Experimental studies of intrinsic kinetics and diffusion during methane steam reforming
The intrinsic kinetics and diffusion behaviour of methane steam reforming have been investigated in this work. Measurements of effective diffusivities of the gases present in methane steam reforming have been carried out by using the steady-state technique over wide ranges of temperature and pressure in a modified Wicke-Kallenbach (W-K) type diffusion apparatus. The effects of diffusion limitation on the reactions were examined at atmospheric pressure in a pellet reactor; and the intrinsic kinetics of methane steam reforming have been studied on a commercial nickel/alumina catalyst (ICI 57-4) in an integral reactor. A simulation study has been carried out to determine the effects of hydrogen removal on the performance of a membrane reactor and the catalyst activity for methane steam reforming. For the measurements of effective diffusivities, the temperature and pressure dependencies of effective diffusivities of gases measured have been obtained, and the tortuosities of pellets used for the gases measured have been estimated by using the parallel path pore model. At ambient pressure, the temperature exponent values range from 1.0 to 1.25. This indicates that the diffusion occurs in the transition region. At pressures up to 1MPa, the diffusion lies mainly in the bulk diffusion region. The pressure exponent was generally less than 1.0 with values lying between 0.4 and 0.85 except for gas-water vapour pairs where it is close to 1.0. The tortuosities estimated for the pellets varied from 1.84 to 2.51 for different gases at ambient pressure, but decreased with increase in pressure. Using the pellet reactor that couples the diffusion and reaction for methane steam reforming, experimental results, which were obtained over catalyst pellets without holes, show that the diffusion rate of methane into the catalyst pellets almost totally controlled the reaction rate. Under such diffusion limitation, other conditions did not show any apparent effects on the reaction. For catalyst pellets that contained four holes, the diffusion limitation on reaction was considerably reduced. However, catalyst activity did not play an important role in affecting the reaction. It was found that the effects of diffusion limitation on the catalyst pellet with a higher activity are greater than on catalyst pellet with a lower activity. For the study of the intrinsic kinetics of methane steam reforming, the effects of temperature, pressure, and ratio of steam/methane on reactions have been investigated experimentally under condition of no diffusion limitation. The effects of total pressure on initial reaction rates indicated that the rate controlling steps of steam reforming are surface reactions between adsorbed species. The experimental results confirmed that both CO and CO2 are primary products of steam reforming. Six detailed reaction mechanisms were considered by combining different adsorption behaviour of methane and steam on the nickel catalyst. For methane steam reforming, accompanied by water gas shift on the catalyst used, intrinsic rate equations were derived by using the Langmuri-Hinshelwood-Hougen-Watson (LH-HW) approach and Freundlich's non-ideal adsorption concept. Applying the method of parameter estimation and model discrimination, the new model was determined and the parameters in this model were determined as statistically significant and thermodynamically consistent. The influence of hydrogen removed on the catalyst deactivated by hydrogen sulphide poisoning and carbon formation has been simulated for methane steam reforming in a tubular catalytic reactor with a hydrogen permeable wall. The effects of the main variables on H2S tolerance and the tendency to carbon formation on the catalyst have been investigated. The simulation demonstrated that the hydrogen removed by the membrane may cause more extensive catalyst deactivation with the H2S tolerance decreasing and the tendency to carbon formation increasing as the proportion of hydrogen removal increased. The simulation also showed that the benefit of using a membrane reactor may not be achieved for feedstocks with a high H 2S level when a high proportion of hydrogen is removed. A higher applied pressure and a more efficient desulphurisation technique need to be employed to compensate for the influence of significant hydrogen removal on the catalyst activity for methane steam reforming in a membrane reactor operated at low temperatures.