The present
work describes models for predicting concentration profiles
of various species in each of the reactors present in a fuel
processing system including a steam reformer, water gas-shift
reactor and a preferential oxidation reactor. These reactor
models incorporate phenomenological reaction schemes in power
law format in order to predict the conversion of the species
as a function of concentration and temperature. A surface film
approach is used rather than the more traditional two-dimensional
boundary layer in order to model the gas on the surface of the
catalyst. The modeling framework is built within the Matlab
Simulink environment to take advantage of available numerical
schemes and optimization algorithms. Only steady state operation
is considered for the reactors with validation occurring against
experimental data obtained from the literature. In addition,
temperature gradients within the reactors are imposed in order
to eliminate the need to model the energy equation of motion.
Parametric studies are performed on each of the individual reactors
by varying the length, catalyst loading, catalyst dispersion
and the effect of temperature drop across the reactor.