Arsenic in Drinking Water and Its Removal

Arsenic in Drinking Water and Its RemovalLiu Zhenzhong12, Deng Huiping', Zhan Jian'1. School of Architecture, Nanchang University, Nanchang Jiangxi 330031, China;2. School of Environment Science and Engineering, Tongi University, Shanghai 200092, ChinaAbstract: Superfluous arsenic in drinking water can do harm tolen and Reimer, 1989).human health. In this paper, a broad overview of the availableThe high arsenic groundwater is mainly distributed intechnologies for arsenic removal has been presented on the basisBangladesh, India, American, German, Japan, Argentina,of literature survey. The main treatment methods included coagu-Mexico, Brazil, Poland and China (Zhao, 2002). Arseniclation- sedimentation, adsorption separation and ion exchange,was classified as a Group A carcinogen by the Unitedmembrane technique, which have both advantages and disadvanStates Environmental Protection Agency. There have beentages. It concluded that the selection of treatment process shouldextensive epidemiological studies showing that chronicbe site specific and prevailing conditions and no process can serveingestion of high levels of inorganic As caused skin cancerthe purpose under divese conditions as each tchnologoy has its(NRC, 1999), at the same time, some documentation of Asown limnitations. In order to gain good results, some methodsexposure also caused cancers of the nasal cavity, trachea,should be improved.bronchus, lung, liver, bladder, colon, kidney, prostate, brain,Key words: arsenic removal, coagulation-sedimentation, adsorp-the lymphatic and hematopoietic tissues as well as thetion, ion exchange, membrane techniquenervous system (Chen and Lin, 1994; Naqvi et al 1994;Remembrance, 2003). According to a recent report by the1 IntroductionUnited States National Academy of Science and UnitedStates National Research Council, even at 3 ug/L of As, theGroundwater in the part of the world has been affectedrisk of bladder and lung cancer is between four and sevenby the arsenic pollution. The arsenic contamination in wa-deaths per 10,000 people. At 10 μg/L, the risk increases toter body was not only caused by the physical geographybetween 12 and 23 deaths per 10.000 people (NRC, 2001).chemistry change, such as the soil and rocks containingIn addition, As can cause high blood pressure and diabetes.arsenic weathering, geology variance, the arsenic mineTriggered by the risk concerned, most of countries havedrenching, the dissolved groundrock, but also caused bychanged their drinking water standard. The US EPA an-human activities, such as mining, metallurgy, waste releasenounced its ruling in October 2001 to lower the maximumof china industy, leather process, chemistry process, dye-contaminant level (MCL) from 50 μg/L to 10 μg/L with aing industry, use of pesticide and insecticide (Chakravartycompliance date of January 22, 2006 (An et al, 2005; EPA,et al, 2002; Smedley and Kinniburgh, 2002).2000a). In China, the new standard of drinking water stan-Common arsenic species in the environment include ar-dard which was promulgated by National Standard Com-senate (As(V)), arsenite ((IW))， dimethylarsinic acid mittee and Sanitation Ministry lowered the arsenic level(DMA), and monomethylarsenic acid MMA) (Gu et al,from 50 ug/L to 10 μg[L after consulted with the developed2005). Arsenic (As(V) and AS([T)) in the inorganic forms countries standard. The new standard will be implementedwere more toxic than those in the organic formns. Most ofin July 2007. The improved drinking water standard posedarsenic was presented in the inorganic forms in the naturenew challenges to water treatment technique.water. When it was in the oxidizing environment, the mainform was arsenate(As(V); when it was in the reducing2 Arsenlc removal from waterenvironment, the main form was arsenite (As(II))Lien etal, 2005). As(V) includes HgAsO4, H2AsO4, HAsO2 ,Mainly, arsenic species were presented in natural watersAsO3, in which the negatively charged arsenate ac- as As (V), As(II), the As valent has a great impact on thecounted