Yu-Chuan Lin1, L. T. Fan1, Shahram R. Shafie1, Keith L. Hohn1, Botond Bertok2, and Ferenc Friedler2. (1) Department of Chemical Engineering, Kansas State University, 1005 Durland Hall, Manhattan, KS 66506, (2) Department of Computer Science, University of Veszprém, Egyetem u. 10., Veszprém, H-8200, Hungary
Much effort has been spent to investigate catalytic generation of hydrogen from methanol partial oxidation (MPO) in light of its practical and theoretical importance. Nevertheless, relatively little has been done to gain in-depth understanding of the catalytic mechanisms or pathways of MPO. The exploration of pathways of any catalytic reaction can be greatly facilitated by exhaustively generating at the outset stoichiometrically feasible and independent catalytic mechanisms or pathways (IPi) from a set of plausible elementary reactions by means of our graph-theoretic method based on P-graphs (process graphs). The method is implemented by resorting to the combinatorial algorithms derived from the two sets of rigorously stated axioms on the basis of the mass-conservation law and stoichiometric principle. A dominant or ultimate pathway or pathways should be searched among this set of pathways through a variety of theoretical and/or experimental means. In the current work, a set of 13 plausible elementary reactions has been proposed for MPO from which 6 IPis have been generated. These IP's, in turn, have given rise to 31 acyclic combined pathways (ACPis). The results have been obtained with a PC (Pentium 4, CPU 3.06GHz and 1G RAM) within 1 second. Such inordinate efficiency of our graph-theoretic method can be attributed to the mathematically exact nature of the method, which is implemented by the algorithms crafted from the rigorously stated axioms.