Topological analysis of catalytic reaction networks: Water gas shift reaction on Cu(1 1 1)

This paper describes our new reaction route network approach for elucidating reaction mechanisms and kinetics of complex catalytic reactions, which is applied to water gas shift reaction on Cu(1 1 1) comprising 19 elementary reaction steps along with detailed kinetics. Three dominant pathways are found, eventually, yielding an 11-step simplified mechanism, from which a predictive rate expression is derived. This provides good agreement not only with the complete WGS mechanism, but also with our experimental data on Cu. Microkinetic mechanism vs. experimental data for Cu under the following experimental conditions: catalyst loading of 0.14 g/cm 3; total feed flow rate of 236 cm 3 (STP) min −1; pressure of 1.5 atm; residence time τ = 1.8 s; feed composition of H 2O(10%), CO(10%) and N 2(balance). ▪ New catalysts are being sought for the water gas shift (WGS) reaction in connection with distributed hydrogen generation. This quest would be aided by improved fundamental understanding of WGS catalysis. We, thus, present here a systematic theoretical approach, namely, Reaction Route (RR) Graph analysis, for developing a comprehensive understanding of the WGS reaction based on a detailed molecular mechanism with a priori kinetics. The RR Graphs follow flow network laws of Kirchhoff, so that they allow not only graphical depiction of the myriad pathways, but also a detailed flux analysis of the network. Thus, we utilize a 19-step mechanism and assemble it into a RR Graph, which is analogous to an equivalent electrical circuit. These steps were picked to be comprehensive without causing undue complexities. A simplification and pruning of the mechanism is then performed based on Kirchhoff's laws of current (rate) and potential (affinity). As a result, the dominant pathways as well as the rate-limiting steps become transparent. Three dominant pathways are found, eventually, yielding an 11-step simplified mechanism, from which a predictive rate expression is derived. This provides good agreement not only with the complete WGS mechanism, but also with our own experimental data on Cu. Prediction of WGS activity on other metals is also provided.