for more; while As(II) includes H;AsO3, HAsO3，removal efet and the incidence of endemic (Korte andin which the uncharged H3AsO, accounted for more (Cul- Quintts of the two arse-中国煤化工Corresponding awuthor. Uiu Vhenzhong 1 liuz79@ 126.com)TYHCNMHGCainese Jourmal of Populasio, Resources and Enioamet 200 VoL.SNo3 23nic species and its redox reaction were reported as followset al, 1994; Sancha, 2000; Scot, et al, 1995). This may be(Luis, 2005):due to some of the aluminum remaining soluble, and pass-Dissociation of As(V)ing through filtration steps, while iron salts completelyH2AsO4→HzAsO,+H* pK,=2.2(1)form particulate iron hydroxide. Meng et al (2000) studiedHAsO,→HAsO3 +H* pKz=7.0(2)the influence of the Si0,-, So2 , CO32 on the As removalwhen the ferric chloride was used. SO~ and CO3- haveHAsO-→AsO2 +H* pKz=11.5(3)litle impact on As at pH 4 -10. While SiO,- has decreasedDissociation of As(I)he As removal efficiency because of its high competitionHAsO3 H4AsO;- +H* pK, =9.2(4)on adsorption site and produced the soluble polymer withRedox reactionthe frric. In addition, Meng et al (2001) removed the arse-HzAsO4+2H* +2e HHAs0, +H2O Eh=+0.56V (5)nic in the well using the ferric chloride and found thatIt was not favorable for arsenite sedimentation becausePO3 and SiO4- were the main anion to decrease the re-of its high solubility. Figure 1 demonstated themoval efficiency.relationship between arsenic species and Eh-pH. ArseniteLiu et al (2005) investigated the effectiveness andwas more difficult to remove than arsenate because of itsmechanism of permanganate enhancing arsenite (As(I))less negatively charged on the surface. Therefore, it wasco-precipitation with ferric chloride. Permanganate significantly enhanced As(II) removal for ferric co-precipita-necessary to oxidize arsenite to arsenate.A oumber of treatment technologies have been reportedtion (FCP)process. With Fe(II) dosage increasing from 2to the date for arsenate and arsenite removal from waters.mg/L to 8 mg/L, As removal increased from 41.3% toMany technologies such as coagulation and sedimentation,75.4% for FCP process; for permanganate oxidation-ferricsorption and ion exchange, membrane separation wereco-precipitation (POFCP) process, however, corresponsiveAs removal increased from 61.2% to 99.3%. Yuan et alapplied for arsenic separation.(2006) adopted the frrate as coagulation and evaluated the1.2performance of ferrate for arsenic removal by experiment.The results showed that the eficiency of As removal can0.85 H,AsO,HAsO,be achieved by 98%. The Optimum pH was 5.5- -7.5. The .0.4oxidative and coagulation time was 10 min and 30 minrespectively. The salinity and hardness did not interfereEh 0H,AsO,AsO,with removal arsenic. This method was very easy and ef--0.4 tfective comparing with the ferric method and KMnO,Ferric method.-0.8Coagulation-sedimentation removing-1262一4古810114adapted to the large -scale water plant which need not in-ptcrease the other treatment facilities with less investmentFig. 1 The relationshlp of arsenic and Eh-pHand convenient operation. However, the disadvantage ofthe method was that the process must follow the clarified2.1 Coagulation- sedimentationtank to remove the produced colloid arsenic and high de-Coagulation-sedimentation was a common technique topendence on the pH. It was not good for the arsenite re-remove the contamination from water. This process caumoval efficiency to depend on the constitute in the water.capture soluble As, transforming it into insoluble reactionAt the same time, large quantity chemicals were suppliedproducts (Edwards, 1994). As may be converted to an in-and the volume of the sludge was huge, other matters weresoluble form by precipitation, co-precipitation and adsorp-also introduced into the water.tion onto ferric or aluminum hydroxides as the ferric and2.2 Membrane techniquealum salts were applied. In general, iron salts were moreeffective at removing arsenite than aluminum salts (Cheng,The ordinary membrane separation removing arsenic中国煤化工YHCNMHG24 Chinese Jounal of Ppulaion, Resoures and Eovircameat 2007 Vol,5 No3included nanofilration (NF) and reverse osmosis (RO). 