CO Oxidation on Au/TiO2

A new mechanism published in Science explains the role of hydroxyls and weakly adsorbed water at the Au/TiO2 interface for CO oxidation. More »

Bimetallic PdCu catalyst for selective oxygenate coupling

This combined theoretical and experimental study shows that a catalyst containing Pd and Cu in a 3:1 ratio dramatically decreased the unselective decarbonylation side reaction, while preserving the high catalytic rates for the desired coupling product seen with Pd-based catalysts. Based on DFT results we propose that Pd enhances the reactivity of exposed Cu sites, while Pd surface atoms are passivated by CO. More »

Outreach Activities in the Computational Catalysis and Interface Chemistry Group

Our group regularly organizes and participates in a broad spectrum of outreach activities. In Spring 2014 we assembled a team to showcase Catalysis & Reaction Engineering research for a sustainable energy future at Earth Day Houston, we demonstrated shape selective catalysis to 6-8th graders, and introduced NSF STEP Forward Campers to High Performance Computing. More »

Novel 2D RuPt Core-Edge clusters with superior CO electro-oxidation activity

In collaboration with Prof. Brankovic we propose that the complex interplay between epitaxial strain, ligand and finite size effects leads to the formation of rippled RuPt monolayer clusters, which provide optimal conditions for a quasi-ideal bi-functional mechanism for CO oxidation, in which CO is adsorbed mainly on Pt, and Ru provides OH to the active Pt-Ru interface. More »

Methanol Dehydration over H-ZSM-5 with Heterogeneous Al Distribution

DFT calculations point to the dominant factors governing the mechanism of methanol dehydration over H-ZSM-5, a frequently used zeolite catalyst with a uneven distribution of active centers in its structure. Modeling results reveal that both reaction pathways can be active at typical reaction conditions depending on the local confinement. More »

Methane Activation

Computational screening for low temperature methane activation catalysts requires a detailed understanding of the transition state structure. This visualization of how electrons re-arrange themselves during the C-H bond breaking process was prepared by Hieu Doan, who won 2nd place in the 2016 Vizapalooza. More »

 

Heterogeneous catalysts improve the efficiency of chemical transformations and are used for the production of fuels, chemicals, and the abatement of harmful emissions. Modern catalysis exists for about 100 years, but catalyst design and development remains by and large a time-consuming experimental trial-and-error process. The discovery of density functional theory (DFT) and the availability of large computational resources have already started to have a significant impact on our understanding of catalysis, and more importantly, they become increasingly applicable for the in silico design of new catalytic materials. The most successful approach for the identification of novel catalysts is known as computational catalyst screening, which encompasses three main steps: (i) the identification of the dominant reaction mechanism and the key reaction intermediates; (ii) the determination of a small set of catalytic reactivity descriptors that can predict reactivity and selectivity trends; and (iii) the calculation of these reactivity descriptors on new catalysts.


In the Computational Catalysis and Interface Chemistry (CCIC) group at the University of Houston (UH) we apply computational catalysis techniques, primarily DFT, to study catalytic process that enable the more efficient use of natural resources. This includes the utilization of abundant natural gas and biomass as fuel and feedstock for the production of valuable chemicals. The use of different fuels in vehicles, e.g. natural gas engines, also requires novel catalytic converters for emissions aftertreatment. Our group is very active in studying such catalysts and we collaborate closely with experimental groups at UH and elsewhere.


The necessary computational resources for our research is provided by a 60 node/720-core computing cluster and is supplemented with generous allocations at the Research Computing Center (RCC) and the Center for Advanced Computing & Data Systems. In addition, we use resources provided by the national supercomputing centers XSEDENERSC, and Argonne National Lab.