1. Alumina and aluminium based adsorbents
1.1 Alumina: As(V) is known to be efficiently adsorbed by aluminium hydroxides, although As(III) is notably less adsorbed. In early experiments, Anderson et al. (1975) investigated the interaction between As(V) and amorphous aluminium hydroxide Al(OH)3 over a range of pH values from 5 to 9 and As(V) concentrations from 0 to 3.2 ppm. At low pH, arsenate adsorption was independent of pH, but as the pH reached higher values, adsorption began to decrease. Moreover, adsorption of arsenate increased with increase in initial arsenic concentration. Furthermore, the adsorption data can be described by Langmuir adsorption isotherm. The effect of pH and competing anions in adsorption process of As(V) onto gibbsite formula was studied by Manning and Goldberg (1996). They found out that the presence of phosphate can reduce As(V) adsorption severely. The effect of competing solutes in As(V) adsorption onto alumina was also studied by Jeong et al. 2007. The authors concluded that there was no effect from chloride and nitrate ions, but slight effects were observed by selenium (IV) and vanadium (V). Both phosphate and silica significantly reduces the adsorption of As(V) with alumina. Han et al. (2013) synthesized mesoporous alumina and calcined them at temperatures 400?, 600? and 800? respectively. They concluded that the kinetics of MA(400) in As(V) removal was faster and higher than CCA(400) (carbon coated alumina). The maximum As(V) removal using MA(400) was 94.7% , which is comparatively much higher than CCA(400) and other adsorbents. MA(400) shows high As(V) removal efficiency in the pH range of 2.5-7.0 and the well fitted adsorption data shows Langmuir isotherm and the maximum adsorption at near neutral pH (pH= 6.6±1) was 36.6mg/g. Due to the fact that MA(400) has highest BET surface area and more number of surface hydroxyl groups, it has the highest As(V) removal efficiency than MA(600) and MA(800). The effect of both organic and inorganic ligands on the adsorption of As(III) and As(V) onto Aluminium oxide was studied by Zhu et al. 2013. The authors concluded that the hindrance of the competing ligands was much higher for As(III) adsorption than for As(V). With the increase in pH the adsorption of As(III) and As(V) decreased in presence of competing ligands.
1.2 Activated Alumina: Activated Alumina (AA) is a physical/chemical process by which ions from feed water are adsorbed to oxidized AA surface (EPA 2000). The feed water is continuously undergoes through the bed for removing the contaminants. The oxidation state of arsenic plays an important role in its removal; As(V) can be adsorbed much more easily than As(III). Typical diameter of AA grains is in a range of 0.3 mm to 0.6 mm and they have a high surface area of approximately 370 m2/g for sorption (EPA 2003). Adsorption study of As(III) and As(V) using AA was performed at different pH values and salinities (Gupta and Chen 1978). They concluded that the maximum arsenic removal takes place in As(V) state. Equilibrium and kinetics of As(III) and As(V) adsorption onto AA was explained by Lin and Wu (2000). The authors observed that the adsorption of As(III) was much less than that of As(V) in most pH conditions. Later, Singh et al. 2001 reported that AA has the ability to remove more than 90 % of As(V), depending on pH conditions, initial arsenic concentration, dose of adsorbent. A suitable adsorbent for As(III) removal was found to be activated alumina (Singh and Pant 2004). The adsorption process had a large dependence on pH, adsorbent dose and contact time. Maximum As(III) removal was achieved 96.2% at a pH value of 7.6 and was independent of initial As(III) concentration. The experimental data of the adsorption process fitted into both Freundlich and Langmuir isotherms, in addition, the adsorption process followed a first order kinetics. Although with the increase in concentration of other ions, there is a decrease in efficiency of AA adsorption. The selectivity sequence of AA adsorption has been established in USEPA 2000:
OH? > H2AsO4 > Si(OH)3O? > F? > HSeO3? > TOC > SO42? > H3AsO3
1.3 Alumina plus manganese oxide: Maliyekkal et al. 2009 described the As(III) removal performance of MnO2-coated-alumina (MOCA). However, the sorption onto MOCA was pH dependent and high removal efficiency was observed between a pH range of 4 and 7.5. The predicted maximum As(III) sorption capacity was found to be 42.48 mg/g, which is much greater than that of AA (20.78 mg/g). As(III) removal mechanism by MOCA takes place in a two-step process, i.e., oxidation of As(III) to As(V) and retention of As(V) on MOCA surface in which As(V) forms an inner surface complex with MOCA.
