Inorganic membrane reactor: Case study : methane steam reforming
Methods of preparation of an inorganic composite membrane and the performance of \ud the membranes obtained have been investigated in this work. The membranes \ud obtained were microporous silica coated y/alumina and dense palladium coated oc- \ud alumina membranes. The preparations of the membranes have been carried out by the \ud dip coating technique and electroless plating method respectively. Characterization of \ud the membranes obtained with single gas permeation measurements at high \ud temperature and various pressures were accomplished. The performance of both \ud membranes has been examined for methane steam reforming on a commercial \ud nickel/alumina catalyst (ICI 57-4). A simulation study on methane steam reforming \ud for both conventional and membrane reactor has been completed to examine the \ud performance of these membrane reactors.\ud For the dip coating method, a microporous silica/y-alumina membrane was obtained \ud and single gas permeations for hydrogen and nitrogen at high temperature have been \ud measured. At 700 K the H2/N2 selectivity was 4.63. From the correlation between \ud temperature and the permeation of nitrogen in the silica membrane it could be \ud concluded that the permeation occurs in the transition region, which is controlled by \ud both the Knudsen mechanism and some viscous flow. At a range of temperature \ud between 623- 743 K the pore diameter calculated by hydrogen permeation is 0.57 nm. \ud The effects of temperature on the permeation of the membrane and hydrothermal \ud stability have been investigated. A hydrothermal exposure at around 723 K for more \ud than 720 hours during methane steam reforming does not make any membrane \ud transformation that leads to pore changes. A dense palladium coated a-alumina membrane was prepared by the electroless \ud plating method. The single gas permeation for hydrogen and nitrogen at high\ud 1 O\ud temperature has been measured. At 700 K the HI and N2 flux were 7.96 cm /cm .min \ud and 0.05 cm3/cm2.min respectively and the selectivity of H2/N2 was 175.03. The \ud pressure dependence of the tfe flux for the Pd membrane was found to be to the power \ud of 0.5, which is the theoretical value. The Fb diffusion through the metal membrane is \ud activated by temperature and can be described by an Arrhenius equation. The \ud activation energy (Ea) is 17.94 kJ/mol for the temperature range of 673-823 K.\ud Both membranes were studied for methane steam reforming by employing them as \ud membrane reactors. The steam reforming focused on the improvement effects of the \ud membrane on the conversion of methane. The effects of pressure, methane feed flow \ud rate and ratio of steam/methane, as well as sweep gas on reactions have been \ud investigated experimentally under conditions of no diffusion limitation.\ud The influence of the membrane with a hydrogen permeable wall on the conversion of \ud methane has also been simulated for methane steam reforming in a tubular catalytic \ud reactor. The effects of the main variables as applied in the experiments of the methane \ud steam reforming have been investigated in the simulation study either. The results of \ud the simulation have been compared to the results of the experimental works.\ud The general behavior is similar for both silica and palladium membrane reactors, i.e.: \ud by selectively removing one of the products from the reaction mixture, the methane \ud conversion can be improved to values higher than the thermodynamic equilibrium composition. From these results, it was concluded that in line with the permeation \ud studies, the performance of the palladium composite membrane was far superior to \ud that of the silica membrane for the steam reforming of methane.