Modelling thermal conductivity in nanostructured materials
The first goal of the present project is to develop methods to simulate thermal conductivity on the atomic scale. Most successful thermoelectric materials consist of atoms from the lower right part of the periodic table where very few inter-atomic potentials have been developed. In the present project we will derive tight binding parameterizations from density functional theory. The potentials will allow the molecular dynamics simulations necessary for the calculation of thermal conductivity.
The second goal is to systematically isolate the role of defect induced strain fields, symmetry breaking, composition gradients and phase boundaries on the lattice thermal conductivity. This will provide important input for phase field and finite element models in the attempt to “design” favorable micro-structures for thermoelectric materials. The work will be focused on Si-Ge systems to elucidate, on an atomistic scale, the role of specific factors of the chemical bonding in these systems, such as the near-instability of the crystal structure.