2.3.1 Activated alumina (AA)The large diameter ions including so3-, cr and Na* canBecause AA was a common, commercially producedbe removed by NP which was a process of relatively lowadsorbent material for water treatment, there was morepressure. While all the ions can be removed by reverseliterature on the use of the AA for As removal relative toosmosis which was a process of higher pressure such asother adsorbents. Activated alumina can remove the arsenicseawater desatation. Compared with the coagulation-effectively in the drinking water at pH 5.5. The principlesedimentation, membrane separation had a higher effi-was that the soluble arsenic (AsO,- and AsO3h) in theciency to remove arsenic.water can be adsorbed on the suface of theA pumber of studies have been performed to examineAA[am-Al(OH)3] and occupied the aluminous octahedronthe removal of arsenic by NF membranes (Chang, 1994;crystal lttice sites (Xiao et al, 2001). This kind of adsorp-Urase et al, 1998). The results showed that NF processestion action can further increase the instability of the AAwere effective for the removal of arsenic. Removal, how-uncrystalloid and be favorable to adsorb more As ion toever, depended upon operating parameters, membrane decease the soluble As in the water.properties and the characteritics of the source water. Re-Compared with other metal oxides such as hydrogenverse osmosis was an effective arsenic removal technology iron oxide, the kinetic sorption of AA was slower than thatproven through bench and pilot scale studies according to aof hydrogen iron oxide. The former spent two days to reachreport prepared for the US EPA (0000 Various pilot stud- half equilibriumn bul the ltter just used a few hours to be-ies reported arsenic removal ranging between 40% 9% come equilibium (liffrd, 1990). The optimum pH towithout specifying the arsenic species removed. Ningadsorb As(II) and As(V) was respectively 5.5 and nearly(2002) reported AstII) reduction at 73%. Bench-scale neural, and the big adsorption capability was respectivelystudies with RO membranes showed As(V) reductions at15 mg/L and 3.5 mg/L (Jiang, 2001; Manning et al, 1998).88%- -96%.In addition, some intemupting ions such as S02, cr,Generally, the membrane process was highly efctivePO3- and F can decrease the arsenic removal. Thismethodcan be used in the concentrating supply waterfor arsenic removal. Membrane also provided an effectiveplants and in the family purify equipment. However, com-barrier to suspended solids, all inorganic pollutants, organicmicropollutants, pesticides and so on. In another word, itmon AA with low sorption capability adapted to thin pHremoved all the ions presented in the water, though somerange and acidic water, As(II) need be oxidated to As(V).minerals were essential for proper growth, remineralizationAfter a period of the operation, regeneration can lose10%- -50% sorption capability. At the same time, aluminumwas required after the treatment. In order to prevent mem-quantity would increase in the treated water because of thebrane contaminant, some pretreatment was necessary. Thelosing of the aluminum. It was known to all that the persis-water should be acidic and pH need be corrected. Thetent adsorbed aluminum caused the diseases such as aged-process was expensive compared to other methods.ness imbecile.2.3 Adsorption2.3.2 Metal (hydrogen) oxideSome materials with big specific surface area and highIron oxide with high surface energy and surface area hassurface energy which have strong adsorption ability canstrong adsorption capability with respect to many inorganicseparate and remove the contaminant to purify water in theions and organic matters. Lots of literatures have reportedprocess of adsorption. This adsorption action may bethat heavy metal ions and organic contaminate can be re-chemistry effect such as surface chemistry coordination or moved by iron (hydrogen) oxide. Several iron oxides re-complex or physical effect such as staticelectric atraction.moved arsenic