1.4 Alumina plus iron oxide: Shugi et al. 2003 investigated the adsorption of As(III) by coating alumina with iron oxide, and found out that the percentage removal of As(III) decreased with increasing initial concentration. However, the equilibrium time of As(III) adsorption was independent of initial arsenic concentration. Hlavay and Polyak (2005) synthesized a unique adsorbent by in-situ precipitation of Fe(OH)3 on the surface of AA. The adsorption of As(III) and As(V) was represented by Langmuir-type isotherm. The authors also pointed to the advantages of the adsorbent, such as: (i) efficient and selective adsorption of As(III) and As(V), (ii) the adsorbent can be directly filled in columns, (iii) a low waste technology can also be developed by using this design. In another study, Masue et al. 2007 prepared an adsorbent by initial hydrolysis of mixed Al3+/Fe3+ salts. They found that approximately equal amount of As(V) adsorption maxima was observed for Al:Fe hydroxide ratios of 0:1 and 1:4 and As(III) adsorption maxima was at 0:1 Al:Fe hydroxide. As(III) and As(V) adsorption decreased with further increase in Al:Fe molar ratio.
1.5 Bauxite: Bhakat et al. (2006) studied the adsorption of As(V) on modified calcined bauxite (MCB). The adsorbent showed excellent As(V) removal (upto 100%) over a wide range of pH values (pH 2 to 8). However, the adsorption intensity increased from 0.099 to 1.37 mg/g in neutral pH range (pH~7) when the initial As(V) concentrations were 0.5-8 mg/l with a dose of 5g/l. Using Langmuir isotherm adsorption, the As(V) removal was best modelled delivering maximum adsorption capacity of 1.566 mg/g. Although the presence of sulfates and EDTA reduced the adsorption efficiency of As(V), there was no appreciable effect from background ions like Ca2+, Fe3+, Cl?, NO3?, PO4? and F?. But the sorption process was almost uninfluenced by the change in temperature.
1.6 Laterite: Maji et al. (2008) used laterite soil as an adsorbent to remove arsenic from groundwater. The optimum adsorbent dose was found to be 20g/l at a 30 min equilibrium time. The adsorption data points were evaluated by the Langmuir, Freundlich and D-R isotherm models. The adsorption rate coefficient (K) and adsorption capacity coefficient (N) obtained values as 1.21 l/(mg h) and 69.22 mg/l respectively. However, the authors have concluded that the results obtained using this method can be further extended to design pilot scale model for future research. They have also claimed that the adsorbent is cost effective and easy to separate from effluent water and an increase in pH or any increase in Fe(III)/Al(III) was not observed due to leaching. Sorption of As(III) and As(V) onto laterite iron concretions (LIC) was investigated by Partey et al. 2008. The sorption capacity increases with increase in temperature. The sorption isotherm data best fitted using Langmuir isotherm model. As(III) sorption increases with increase in pH solution up to pH 7 and then decreases with further increase in pH, however, As(V) sorption shows little change with increase in solution pH. As(V) sorption increases with the increase in ionic strength of solution. The authors have also claimed that LIC is cost effective, user friendly, requires no pre-treatment, and has the ability to remove both the arsenic species over a wide range of pH. Arsenic sorption mechanism onto LIC using electrophoretic mobility (EM) measurements were performed by Partey et al. 2009. They investigated that both As(III) and As(V) form inner-sphere complexes on LIC. Sorption of As(III) onto LIC was markedly reduced due to the increasing ionic strength. Although As(III) sorption was effected by the presence of sulfate ions, but the presence of phosphate reduced the sorption of both As(III) and As(V) on LIC.
