Free metal and metal-oxide clusters: beyond the static, monostructure description

Luca Ghiringhelli
Fritz-Haber-Institut der Max-Planck-Gesellschaft

One of the challenges of present and near future conversion, storage, and transportation of energy is its catalytic transformation into fuels, e.g. in the form of hydrocarbon molecules. Catalysis is also of great importance for pollution control. The widespread description and modeling of heterogeneous catalysis got stuck in the concepts and methods of the last century. Most (nearly all) catalysis researchers focus on energies and energy barriers, however the dynamics and statistical mechanics are also important. A catalyst is usually not working efficiently from the very moment that the process starts, but a macroscopic “induction period” must rather elapse. In fact, the catalytically active phase may exist only in a narrow range of the external conditions. Not only the surface composition but also surface morphology can change in the course of a catalytic process. Nanostructures of various shapes, point defects, extended defects such as steps, dislocations, and stacking faults, can result from interaction of the surface with the reactive environment. This latter aspect implies that the operating catalyst will never be the pristine material that is initially introduced in the reactor. Furthermore, the difference between “real-life catalysis” and “UHV surface chemical reactions” reflects the so-called “materials” and “pressure gap”. By focusing on gas-phase metal and metal-oxide (sub)nano clusters in a reactive atmosphere, I show how the accurate (and validated) knowledge of the (ab initio) potential energy surface cluster + reactant systems can be bridged to simple thermodynamic considerations for modeling the prediction of (meta)stable structures and compositions of the systems at realistic environmental conditions (finite temperature and pressure of the reacting atmosphere). One of the most delicate assumptions of such model is that the configurational entropy is approximated by the harmonic vibrational entropy. This assumption is not always justified. In fact some (but not all) (sub)nano cluster systems exhibit even at moderate temperatures anharmonic behavior associated with the coexistence of a multiplicity of structures: fluxionality and liquid-like behavior. I will show how massively parallelized (ab initio) replica exchange molecular dynamics and efficient reweighting techniques offer a way to describe the relative population of such higly-anharmonic systems, seamlessly bridging the vibrational to the configurational-change time scales. I also show that the accurate description of the PES is always necessary: temperature and time average do not necessarily smear out the inaccuracies! Nonetheless, (good) force field are valuable (timesaving) starting points for an ab initio structural scanning.

Presentation (PDF File)

Back to Workshop II: Fuels from Sunlight