effectively such as amorphous hydrous fer-Adsorption was one of the most efctive methods to re- ric oxide, crsalline hydrous frric oxide (errihyrie),move the arsenic in the water, the common adsorbents a FeOOH, hematite, magnetitie and goethite (Goldberg,included activated alumina, activated carbon, function resin2002; Jackson and Miller, 2000; Jain and Loeppert, 2000;and metal oxide, etc.Manning et al 1998; Raven et al, 1998). The iron hydroxide中国煤化工YHCNMHGChinee Joumal of Populain Resoures and Enionment 2000 VoL.5No.3 25and its polymer with high adsorption and fast kinetic whichrespectively 163- -184 BV and 149 -165 BV (Joshi andhave the biggest capability of adsorption were the mostChaudhuri, 1996). The iron oxide coated sands has loweffective. However, some shortcomings with iron hydrox-saturation capacity of arsenite and arsenate. When theide still existed, for example, some anions have an adversecoated iron oxide quantity was 0.1% 0.2% or 1%- -2%, theinfluence on the arsenic removal such as so4~, cr, F,saturation quantity was respectively 0.041 mg/L and 0.043PO广and SiO3. The ltter two have relatively strongermg/L (Thirunavukkarasu, 2003).impact than the former three. In addition, the arsenic re-The shortcoming of the process was that the granularmoval was infuenced by pH markedly. The removal ef-size must be large enough to reduce the stream resisanceciency decreased quickly as pH was over 8.5. Arsenatewhich decreased the sorption capacity of the unit volumeremoval was much better than arsenite by iron hydroxide. and quality. The larger volume sorption unit and frequentlyOn the other hand, the robustness and mechanical strengthregeneration operation was needed because of the low ad-of the granular iron hydroxide was not very good and sorption capacity. Because sands belonged to the silionneeded improvement, the headloss pressure was producedoxide, which combined with the iron oxide and hydroxidequickly with time and became more significant after back-less strongly, the iron oxide was subject to break off.washing, the larger the particles, the less adsorption capac-2.3.4 Activated carboniy. While the regeneration of granular iron hydroxideseemed feasible, it generates an alkaline solution with highActivated carbon, either granular or powered, waslevels of arsenate, which requires further treatment andwidely used as an adsorbent for water and advanceddisposal (Gu et al, 2005; Selvin et al, 2000).wastewater reatment. It was capable of adsorbing a wideAs far as iron hydroxide had poor adsorption to arsenite,of organic contaminants and heavy metals and was desig-Lakshmipathiraj et al (2006) combined the advantages ofnated as the best available technology by the US ( Gu et al,Mn and Fe, synthesize a suitable adsorbent, Mn-substiuted2005; James, 1985). Arsenic adsorption onto virgin acti-iron oxyhydroxide (MIOH), which could remove both ar-vated carbon was minimal and regeneration was difficult,senite and arsenate from aqueous solutions with consider-so it cannot be directly applied for arsenic treatment (Dausable eficiency. Mn-substituted iron oxyhydroxide (MIOH)et al, 2004; Deng et al, 2005). Literature has, however,efficacy was studied for the removal of arsenite and arse-shown that the adsorption on activated carbon can be sig-nate from aqueous solutions. The maximum uptake of ar-nificantly increased by treatment with various metal com-senite and arsenate was found to be 4.58 and 5.72 mg/gpounds (Huang and Vane, 1989; Reed et al, 2000). Somerespectively. Adsorption was best described by Langmuiriron compounds were impregnated into activated carbon,isotherm and activation energies as calculated from aresuling in enhanced As sorption (Gu et al, 2005; Huangmodel-free isoconversional method were found to be on theand Vane, 1989). Enhanced arsenic adsorption was simi-order of 15- 24 and 45- -67 kJ mol-' for arsenate and ar-larly observed with copper-treated and zirconium-treatedsenite, respectively.activated carbon (Birgit et al, 2004; Manju et al, 1998).2.3.3 Iron oxide coated sandsBirgit (2004) adopted Zr