Computational discovery of new catalytic materials holds great promise for energy and environmental technologies. However, the energy accuracy of current computational methods severely limits the power of this approach.  Thus, much effort is being devoted worldwide to improve accuracy in high-throughput computational methods like DFT with van der Waals corrections.  To guide this effort, accurate experimental energies for adsorbed species are required.  While adsorption energies are known for many molecularly adsorbed species, few values are known for adsorbed molecular fragments (except H, O and N) and molecularly adsorbed species which dissociate upon heating.  Here we review calorimetric measurements of the adsorption energies of molecular fragments and molecules which tend to dissociate on single crystal surfaces.  Entropies of adsorbates are also very important in modeling catalytic reaction rates.  We also review trends in the measured entropies of adsorbates that reveal some errors in estimating entropies using the harmonic approximation with DFT-derived frequencies.  Computational catalysis has had significant success in building microkinetic models for multi-step catalytic reactions which reveal important new pathways that are required to understand experimental activity measurements.  We also describe a method for analyzing such microkinetic models that quantifies the extent to which each elementary step and each adsorbed intermediate controls the net reaction rate.  This approach can be applied to find the descriptors needed in computational catalyst discovery.Work supported by NSF and DOE-OBES Chemical Sciences Division.

Experimental benchmarks for computational catalysis

246th National Meeting (2013)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.