CO-PrOx reactor design by model-based optimization

DIEGO G. OLIVA; JAVIER A. FRANCESCONI; MIGUEL C. MUSSATI; PIO A. AGUIRRE

JOURNAL OF POWER SOURCES

ELSEVIER SCIENCE BV

Año: 2008 vol. 182 p. 307 - 316

0378-7753

This work analyzes the CO-PrOx reactor design as a component of the CO clean-up system of the ethanol processor for H2 production applied to PEM fuel cells. The operating conditions of the processor require compact and lightweight pieces of equipment and efficient operation at different conditions. An eggshell catalyst type of Pt/Al2O3 is considered. One-dimensional heterogeneous catalytic reactor model accounting for interfacial gradients is used to optimize the PrOx reactor. Different reactor components are added gradually to illustrate how the system dimensions and configuration change after optimization.The optimization problem determines the optimal reactor length, reactor diameter, catalyst particle diameter, inlet reactants temperature and insulating material thickness that minimize the total system volume. On these model-based results, the final reactor design is mainly governed by the presence of hemispherical heads (distributor and collector). Different inlet CO compositions and power generation targets are analyzed. According to the inlet CO level, more than one catalytic stage is required to meet design goals and fulfill process constraints. The model-based reactor optimization of the pseudo-adiabatic operation allows obtaining both designs for reducing volumes and optimal operating conditions that allows conventional reactor technology. Afterwards, simulation runs based on a rigorous one-dimensional heterogeneous catalytic reactor model accounting for intra-particle gradients are performed using data obtained from model-based optimization results. The aim of these simulations is to verify feasibility of theoptimal design obtained from the proposed one-dimensional heterogeneous catalytic reactor model without intra-particle gradients, which is intended to approximate an egg-shell catalyst behavior. The present work reflects clearly the advantages of applying mathematical programming techniques to optimize both design and operation conditions of the PrOx reactor.