treated activated carbon to re-In light of the deficiency of granular iron bhydroxide,move arsenic in the wastewater, and the efficiency wassome iron oxide was coated on the sands to remove thehigh, but it was not suitable for drinking water because ofcontamination in the water. It was low cost operation forits toxic. Then Birgit coated Cu and Fe to the carbon, bothiron oxide coated sands beds and it adapted to the smallof arsenite and arsenate removal effect were improvedequipment and family usage. In the trial, the iron oxidemarkedly. Huang et al (1989) reported that activated car-coated sands volume was 50 mL, diameter was 11 mm,bon traited ferrous salts improved the arsenate removalvelocity was 1 cm/m, detention time was 50 min, the initialefficiency because of the producing ferrous arsenate com-concentration was 1 000 ug/L, according to WHO drinkingplex. Gu et al (2005) posed that frrous ion can impenetratesanitary standard, the arsenic goal level was 10 ug/L. The,into activated carbon inner boles and be oxided to ferric ionequipment operated for ten times, the penetrating volumeby oxidants. Then ferric ion can combine multifunctionof the iron oxide coated sands for arsenite and arsenate wasgroup to enhance the sorption capacity of arsenic (Fig. 2).中国煤化工HCNMHGPore surface ofGACFe'difusion into internal poresFe*oxidation by O2, H2O2,orNaClOFe-OH器feOHH.AsO2. HAsO7-diffusion into internal poresFe-O-AsO,Fe-O-Aso,feO-AsO,H,AsO_. HAsO2- complexationwith the GAC-Fe surfaceFig.2 An ilustratve model for preparation of As-GAC and arsenic adsorptionThe sorption capacity increased after activated carbonbind with a transition metal such as copper and iron, and (b)was traited. However, some shortcomings existed whenmetal ions that was immobilized to the functional groups ofmetal-impregnated activated carbon was used. The metal the hosting resin. While sharing many common featuresion was subjected to leak from the media in the process ofwith standard ion exchangers, a ligand exchanger employsbackwashing, then the second contamination of watertransition metal ions as its terminal functional groups. As asource was caused and the sorption capacity decreased. Inresult, ligand exchange involves concurrent Lewisorder to improve the combination ability with metal ions,acid-base (LAB) interactions (metal-ligand complexation)activated carbon was substituted with the ion exchangeand electrostatic interactions between the fixed metal ionsresin and fiber, to remove arsenic in water.and target anionic ligands. While conventional anion ex-changer selectivity for various anions was govermed by2.3.5 Ion exchange resin and fiberelectrostatic interactions, the afinity of a PLE was pre-Ion exchanged resin has many kinds according to the dominated by both the ligand strength and ionic charge ofmaterial, making method and purpose. As far as the func-the ligands (An et al, 2005).ion group was concerned, there were strong (weak) acidIn the 1980s, Chanda et al (1988) impregnated iron tocanion exchanged resin, strong (weak) base anion ex-weak base macroporus chelating resin to remove As (II)change resin, redox resin, heat regeneration resin and che-and As (V), the results showed that As (I) and As (V)lating resin (Shao, 1989). lon exchange technique was re- were prior to other anions reacting with the resin at a lowgarded as one of the best available technology for arseniclevel of As. When equilibrium level reached, As (1I)wasremoval (EPA, 2000a). Current commercial ion exchanger lower than As (V) for one order of magnitude. Chanda alsowas prone to be saturated because of its poor selective tostudied the sorption principle of iron impregnated chelatingarsenic and suffered from competing with other anions.resin and thought iron was not toxic metal, and the adsorp-Due to the lack of As-selectivity, current ion exchange retion activity was prior to other metal ions such as copperquired frequent regeneration and lots of regeneration brineand nickel ion. In order to effectively select anions, thewas used, which in turn resulted in a large quantity