchange capacity. A (full factorial) 2 3 design calls for 8 runs was used. Three more runs were set at the center point making 11 runs in total. The design matrix and analysis matrix were prepared before running the experiment; results were recorded in the analysis matrix as every run was completed. Analysis of the results was then done to obtain an equation that describes the dependence of arsenate-exchange capacity on the various factors. The system was stopped as the 10-ppb break through occurred. 2.5 Analytical procedures The chemical analysis of the samples was performed by a visual method Arsenic detection kit, Hach – 28228-00, (using the Hach method) for the detection of arsenic from 0–500 ppb. This test kit was used for the analysis of most samples. It was purchased from Anachemia Canada Inc. The kit included reaction vessels, chemical reagents (sulfamic acid and powdered zinc) and test strips. Representative samples were also sent to an accredited laboratory for ICP-MS detection. 3 RESULTS AND DISCUSSION 3.1 Preliminary test results with As(III) Initial tests were performed to determine if As(III) could be removed by the ion-exchange media without oxidation. Using an input concentration of 1600ppb As(III) into the ion exchange column, it was shown that the concentration of As(III) in the column effluent was approximately the same as the input, indicating negligible removal of this form of arsenic. Subsequently, the oxidizing media was evaluated for its ability to remove As(III). The results are shown in Figure 2. They indicate that the removal of the As(III) was complete until saturation of the column started to occur at a little before 1500 ml. After 2 L of media had passed through the column exhaustion had occurred. Therefore the use of the oxidizing media is a requirement for removal of As(III) as arsenic III needs to be converted to arsenic V prior to treatment. Ion exchange 241 0 500 1000 1500 2000 0 1000 2000 3000 4000 Volume (ml) Concentration (ppb) Figure 2. Sorption of arsenic (III) on oxidizing media. Copyright © 2005 Taylor & Francis Group plc, London, UK does not remove As(III) because As(III) occurs predominantly uncharged (H 3 AsO 3 ) in water with a pH value of less than 9.0 (Clifford 1999). The predominant species of As(V) are H 2 AsO 4 Ϫ and HAsO 4 2Ϫ which are negatively charged, and thus are removable by ion exchange. If As(III) is pres- ent, this it must be oxidized to As(V) before removal by ion exchange (Clifford 1986). 3.2 Sorption of As(V) Preliminary tests determined the levels and intervals of the factors (Table 1). The concentration range of arsenate 500–1600 ppb is typical of real groundwater in Bangladesh and therefore these levels were chosen for the range of the tests. The velocity range 0.65–1.94cm/min (giving an empty bed contact time (EBCT) range of 2–6 min) is also typical for most ion-exchange resins. Based on these parameters, 11 runs were conducted. The results are shown in Table 2, it can be seen that the arsenate capacity increases at lower pH values. At the lower pH the ionic form of arsenate H 2 AsO 4 Ϫ would dominate compared to the higher pH where HAsO 4 2Ϫ would dominate (Kang & Kawasaki 2000). The resins have been shown to be not sensitive to pH within the range of 6.5 to 9 (USEPA 2003). The difference is more pro- nounced at the lower arsenate concentration. Increasing the velocity at the lower concentration did not have any significant effects. However, at the higher concentration, the capacity decreased as the velocity increased. The contact time was not sufficient for the higher concentration at the higher velocity. To evaluate the reproducibility of the experiments, the values for the mid-range parameters were evaluated. The capacity of the ion exchange resin for the three repetitions were 0.030, 0.029 and 0.029 meq/g resin, indicating excellent reproducibility. Residual concentrations are shown in Figure 3a, and 3b for the various experiments. The lower arsenate concentrations were achieved at the lowest values for all parameters. In all cases, how- ever, the concentrations were below the current guidelines of 10ppb. However, if the guidelines 242 Table 1. Full factorial design for the sorption experiments, type 2 3 . Setting As(V) (ppb) pH Linear velocity (cm/min) Upper 1600 8.5 1.94 Middle 1050 7.5 1.30 Lower 500 6.5 0.65 Table 2. Determination of arsenate-exchange capacity as a function of initial arsenate concentration, pH and velocity. Operating parameters Initial As(V) Velocity Conc. (ppb) pH (cm/min) Capacity (meq/g) 500 6.5 0.65 0.030 1600 6.5 0.65 0.040 500 8.5 0.65 0.010 1600 8.5 0.65 0.030 500 6.5 1.94 0.030 1600 6.5 1.94 0.030 500 8.5 1.94 0.010 1600 8.5 1.94 0.025 Copyright © 2005 Taylor & Francis Group plc, London, UK were lowered to 10ppb, only lower concentrations of initial arsenate would be able to be treated to achieve acceptable concentrations of arsenate in the effluent. Figure 4 indicates the As(V) removal efficiencies. These rates are higher than the maximum 95% rate according to the USEPA for ion exchange as a Best Available Technology (BAT). 