Water Qual. Res. J. Canada, 2006 • Volume 41, No. 2, 198–209 Copyright © 2006, CAWQ 198 Hydrous Iron(III)-Tin(IV) Binary Mixed Oxide: Arsenic Adsorption Behaviour from Aqueous Solution Uday Chand Ghosh,* Durjoy Bandyopadhyay, Biswaranjan Manna and Manik Mandal Department of Chemistry, Presidency College, 86/1, College Street, Kolkata 700073, West Bengal, India Groundwater in the Bengal delta plain is primarily in an anoxic environment and As(III)/As(total) ratios reported are in the range of 0.60 to 0.90. Most of the studied adsorbents showed greater adsorption affinity for less toxic As(V) than for more toxic As(III). The synthesis and physicochemical characterizations of iron(III)-tin(IV) binary mixed oxide composite with sys- tematic arsenic adsorption behaviour are reported here. The parameters studied are the effect of pH, adsorption capacity with varying arsenic loads, contact time for kinetics and isotherm modelling by batch method. In the pH range of 3.0 to 9.0, the adsorption of As(III) is found to be nearly the same while that of As(V) decreases with increasing pH for the studied arsenic concentrations (5.0 to 20.0 mg L -1 ). The adsorption data were analyzed using various kinetic and isotherm model equations. The rate of adsorption of As(III) was faster than that of As(V) at equilibrium pH ~7.0 and temperature 30(± 2)ºC. The adsorption kinetic data for arsenic fitted best with the Ho’s pseudo-second order [R 2 ~0.99–1.00 for As(III) and ~0.97–0.98 for As(V)] and parabolic diffusion (R 2 ~0.98–0.99) equations. The present kinetic data fitted the power function and simple Elovich kinetic equations well (R 2 = ~0.97–1.00), except the data obtained for As(III) from the 20.0 mg L -1 solu- tion. The analyses of adsorption data for As(III) and As(V), respectively, at equilibrium pH 6.8(±0.1) and 7.3(±0.1) with three different two-parameter isotherm model equations show the fitting order: Langmuir > Freundlich >Temkin. The mono- layer adsorption capacity (θ0, mg g -1 ) values evaluated from the Langmuir isotherm for As(III) and As(V) are 43.86 and 27.55, respectively. Key words: arsenic, adsorption, hydrous iron(III)-tin(IV) mixed oxide, isotherm, kinetics, pH * Corresponding author; ucg@vsnl.net Introduction Naturally occurring high arsenic levels in groundwater of the Bengal delta plain pose health risks to the public in India and Bangladesh, because people in both coun- tries are dependent upon groundwater as the main drinking water source. Many other countries worldwide have also faced similar problems (Smedley and Knniburgh 2002). Among the various arsenic species, As(III) is reported to be the most toxic (Ferguson and Davis 1972; Cullen and Reimer 1989; Korte and Fer- nando 1991), because As(III) is soft in nature, and com- bines readily with the soft thiol (-SH) group of the cys- tein residue in protein. Underground aquifers at a depth of 30 to 40 m in the Bengal delta plain are in a strong reducing (anoxic) environment, and As(III) to As(total) ratios are 0.6 to 0.9 (Harvey et al. 2002). Thus, As(III) is the major species in Bengal delta plain groundwater, and direct As(III) extraction from water into the solid phase is thought to be important in light of present-day con- cepts of “green chemistry” and public health. Numerous methods have been reported for removing arsenic from water. Among them, adsorption onto a solid surface has been found to be the most convenient in the field for third world countries where people are unable to deal properly with toxic sludge. Adsorbents studied to date are amorphous aluminium oxide (Anderson et al. 1976; Gulledge and O’Conor 1973), activated alumina (Rosenblum and Clifford 1984; Singh et al. 2001), acti- vated carbon (Huang and Fu 1984), amorphous iron hydroxide (Pierce and Moore 1982), hydrous iron oxide (Hsia et al. 1994; Wilkie and Hering 1996), granular fer- ric hydroxide (Driehaus et al. 1998; Thirunavukkarasu et al. 2003a), ferrihydrite (Fuller et al. 1993; Raven et al. 1998; Jain and Loeppert 2000), goethite (Sun and Doner 1998), crystalline hydrous ferric oxide (Manna et al. 2003), red mud (Altundogan et al. 2002), hydrated zirco- nium oxide (Manna et al. 1999), crystalline hydrated tita- nium oxide (Manna et al. 2004), iron oxide-coated sand and ferrihydrite (Thirunavukkarasu et al. 2001, 2003b), Fe(III)-Si binary oxide (Zeng 2004) and iron oxide impregnated activated alumina (Kuriakose et al. 2004). Many of the adsorbents are shown to be more efficient in removing As(V) than As(III). Crystalline hydrous ferric oxide can adsorb both arsenic species with nearly equal efficiency (Manna et al. 2003) and hydrous stannic oxide showed higher efficiency for As(III) (Manna 2005) when contamination levels in aqueous solutions were low. The underground aquifers of the Bengal delta plain at a depth of 30 to 40 m contain 60 to 90% As(III) of total