of hosting resin must have positive charge, which was anionvolumes wastewater containing arsenic (Clifford, 1999).exchanger. Such polymers have high affinity to metal ionsGenrally, a polymeric ligand exchanger (PLE) was and prevented to strip off from the hosting resin. An (2005)composed of (a) a cross-linked hosting resin that can firmlyprepared a polymeric ligand exchanger (PLE) by loading中国煤化工MYHCNMHGClinese Jounal of Population,n Resouces and Erirnmen 2007 Vol. sNo.3 27Cu2+ to a commercially available chelating ion exchange surface area than gel resin which can adsortb more ironresin. Results from batch and column experiments indi- inside. Strong base anion resin with positive charge re-cated that the PLE offered unusually high selectivity for moved more arsenic than strong acid canion. Compared thearsenate over other ubiquitous anions such as sulfate, bi- iron-loaded resin with granular frric hydroxide, Cumbercarbonate and chloride. Because of the enhanced arsenatefound the former removed more arsenic than the latter, butselectivity, the PLE was able to treat ten times more bed the sorption kenetic was slow. Diffusion inside the granularvolumes (BVs) of water than commonly used SBA resins.maybe velocity limited process.Lenoble et al (2004) loaded manganese dioxide on aBecause the adsorption velocity of ion exchange resinpolystyrene matrix anionic commercial resin in chloride was low, ion exchange fiber was used to remove arsenic.form, called R-MnO2, which was tested for As(V) retention Ion exchange fber, a kind of surface adsorption and sepa-and for As(II) simultaneous oxidation and removal. Equi-ration material in fiber type, was 20 300 μm in diameterlibrium was reached in 2 b and isotherms showed thatwhich composed of lots of single silk. Compared with tra-R-MnO2 maximal capacities towards As(II) and As(V) ditional ion exchange resin, ion exchange fber has highwere, respectively, 0.7 and 0.3 mmol/g. Various mecha-stable chemistry, short distance of transfer, fast adsorptionnisms involved in As(II) and As(V) as follows (Fig. 3).and disadsorption, comprehensive purifying, good washingand regeneration capacity, low energy consumption andsmall fluid resistance (Xu et al, 2005).(1Liu et al (2002) prepared a new type of ion exchangeMnO,fiber to remove arsenate from water. The batch sorption> As(V) and Mn2*experiments showed that fibrous sorbent had high sorption(2) Vcapacity and good kinetic property for arsenate ion. Thesorption kinetic data can be described by LagergrenMoO,3)pseudo-second order rate equation very well. GreenleafManganese arsenate precipitate(2006) loaded nanoparticle hydrated iron oxide (HFO) onFig. 3 Various mechanisms involved in As(II) retenton by R-ion exchange fiber to sorb arsenic. Compared with the cor-responding resin, the efficiency was higher, sorption andMnO2desorption velocity was more fast. Other anions have lessLuis (2005) loaded nanoparicle iron hydroxide on the infuence on it While the marked shorcoming of ion ex-polymer resin producing four polymers (HCDX-M, HCIX-G change fiber was its high cost.HAIX-M,HAIX-G) to remove arsenic. Lewis acid-typefunctional groups existed between iron hydroxide and hy-3 Conclusionbrid polymeric sorbents. Lewis acid-base LAB, electric-static action and complexation combined two matters t0-The literature survey has indicated that each of the dis-gether. Luis proposed the main reason and process to ad-cussed techniques can remove arsenic under specified con-sorb arsenic by iron hydrogen as fllowsditions. Different techniques have different advantages and=FeOH2→H* +=FeOH pK(5) disadvantages. Specific technique can be selected accord-=FeOH→H* +=FeO~ pK2(6)ing to specific conditions. The existing methods should beperfected to gain the best effect.(=FeOH$)(CT)+H2AsOz- B+L + .