Separation of chlorine gas with membranes has been studied for more than 10 years, and the current article is a review of the progress and pitfalls which have been documented over these years. Based on this documentation. a suitable high performance membrane is expected to be designed within reasonable time. For such a membrane to be of commercial interest the material must prove to have sufficient durability, permeability, selectivity and not be too expensive. The materials evaluated are: PDMS, Polychloroprene, Fluorel (R), Teflon (R), Matrimid (R), Nafion (R), Hyflon (R), carbon membranes (polyimide and cellulose precursors), tubular glasses (4 and 2 nm nominal pore size), surface-modified tubular glasses, glass hollow fibre and Hyflon with silica or glass as mixed matrix membrane. None of the membranes can, as per date, fulfil all the requirements for being commercially attractive. Closest are the PDMS used at low temperature (high performance, but low durability at higher temperature). Hyflon (R) (good durability, but low selectivity), C18 and Pf-C10 surface-modified glasses (2 nm) (medium selectivity and structural integrity). In an attempt to combine the problems with the Hyflon and the glasses, the according polymer glass mixed matrix membrane are under investigation. (C) 2008 Elsevier B.V. All rights reserved.