Heterogeneous catalyst design using stochastic optimization algorithms

We describe the application of two stochastic optimization algorithms to heterogeneous catalyst design. In particular, we discuss the optimal design of a two-component catalyst for the diffusion limited A + B --> 0 and A + B2 --> 0 reactions in which each of the reactants are adsorbed specifically on one of the two distinct catalytic sites. The geometric arrangement of the catalytic sites that maximizes the catalyst activity is determined by the use of a genetic algorithm and a simulated annealing algorithm. In the case of the A + B --> 0 reaction, it is found that the catalyst surface with the optimal active site distribution, that of a checkerboard, is approximately 25% more active than a random site distribution. A similar increase in catalytic activity is obtained for the A + B2 --> 0 reaction. While both the genetic and simulated annealing algorithms obtain identical optimal solutions for a given reaction, the simulated annealing algorithm is shown to be more efficient.