Final Report Summary - SURFOPTIM (Complexity adapted theoretical studies on surface optimisation and reaction modelling)
The project SURFOPTIM is aimed at understanding and quantifying the implications brought by the complexity of the energy landscapes that dominate molecule-surface interactions. In order to study bond breaking and forming processes of adsorbed molecules, relevant in heterogeneous catalysis, it is vital to have an accurate description of the potential energy surfaces (PES) that describe the interactions between adsorbates and the substrate. First principles calculations based in the Density Functional Theory (DFT) have proved to be a powerful tool in this kind of studies, and a large fraction of the current theoretical modelling of chemical processes on surfaces is carried out by this means. However, the study of the energetics of bond breaking and forming processes of adsorbed molecules, even in the apparently simple case of a diatomic molecule, is not exempt of difficulties derived from the high dimensionality of the system, such as the existence of several metastable adsorption configurations and pathways for diffusion and dissociation of the adsorbed species. In addition to this complexity, which is reflected in the intricate topography of the PES, dynamical effects also play a major role. These effects, which are often neglected when modelling reactions on surfaces, may be however responsible for enhanced or reduced catalytic activity of a surface in apparent contradiction with the predictions made from the energies calculated by DFT. For example, a molecule that approaches the surface with a certain kinetic energy or that is vibrationally excited may be able to follow a pathway towards dissociation that departs from the one designated as the lowest barrier by the DFT. It is the objective of SURFOPTIM to investigate the interplay between the energy landscape complexity and the system dynamics in cases where the "static" parameters, such as adsorption configurations transition states, lie in the same energy scales as the dynamical ones, i.e. in the range of 0.01-10 eV. Last but not least, the role of the surface in the dissipation of energy, either via vibrations of substrate atoms or electron-hole pair (ehp) excitations, has been accounted for in some relevant cases.
Among others, the following two case studies have been used as benchmarks for our methodological developments and as reduced models of phenomena that, we believe, are transferable to other systems:
- H diffusion on Pd: many previous studies had already determined the energy landscape of this system, which is particularly relevant for its catalytic properties. In fact, it is known that C=C bonds are more easily hydrogenated on Pd-based catalysts when H is diluted in the outer metal layers. Therefore, methods that allow for manipulation of this subsurface distribution of hydrogen atoms are being seeked by the heterogeneous catalysis community. In this respect, we have put together a model that yields quantitative rates of diffusion under realistic experimental conditions importing ideas for ballistic electron transmission and electron promoted vibrations. This part of the project has stimulated further work in order to devise an efficient method for subsurface mass transfer manipulation via electron pumping with a Scanning Tunnelling Microscope.
In addition, we have studied the dissipation mechanisms during the dissociation of H2 molecules on Pd surfaces. We have found that the ehp mechanism is around five times faster than the vibrational one, and that dissipation takes place during the relaxation of the dissociation products rather than being associated to the bond breaking itself. This result will have implications in the future interpretation of the so-called non-adiabatic effects during surface reactions.
- N atoms and N2 molecules on Ag: we have investigated the impact that preadsorbed N atoms on the surface has on the PES for a beam of partially dissociated atoms impinging on the Ag surface. In particular, we find that the changes in the PES corrugation are dramatical and recombination events proceed by a "pick-up" or Eley-Rideal (ER) mechanism with high efficiency. Interestingly, it was believed so far that efficienct ER reactions were restricted to lighter atoms. Our calculations, which contribute to explain a series of existing molecular beam experiments, can be generalised to other cases where surface reactivity is likely to be altered by preadsorbed species. Our findings make clear that these kind of processes are not as marginal as it was believed. In order to account for all the partaking species, a too high dimensional configurational space would have been required. Our approach has consisted in constraining the simulations to the relevant degrees of freedom at each length-scale, in such a fashion that only 3D or 6D PES were needed to obtain a reasonably accurate description of the mechanism.
Among others, the following two case studies have been used as benchmarks for our methodological developments and as reduced models of phenomena that, we believe, are transferable to other systems:
- H diffusion on Pd: many previous studies had already determined the energy landscape of this system, which is particularly relevant for its catalytic properties. In fact, it is known that C=C bonds are more easily hydrogenated on Pd-based catalysts when H is diluted in the outer metal layers. Therefore, methods that allow for manipulation of this subsurface distribution of hydrogen atoms are being seeked by the heterogeneous catalysis community. In this respect, we have put together a model that yields quantitative rates of diffusion under realistic experimental conditions importing ideas for ballistic electron transmission and electron promoted vibrations. This part of the project has stimulated further work in order to devise an efficient method for subsurface mass transfer manipulation via electron pumping with a Scanning Tunnelling Microscope.
In addition, we have studied the dissipation mechanisms during the dissociation of H2 molecules on Pd surfaces. We have found that the ehp mechanism is around five times faster than the vibrational one, and that dissipation takes place during the relaxation of the dissociation products rather than being associated to the bond breaking itself. This result will have implications in the future interpretation of the so-called non-adiabatic effects during surface reactions.
- N atoms and N2 molecules on Ag: we have investigated the impact that preadsorbed N atoms on the surface has on the PES for a beam of partially dissociated atoms impinging on the Ag surface. In particular, we find that the changes in the PES corrugation are dramatical and recombination events proceed by a "pick-up" or Eley-Rideal (ER) mechanism with high efficiency. Interestingly, it was believed so far that efficienct ER reactions were restricted to lighter atoms. Our calculations, which contribute to explain a series of existing molecular beam experiments, can be generalised to other cases where surface reactivity is likely to be altered by preadsorbed species. Our findings make clear that these kind of processes are not as marginal as it was believed. In order to account for all the partaking species, a too high dimensional configurational space would have been required. Our approach has consisted in constraining the simulations to the relevant degrees of freedom at each length-scale, in such a fashion that only 3D or 6D PES were needed to obtain a reasonably accurate description of the mechanism.