Intraparticle diffusion and adsorption of arsenate onto granular ferric hydroxide (GFH)

Porous iron oxides are being evaluated and selected for arsenic removal in potable water systems. Granular ferric hydroxide, a typical porous iron adsorbent, is commercially available and frequently considered in evaluation of arsenic removal methods. GFH is a highly porous (micropore volume approximately 0.0394+/-0.0056 cm(3)g(-1), mesopore volume approximately 0.0995+/-0.0096 cm(3)g(-1)) adsorbent with a BET surface area of 235+/-8 m(2)g(-1). The purpose of this paper is to quantify arsenate adsorption kinetics on GFH and to determine if intraparticle diffusion is a rate-limiting step for arsenic removal in packed-bed treatment systems. Data from bottle-point isotherm and differential column batch reactor (DCBR) experiments were used to estimate Freundlich isotherm parameters (K and 1/n) as well as kinetic parameters describing mass transfer resistances due to film diffusion (k(f)) and intraparticle surface diffusion (D(s)). The pseudo-equilibrium (18 days of contact time) arsenate adsorption density at pH 7 was 8 microg As/mg dry GFH at a liquid phase arsenate concentration of 10 microg As/L. The homogeneous surface diffusion model (HSDM) was used to describe the DCBR data. A non-linear relationship (D(S)=3.0(-9) x R(p)(1.4)) was observed between D(s) and GFH particle radius (R(P)) with D(s) values ranging from 2.98 x 10(-12) cm(2)s(-1) for the smallest GFH mesh size (100 x 140) to 64 x 10(-11) cm(2)s(-1) for the largest GFH mesh size (10 x 30). The rate-limiting process of intraparticle surface diffusion for arsenate adsorption by porous iron oxides appears analogous to organic compound adsorption by activated carbon despite differences in adsorption mechanisms (inner-sphere complexes for As versus hydrophobic interactions for organic contaminants). The findings are discussed in the context of intraparticle surface diffusion affecting packed-bed treatment system design and application of rapid small-scale column tests (RSSCTs) to simulate the performance of pilot- or full-scale systems at the bench-scale.