(7)References(= FeO2)(H2AsOz)+Cr(= FeOH2 )+ H2AsO2- LB→>(E FeOH)(HAsO2)+H* (81. An B R, Stciowinder T R, Zhao D Y, 2005. Selective地Luis found the removal effect of macroporous resin wasmoval of arsenate from drinking water using a polymericbetter than gel resin. Perhaps macroporous resin has moreligand exchanger. Water Research, 39: 4993- -5004中国煤化工YHCNMHG28 Chinese Journal of Population, Resores and Eovironnent 2007 Vol. 5No.32. Chakravarty s, Dureja V, Battacharyya G et al, 2002.15. Gu Z M, Fang J, Deng B L, 2005. Preparation and evalua-Removal of arsenic from groundwater using low cost fer-tion of GAC-BASED iron-containing adsorbents for arse-nuginous manganese ore. Water Res, 36(3): 625- -632nic removal. Eoviron Sci Technol, 39: 3833 -38433. Chanda M, Driscou K F, Rempel G L, 1988. Ligand ex-16. Huang C R, Vane L M, 1989. Enhancing As$+ removal by achange sorption of orsenate and orsenite anions by chelat-Fe+t. treated activated carbon. J Water Pollut Control Fed,ing resins in frric ion form 1: Iminodiacetic chelating61: 1596-1603resin Chelex 100. Reactive Polymers, 8; 85- 8917. Jackson B P and Miller W P, 2000. Effectiveness of pho-4. Chang S D, 1994. Critical Isues in Water and Wastewaterphate and hydroxide for desorption of arsenic and seleniumTreatment, ASCE, Boulder, Cospecies from iron oxides. Soil Sci Soc Am J, 64:5. Chen CJ and Lin L I, 1994. Human carcinogenicity and1616- -1622atherogenicity induced by chronic exposure to inorganic18. Jain A and Loeppert R H, 2000. Effect of competing anionsarsenic. In: Nriagu J O, ed. Arsenic in the Environment,on the adsorption of arsenate and arsenite by frihydirte. JPart I: Human Heath and Ecosystem Effects. New York:Environ Qual, 29: 1179 1184Jobn Wiley & Sons, Inc. 27: 109-13119. James M, 1985. Montgomery Consulting Engineers, Inc.6. ChengR C, Sun L, Wang H C, Beuhler M D, 1994. En-Water Treatment Principles and Design; Wiley&Sons:hanced coagulation for arsenic removal. Journal of Ameri-NewYorkcan Water Work Association, 79 -9020. Jiang J Q, 200. Removing arsenic from groundwater for7. Clifford D, 1999. lon exchange and inorganic adsorption inthe developing world-a review. Water Science and Tech-water quality and treatment. In: A Handbook of Commu-nology, 44(6): 89 98nity Water Supplies, American Water Works Association.21. Joshi A and Chaudhuri M, 1996. Removal of arsenic fromMcGraw- Hil, New York, NY9.1- 991ground water by iron oxide - coated sand. Jourmal of Envi-8. Cullen W R and Reimer K J, 1989. Arsenic speciation inronmental Engineering, 769-771the environment. Chem. Rev. 89:713-764Korte N E and Quintus F M s, 1991. A review of arsenic(I)9. Daus B, Wennrich R, Weiss H, 2004. Sorption materials forin groundwater. Critical Review in Environmental Control,arsenic removal from water: A comparative study. Water21:1-39Research, 38: 2948 -295423. Lakshmipathiraj P, Narasimhan B R V, PrabhakarS, et al,10. Deng B, Caviness M, Gu z, 2005. Arsenic removal by ac-2006. Adsorption studies of arsenic on Mn-substituted irontivated carbon-based materials. In: Advances in Arsenicoxyhydroxide. Joumal of Colloid and Interface Science,Research: Integration of Experimental and Observational304: 317-322Studies and Implications for Mitigation, ACS Symposium24. Lenoble V, Laclautre C, Serpaud B, et al (2004). As(V)他-Series; Washington, DC: American Chemical Societytention and As(I) simultancous oxidation and removal on11. Edwards M, 1994. Chemistry of arsenic removal duringa MnOr-loaded polystyrene resin. Science of the Total En-coagulation and Fe-Mn oxidation. Joumal of the Americanvironment, 326: 197 -207Water Works Association, 86(9): 64-7825. Lien H L and Wilkin R T, 2005. High-level arsenite re-12. EPA, 2000a. Arsenic removal from drinking water by ionmoval from groundwater by zero-valent iron. Chemosphere,exchange and activated alumina plants. Cincinnati, OH59: 377-386Office of Research and Development. EPA/600/R- 00/08826. Liu RP, Li X, Xia S J, 2005. Efectiveness and mechanism13. Goldberg s, 2002. Competitive adsorption of arsenate andof permanganate enhancing arsenite co-precipittion witharsenite on oxides and clay minerals. Soil Sci Soc AmJ, 66:ferric chloride. Environment Science, 26(1): 73 -76 (in413- 421Chinese with English abstract)14. Grenleaf J E, Lin J C, Sengupta A K, 2006. Two novel ap-27. Liu R X, Wang Y X, Tang H X, 2002. Removal