Abstract
Granular ferric hydroxide (GFH) is often used for fixed bed adsorbent (FBA) columns in groundwater purification units around the world to remove arsenate contaminations. Groundwater can contain also other toxic (e.g., antimonite and vanadate) and non-toxic oxo-anions (phosphate and silicic acid) that are known to affect FBA lifetimes. Therefore, understanding the breakthrough of toxic compounds intended for removal by FBA is essential to their design, and is important to predict accurately breakthrough curves (BTCs) for FBAs in waterworks to plan future operating costs. Rapid small-scale column tests (RSCCT) and pilot-scale FBA were used to simulate vanadate BTCs for complex groundwater chemistries. The BTCs were simulated successfully using a homogeneous surface diffusion model (HSDM) combining equilibrium chemical adsorption and kinetic mass transfer. Adsorption parameters for various groundwater compositions were predicted using the CD-MUSIC surface complexation model, which was set up for the first time for akaganéite-based granular ferric hydroxide with a competitive multi-solute system. The results indicated that V(V) is less prone to competitive adsorption effects, and use of the homogeneous surface diffusion model to predict the BTCs requires then the kinetic mass transfer Biot number to be used as the only fitting parameter. On the other hand, a concentration overshoot could be observed for the two weaker absorbed oxo-anions arsenate and phosphate because of displacement by the vanadate. Results of pilot scale test column BTCs of vanadate for three waterworks with different groundwater compositions could be favorably extrapolated with a unique Freundlich constant kF of 3.2 derived on basis of the multi-solute CD-MUSIC model, and a unique Biot number of 37 fixed for all three different test sites.