However, the same reactivity that makes methane an attractive fuel makes methane conversion a difficult process to control. Computational chemist Tom Cundari (University of Memphis) is studying one way to do this--the design of new catalytic agents for methanol production. Cundari and associates are employing a scalable computing approach, including use of the resources of the Cornell Theory Center (CTC), to simulate methane reactions and suggest catalysts that are likely candidates for experimental evaluation.
The candidates include a group of highly reactive compounds that have transition metal atoms (with atomic numbers ranging from 21 to 106) at their centers. Before the researchers could simulate reactions involving these compounds, they had to simplify the models of the particular atoms of interest. This required groundbreaking work on the transition metal atoms. They used the IBM RS/6000 Scalable POWERparallel Systems (SP) at CTC to derive highly accurate descriptions of the cores of these more complex atoms, electronic core potentials (ECP), from supercomputer calculations.
After these descriptions were calculated and experimentally verified, they were added to a database of ECPs that was used to replace all the electrons in a given atom except those in the outer layer. This method allows the calculations to focus on the layer, or shell, of electrons that plays the most important role in the chemistry. Cundari then used simulations of methane elimination (the reverse reaction of activation) to fine-tune the ECPs.
The next step was to use the model to examine methane activation, the real reaction of interest. This work, which can be conducted on a workstation thanks to the availability of the ECP database, has provided some interesting results. For example, computer simulations of methane activation show weak interactions between the catalyst and the methane molecules prior to the beginning of the reaction.
Understanding these precursor interactions is important because they set the stage for the crucial methane activation step. Catalysts that interact selectively with methane could help steer the reaction process more effectively toward methanol. Cundari's work is not only suggesting new directions for experiment, it is also expanding the horizons of computational chemistry by expanding the palette for simulation.
Tricky Transformation
Remote Resources
Controlled Conversion