Abstract
A two-dimensional mathematical model was formulated and solved for a catalytic membrane reactor in a tube-and-shell configuration. The model was compared to experimental conversions for the reversible gas-phase catalytic dehydrogenation of ethylbenzene to styrene. No adjustable parameters were used; the ceramic membrane diffusion coefficients and the reaction kinetics for the iron oxide catalyst were determined from independent experiments. The model correctly predicted the 20% increase in conversion over conventional fixed bed operation, when a permselective alumina membrane was used as the fixed bed reactor tube. The porous tube wall allowed the selective removal of hydrogen from the catalyst bed. Changes in conversion that occurred when tube- and shell-side flow rates were changed were also correctly predicted. The model predicted a small increase in conversion (2%) when catalyst was deposited in the membrane wall, which was less than the experimentally observed increase (an average of 10%).