This study has investigated mixes of South African sands and zero valent iron (ZVI) for the design of low cost permeable reactive barriers (PRB) for Arsenic removal. After preliminary batch tests conducted to select the substrates with the highest Arsenic removal capacities and to identify the best operative conditions, column tests were performed to simulate environmental conditions that occur when a PRB is interested by a contaminated effluent. Column experiments were carried out in two columns of different sizes (small and large columns) using two mixes of Gotthart Maier Metallpulver GmbH ZVI (GmbH ZVI) and Berea red sand (BRS) (75-25% and 50-50% of column volume) and only GmbH ZVI as filling material. In small column tests, higher amounts of BRS reflected in 100% of Arsenic removal both at 8.4 ml min(-1) and 16.8 ml min(-1) flow rates. In large column tests, 100% of Arsenic removal was observed for all flow rates investigated. A first-order kinetic model was used to describe the Arsenic adsorption kinetics. (C) 2014 Elsevier B.V. All rights reserved.