243 0 2 4 6 8 0 500 1000 1500 2000 Input As(V) concentration (ppb) Residual As(V) concentration (ppb) a pH=6.5 pH=7.5 pH=8.5 0 2 4 6 8 10 0 500 1000 1500 2000 Input As(V) concentration (ppb) Residual As(V) concentration (ppb) b pH=6.5 pH=7.5 pH=8.5 Figure 3. Dependence of effluent concentration on pH and initial As(V) concentration. (a) velocity ϭ 0.65 cm/ min; (b) velocity ϭ 1.94 cm/min. 98.0 98.4 98.8 99.2 99.6 100.0 6.5 8.5 pH As(V) Removal Efficiency (%) High As; Low Velocity Low As; Low Velocity High As; High Velocity Low As; High Velocity Figure 4. Effect of pH on As(V) removal efficiency. High As represents initial concentration of 1600 ppb, low As concentration represents 500ppb, high velocity represents 1.94 cm/min and low velocity represents 0.65 cm/min. Copyright © 2005 Taylor & Francis Group plc, London, UK 3.3 Statistical analysis of the data Statistical analysis of the data by elimination of insignificant coefficients by Student’s criterion and the adequacy test by Fisher’s criterion was performed and indicated that the pH has the most influence on exchange capacity. The following equation was thus derived from the analysis to relate the ion exchange capacity to the arsenic input concentration, pH and velocity: (1) where C as is the initial As(V) concentration and v is the velocity. The values generated by this esti- mation had a maximum deviation of 2.6% from the experimental one. Regarding the effluent As(V) concentration, the following equation was derived: (2) The maximum deviation between the experimental and estimated values was 7.5% in this case. 3.4 Regeneration of the ion exchange column The material balance of the sorbed (sorption stage) and desorbed (regeneration stage) arsenate was determined observed (Table 3). The ion-exchange column was regenerated and the spent regenerant analyzed to determine the amount of arsenate the resin sorbed in the sorption process as well as the amount of arsenate that was possible to have desorbed from the exhausted resin in the regeneration stage. This analysis determines the nature of the sorption process, i.e. it gives the answer whether the sorption process is indeed ion exchange and is reversible. An equal amount of arsenate sorbed and desorbed allows make the statement that the sorption process is reversible. In the sorption stage arsenate ions displaced chloride ions but in the regeneration stage chloride ions displaced almost all the arsenate ions and thus recovered the resin as seen in Table 3. 4 CONCLUSIONS This paper does not attempt to cover the complex of geological, economic and legal problems con- nected with the remediation of arsenic from groundwater but was limited to the optimization of the ion-exchange technology. There are still many problems that have yet to be solved, such as the effect of sulfate, nitrate and nitrite on sorption capacity and the minimization of the wastes pro- duced during the regeneration of the ion-exchange column. Therefore, attention can be drawn to the following preliminary conclusions of this research. • Feasibility of removing high levels of arsenic concentration at higher than 98% removal was obtained for all conditions evaluated • Sorption mechanism is anionic exchange for arsenate ions 244 Table 3. Sorption and desorption of arsenate from the ion exchange column for lowest concentration of arsenate, pH and velocity. As(V) removed As(V) in As(V) adsorbed As(V) desorbed by regeneration spent solutions Volume Bed volumes (g) (g) (%) (ppm) of wash (mL) 775 2810 2725 97 11.2 244 Copyright © 2005 Taylor & Francis Group plc, London, UK • Arsenite cannot be removed without oxidation to arsenate • Increasing the pH from 6.5 to 8.5 decreases sorption capability of the resin • Polynomial dependencies of arsenate-exchange capacity of different resins on various factors • Ion exchange column can be regenerated almost completely as 97% of the arsenic could be removed from the column. Further work will involve optimization of the regeneration processes to reduce waste generation. The same statistical analysis approach will be used. ACKNOWLEDGEMENTS The authors would like to thank NATEQ for the partial financial support for this project. REFERENCES Boyle, R.W. & Jonasson, I.R. 1973. The geochemistry of arsenic and its use as an indicator element in geo- chemical prospecting. J. Geochem. Explor. 2: 251–296. Clifford, D.A. 1986. Synthesis of Dual Mechanism Ion Exchange/Redox Resins and Ion Exchange/Coordinating Resins with Applications to Metal Ion Separations. Discussion of International Symposium on Metals Speciation and Recovery, Chicago, IL, August. Clifford, D.A. 1999. Arsenic Removal Options: Activated Alumina Adsorption, Anion Exchange, and Iron Coagulation-Microfiltration. AWWA Arsenic Technical Work Group Treatment Meeting, Resolve Inc. Washington, D. C., February 18–19. Kang, M. & Kawasaki, M. 2000. Effect of pH on the removal of arsenic and antimony using reverse osmosis membranes. Proceedings of the conference on membranes in drinking and industrial water production, volume 1, pages 489–494. ISBN 0-86689-060-2, October, L’Aquila, Italy: Desalination Publications. Kinniburgh, D.G. & Smedley, P.L. 2001. Groundwater studies of arsenic contamination in Bangladesh. BGS Technical Report WC/00/19. USEPA 2003. Arsenic Treatment Technology Evaluation Handbook for Small Systems. Office of Water (4606M), EPA 816-R-03-014, July. Wang, L, Chen, A. & Fields, K. 2000. Arsenic Removal from Drinking Water by Ion Exchange and Activated Alumina Plants. EPA/ 600/R-00/088, October. 245 Copyright © 2005 Taylor & Francis Group plc, London, UK . significant fac- tors in the transport of arsenic. Arsenic sulfides can be oxidized, releasing arsenic to the environment. 237 Natural Arsenic in Groundwater: Occurrence, Remediation and Management. or semi-arid inland basins, for example in parts of Argentina and the United States but the existence of arsenic in soluble form in the anaerobic groundwater of alluvial and deltaic plains was. Jonasson 1973). Arsenic is a contaminant that originates from and is transported to natural waters through erosion and dissolution of arsenic- containing rocks and soil. Climate and redox potential