Team proves streched surfaces make better catalysts

Manos Mavrikakis (left) and Jeff Greeley (18K JPG)

NO molecules adsorbed on an Ru surface dissociate preferentially at edge dislocations. This long-discussed effect was demonstrated with the aid of scanning tunneling microscope images that show a higher concentration of nitrogen molecules near such defects. (31K JPG)

Writing the "frontispiece" of the June issue of Angewandte Chemie International Edition, Assistant Professor Manos Mavrikakis and graduate student Jeff Greeley, in collaboration with colleagues in Germany, provide the first atomic-scale evidence for enhanced catalytic reactivity at stretched surfaces. The team offers microscopic evidence showing that molecules tend to adsorb and break on metal surfaces in locations where individual metal atoms are further apart than others, and they demonstrate that this behavior agrees with results from theoretical calculations.

"We also proved with our studies that the reaction rate for this specific bond-breaking event can be several orders of magnitude higher at the defect compared to the equilibrium lattice constant surface," says Manos. "That is extremely important because it means that one defect site can accomplish more work than a million non-defective sites in the same amount of time. The turnover rate, or rate at which molecules adsorb and dissociate, at the defect site overwhelms all contributions to reactivity from the large area of the catalyst that does not contain defects. As a result, turnover rates achieved at a few defect sites on catalytic particles can be responsible for the entire reactivity of metallic nanoparticles, such as those typically used in industrially important catalytic processes."

Bi-metallic catalysis is a bread-and-butter technology for many industries including petrochemical, pharmaceutical, polymer and energy industries, as well as for pollution-prevention. For years, it had been theorized that molecules on the surface preferentially adsorb at the stretched part of the defect and preferentially dissociate in the area of the defect. By comparing lattice constants via atomic-resolution scanning tunneling microscopy in connection with state-of-the-art quantum mechanical analysis, the research team confirmed these theories and put a solid foundation under techniques to generate better catalysts through strain.

"If you can be sure to provide defective areas, you can reduce the amount of catalyst that you need for processes where bond breaking is the rate limiting step. That is directly related to the particle size in catalysis," says Manos. "The smaller the particles you have, the more defects you create, the more steps you have compared to terraces on the surface of the catalytic particle. Terraces are the most benign, so you go to smaller and smaller particles to generate more and more steps and therefore increase the reactivity."

The work also holds significance for understanding the growth and electronic structure of layered metal nanostructures, which could play an important role in a number of technological applications beyond catalysis. Similarly, surface strain plays a critical role for the surface chemistry and properties of other solid materials, including semiconductors.