Coal gasification is currently being developed as a cleaner alternative to conventional combustion technology. To optimise H2 production in this process, a water gas shift reaction is utilised to convert all CO with H2O to produce CO2 and H2. Typically industrial processes involve a two-step reaction system followed by a downstream H2 purification system, though attracting significant inefficiencies and high capital costs. Replacing a conventional unit process with a membrane reactor in this application is foreseen to provide major advantages: • Removing H2 from the reaction in-situ, a membrane reactor can minimise downstream processing and associated capital and operational costs. • Shift the reaction to higher conversions, improving efficiencies and reducing CO in the outlet. • Provide a purified H2 stream for use in PEM fuel cells, while concentrating the CO2 stream at high pressure for possible sequestration. If the concept of membrane reactor is to be adopted in coal gasification, important material improvements and operational challenges must be overcome before commercialisation can be realised. In addition, the water gas shift reaction has only recently gained interest for membrane reactors and is currently lacking comprehensive research on the effects of operating conditions on both the conversion and separation found within the unit. To this end, these are strong motivations of this work to contribute with knowledge in this field of research. This thesis examines the effects of operating conditions such as temperature, pressure, space velocity, sweep gas rate and feed water ratio on the performance of a water gas shift membrane reactor as compared with a conventional reactor. Novel cobalt silica molecular sieve membranes were used with conventional low temperature water gas shift reaction CuZnAl2O3 catalysts. Two type of membrane reactor configuration were investigated: a small flat template with catalyst on the feed side, and a scale up tube membrane with catalyst placed also in feed stream, the inner shell of the tube membrane. The cobalt silica membranes complied with activated transport, following a flux dependency gas permeation, where He and H2 permeance increased with temperature whilst N2, CO and CO2 showed the opposite effect. Best single gas selectivities were very high, with values of 4500 (He/N2) and 1100 (H2/CO2). In addition, the energy of activation for He and H2 was also very high, in excess of 9-10 kJ.mol-1, clearly indicating the high quality of the membranes employed in this study. It was found that the MR improved CO conversions for a range of space velocities as a function of temperature, which was attributed to both activate transport property of the membrane and increased conversion. Below equilibrium limits this provided an improved H2 production of 5 – 12% at 200-250oC as the removal of H2 through the membrane allowed enhanced conversion. With a set feed rate, the optimum advantage of the MR was seen at a water ratio of 1 as the lower equilibrium limits allowed greater potential for conversion enhancement. With increasing excess water this advantage decreased from 7% down to 0.5% at 300oC. The use of pressure and sweep rate was used to optimise the membranes permeation rate and selectivity. While pressure (or driving force) provided the highest potential for increasing permeation (or flow rate), temperature in tandem with pressure provided the greatest improvement in membrane selectivity, thus increasing H2 concentration from 95 – 99% in the permeate stream. Detailed study of permeate concentrations with changing conditions was undertaken to provide an understanding of the transport properties of silica membranes. It was observed that membrane selectivity and permeation decreased with the gas composition (ie Single>Binary>Ternary). Nevertheless, for separation of a ternary mixture at increased temperatures (250oC) the membrane could provide up to 99% purified H2 while reducing CO down to 700ppm. Competitive gas permeation regimes are an industrial reality which is seldom addressed in membranes for high temperature gas separation. The effect of gas mixtures on permeation and selectivity was attributed to several factors: chemical potential (or driving force) of the feed gas mixture, blockage of micropores by large molecules (CO2 and CO) which in turn affects the percolation of H2. As a result, gas separation was reduced for higher CO and CO2 feed concentrations, leading to a significant reduction in the H2 flow rate. Temperature played a vital role in this competitive process, as H2 diffusivity and CO, CO2 adsorption followed an inverse trend. Thus, increasing temperature led to higher H2 pore diffusivity, while decreasing the competitive effect of CO and CO2 adsorption. The use of cobalt modified silica to improve the hydrothermal stability of the membranes was investigated for use in the water gas shift reaction. It was found that the addition of cobalt stabilised the silica pore network, maintaining microporosity after exposure to steam. This is validated with long term stability testing in a water gas shift membrane reactor, where it was seen that the membrane could provide up to 95% H2 concentration in the permeate for over 200hrs of MR operation. This provided novel work, establishing the feasibility of these membranes for long term testing and operation in an industrial WGS MR.