Bulk CeO2 has been recently reported as a promising catalyst for the oxidation of HCl to Cl-2. In order to undertake the scale up of this system, various oxides (TiO2, Al2O3, and low- and high-surface area ZrO2) have been evaluated as carriers. Supported CeO2 catalysts (3-20 wt.% Ce) prepared by dry impregnation were isothermally tested at the lab scale. Their performance was ranked as: CeO2ZrO2 >> CeO2/Al2O3 >= CeO2/TiO2. Kinetic data revealed a lower activation energy and a similar activity dependence on the partial pressure of O-2 for CeO2/ZrO2 compared to bulk CeO2. Pilot-scale testing over 3-mm pellets of this catalyst evidenced outstanding stability for 700 h on stream. In-depth characterization of the carriers and fresh catalysts by N-2 sorption, Hg porosimetry, X-ray diffraction, temperature-programmed reduction with H-2, Raman spectroscopy, electron microscopy, and in situ prompt gamma activation analysis, enabled to rationalize the different catalytic behavior of the materials. ZrO2 stabilizes nanostructures of CeO2 and a Ce-Zr mixed oxide phase, which offer high dispersion and improved oxidation properties. The catalyst also shows reduced chlorine uptake, and overall stands as a better Deacon material compared to bulk CeO2 and other supported systems. CeO2 is present on Al2O3 predominantly as well-distributed nanoparticles. Still, alumina does not induce any electronic effect, thus the supported phase behaves similarly to bulk ceria. TiO2, likely due to the structural collapse and dramatic sintering suffered during calcination, leads to the formation of very large ceria particles. Based on our results, CeO2/ZrO2 appears as a robust and cost-effective alternative to the current RuO2-based catalysts for large-scale chlorine recovery. (C) 2012 Elsevier B.V. All rights reserved.