1.7 Red Mud: Red mud is a byproduct of bauxite which is generated in the production of alumina. The cost of this adsorbent is lower than that of activated carbon and activated alumina (Altundogan et al. 2000 and 2002). Experiments by Altundogan et al. 2000 showed that the removal of As(III) was favored by the alkaline aqueous medium at pH 9.5 and As(V) removal was effective from a pH range of 1.1 to 3.2. The arsenic adsorption process obeyed Langmuir adsorption isotherm model. Thermodynamic calculations showed that As(III) adsorption process was exothermic, while As(V) adsorption was endothermic. Altundogan et al. 2002 indicated that the increase in adsorptive capacity of red mud can be reached by simply stirring red mud in a 1M HCl solution under atmospheric conditions. The process of arsenic adsorption using this method is pH dependent, where the optimum range of pH for As(III) sorption was 5.8-7.5 and 1.8-3.5 for As(V) sorption. The maximum removal for As(III) was 87.54% at pH 3.50 and for As(V) was 96.52% with a final pH of 7.25. The initial arsenic concentration in the aqueous solution was10 mg/l, with activated red mud dosage of 20g/l, for a contact time of 1 hour and at a temperature of 25 ?. The adsorption data followed a first-order kinetics with Langmuir adsorption isotherm. Later Altundogan et al. 2003 investigated arsenic removal by using liquid phase of red mud (LPRM). As(V) co-precipitates with aluminium hydroxide when the highly alkaline arsenic contaminated solution and LPRM are neutralized with acid solution. An effective removal of As(V) was obtained at a pH range of 5.5-8 when LPRM/(As(V) solution) ratio was 1/5 (v/v). When the initial arsenic concentration of the solution was 50 mg/l, more than 90% of As(V) was removed. However, for the effective removal of As(V),the amount of aluminium precipitation played an important role. Fuhrman et al. (2004a) investigated the increase in adsorption of As(V) through combined acid and heat treatment of seawater-neutralized red mud (Activated Bauxsol). The efficiency of As(V) adsorption by combined acid and heat treatment reached roughly up to 100% from 89%. These figures were reached when the initial As(V) concentration was 2 mg/l and the dosage of activated bauxsol (AB) was 5 g/l at pH 4.5 ± 0.1. Despite the presence of competing anions at the concentrations likely to be in natural waters, the removal efficiency was not disturbed by the usage of such amount of AB dose, but the efficiency was highly reduced when 0.5 g/l AB was used. Fuhrman et al. 2004b studied the adsorption of arsenic onto AB as a function of pH, particle size, contact time, initial arsenic concentration, AB dosage, and temperature. The authors concluded that the optimum pH for the removal for As(III) was 8.5 and for As(V) was 4.5. The removal efficiency of As(V) reached close to 100% independent of initial As(V) concentration, however with the initial concentration of As(III) there was a change in removal efficiency. The adsorption process followed a pseudo-first-order rate expression in that the process pseudo-equilibrates in 3 hours for As(V) and 6 hours for As(III) adsorption. The adsorption data fitted the Langmuir isotherm model. In order to get high arsenic removal efficiency using AB, the authors recommended that As(III) has to be oxidized to As(V). Hence AB can be a very efficient unconventional adsorbent for removing As(V) from water. Arsenic removing potentials of Bauxsol, Bauxsol modified by acidification (aBauxsol), and AB were compared as a function of sorbent dose, contact time, and initial arsenic content (Vukasinovic et al. 2012). AB was found to be most effective in As(III) and As(V) removal. The kinetic studies showed that AB sorbent was most effective in removing arsenic from water that reached to a sorption capacity of 1.49 mg/g at pH 7.0 with an initial arsenic concentration of 10 mg/l. The high removal efficiency of AB sorbent was dedicated to the presence of relatively high iron content. The experimental data was best fitted in Langmuir adsorption isotherm model. When the dosage of AB sorbent was 6 g/l, the effect of competing anions on arsenic adsorption was negligible, but when the dosage was reduced to 0.6 g/l, there was a great repression in arsenic adsorption. The order of interference of anions was found to be (on molar basis): phosphate>sulfate>bicarbonate.
1.8 Layered double hydroxides (LDHs): The adsorption of As(V) on conditioned LDH was investigated by Yang et al. 2006. The rate of As(V) adsorption on conditioned LDH increases with the decrease in particle size of the adsorbent, while the adsorption capacity of As(V) on conditioned LDH was independent of particle size. Adsorption data showed a well fitted Sips-type adsorption isotherm model. Results from Yang et al. 2005 shows that the adsorption capacities of As(V) on calcined LDH are higher than on uncalcined LDH. In fact, even in the presence of competing anions like carbonates and phosphates, LDH can be a promising adsorbent for removal of As(V). pH does not significantly reduce the adsorption of As(V) on calcined LDH. However, As(III) adsorption is much more difficult than As(V) both by calcined and uncalcined LDHs. As(V) sorption on Lithium/Aluminium layered double hydroxide intercalated by chloride (Li/Al LDH-Cl) was investigated by Liu et al. 2006. As(V) sorption maximum of Li/Al LDH-Cl was almost six-fold higher than gibbsite. The sorption mechanism on Li/Al LDH-Cl showed pH sensitivity from pH 4.0 to 7.0 but was insensitive above pH 7.0. EXAFS analysis showed that along with Al in the edges of Al(OH)3 layers, As(V) reacted with Li which are located in the vacant octahedral sites within Al(OH)3 layers. The high sorption capacity of Li/Al LDH-Cl than gibbsite was ascribed to the presence of intercalated Li cations which served as the permanent sorption sites for As(V). Wang et al. 2009 concluded in his study that the primary driving force for As(V) adsorption on LDHs is Coulombic attraction between the negatively-charged adsorbate (i.e., As(V)) and positively-charged sites of adsorbents. The competing effect of anions on As(V) adsorption by various LDHs generally are as follows: H2PO4?>HCO3?>SO42?>F?>Cl?.