of arsenateplications of ion exchange fibers arsenic removal andby a new type of ion exchange fiber. Environment Science,chemicalfree softening of hard water. Environmental Pro-23(6): 88-91gress, 25(4): 300-31128. Luis C, 2005._ Polymer-supported hydrated Fe oxide中国煤化工MYHCNMHGChinese Journal of Populaion, Resources and Envioumneat 2007 VoL.5No.3 2nanoparticles: characterization and environmental applica-41. Scot K N, GreenJ F Do H D, Mclean S J, 1995. Arsenictions Disertation]. Lehigh Universityremoval by coagulation. Joumal of American Water Work29. Manju G N, Raji C, Anirudhan T s, 1998. Evaluation ofAssociation, 114-126conuthusk carbon for the removal of arsenic from water.42. Selvin N, Messham G Simms J, et al, 2000. The develop-Water Res, 32(10): 3062- -3070ment of granular fermic media -arsenic removal and addi-30. Manning B A, Fendorf S E, Goldberg s, 1998. Surfacetional uses in water treatment. In: Proceedings - -Watertructures and stability of arsenic on gocthite: Spectro-Quality Technology Conference, Salt Lake City, UT,scopic Evidence for Ine-Surface Complexes. Environmen483- 494Sci Technol, 32: 2383 -238843. Shao L, 1989. lon exchange resin in water treatment. Bei-31. Meng X G Bang s, Korfiais G P, 2000. Effects of silicate,jing: Water Conservancy and Electric Power Press (in Chi-sulfate, and carbonate on arsenic removal by feric chloride.nese with English abstract)Water Research, 134(4): 1255 -126144. Smedley P L and Kinniburgh D G 2002. A review of the32. Meng X G Krfratis G P, Christodoulatos C, et al, 2001.source, behavior and distribution of arsenic in natural wa-Treatment of arsenic in Bangladesh well water using aers. Applied Geochemistry, 17(5): 517- -568bousehold co precipitation and filtration system. Water 45. Thirunavukkarasu 0 S, Viraraghavan T, Subramanian K s,Research, 35(12): 2805- 28102003. Arsenic removal from drinking water using iron 0x-33. National Research Council, 199 Arsenic in Drinking Waide-coated sand. Water, Air and Soil Plluion, 142: 95-111ter. Washington D C: National Academy PressUrase T, Jeongik Oh, Yamamoto K, 1998. Effect of pH on34. Naqvi S M, Chetana V, Singh H, 1994. Toxicity and me-rejection of different species of arsenic by nanofiltration.tabolism of arsenic in vertebrates. In: Arsenic in the Envi-Desalination, 117(20): 11-18ronment Part I: Human Heath and Ecosystem EffectUS-Environmental Protection Agency, 20006. Technolo-New York: John Wiley & Soas, Inc,27: 5591gies and costs for removal of arsenic from drinking water,35. Ning Y R, 2002. Arsenic removal by reverse osmosis. DerEPA-8 15-R-00-028, available from EPA websitesalination, 143: 237-24148. XiaoTE, Hong B, Yang Z H, 2001. Hydrogeochemistry of36. NRC, Arsenic in drinking water: 2001 Update. Nationalarsenic and its environmental effects. geological scienceResearch Council. Washington D C: National Academyand technology information, 20(1): 71-76 ( in Chinesewith English abstact)37. Raven K P, Jain A, Loeppert R H, 1998. Arsenite and arse-XuT J and Du M N, 2005. Research on application ofnate adsorption on femihydit: Kinetics, equilibrium, andion-exchange fber in water treatnent. Hi-Tech Fiber &adsorption envelopes. Environ Sci Technol, 32: 344 349Application, 30: 44 -4838. Reed B, Vaughan E R, Jiang L Q, 2000. ASII), As(V), Hg,50. YuanBL,LiK L, DengL L, et al, 206. Removal of Asand Pb removal by Fe-oxide impregoated activated carbon.(四by ferate oxidation-coagulation from drinking water.J Environ Eng Div (Am Soc Civ Eng), 126: 869 873Environmental Science, 27(2): 281-284 (in Chinese with39. Remembrance L N, 2003. Arsenic removal from drinkingEnglish abstract)Water [Dissertation]. University of Idaho51. Zhao Y P, 2002. Arsenic and fluoride removal from aque-40. Sancha A M, 2000. Removal of arsenic from drinking wa-ous solution by Fe (D)loading ligand exchange cottonter spplie: Chile experience. Water Supply, 18(1): 621-cellulose adsorben Disseration]. Tanjin: Nankai Univer-24sity (in Chinese with English abstract)中国煤化工MYHCNMHG30 Cainese Journal of Ppulaion, Resoures and Eniromeat 2000 VoL. 5 No.3