Two PhD studentships are available with an autumn 2019 start date.
Diffusion and reaction of atomic and molecular hydrogen at metal surfaces underpins a wide range of technological applications, including hydrogen dissociation in fuel cells, photoelectrochemical water splitting, hydrogen storage, and heterogeneous catalysis. The small mass of hydrogen means that quantum nuclear effects govern its chemical interaction with metal surfaces. In addition, electronic excitations in the metal can also affect the chemistry via so-called “electronic friction effects”. Recent experiments suggest that there is a rich interplay between nonadiabatic and quantum tunnelling effects, calling for improved theories to provide a mechanistic understanding of these findings.
The two projects will involve the development and application of new quantum dynamical simulation methods to study the interplay between quantum tunneling and electronic friction in hydrogen metal chemistry. The students will develop an approach that combines ab initio electronic structure calculations and the path-integral molecular dynamics framework for nuclear quantum effects. The two projects will be closely aligned. Project 1 will involve the analytical method development and its application on hydrogen-metal diffusion. Project 2 will involve computational software development and its application to reactive hydrogen chemistry at metal surfaces.
The students will employ state-of-the-art electronic structure theory, path-integral molecular dynamics methods and contribute to numerical and analytical method development. The project will further involve computations on national- and international-scale high-performance computing facilities, and enhanced data analysis and visualization.
Figure: Reaction rates of hydrogen-metal chemistry are affected by (A) electron-nuclear coupling due to low-lying electronic excitations in the metal (so-called electronic friction forces) and (B) atomic quantum tunnelling effects which lead to deviations from classical rate equation models.