Molecular Level Assessment of Thermal Transport and Thermoelectricity in Materials: From Bulk Alloys to Nanostructures
The ability to manipulate material response to dynamical processes depends on the extent of understanding of transport properties and their variation with chemical and structural features in materials. In this perspective, current work focuses on the thermal and electronic transport behavior of technologically important bulk and nanomaterials. Strontium titanate is a potential thermoelectric material due to its large Seebeck coefficient. Here, first principles electronic band structure and Boltzmann transport calculations are employed in studying the thermoelectric properties of this material in doped and deformed states. The calculations verified that excessive carrier concentrations are needed for this material to be used in thermoelectric applications. Carbon- and boron nitride-based nanomaterials also offer new opportunities in many applications from thermoelectrics to fast heat removers. For these materials, molecular dynamics calculations are used to evaluate lattice thermal transport. To do this, first, an energy moment term is reformulated for periodic boundary conditions and tested to calculate thermal conductivity from Einstein relation in various systems. The influences of the structural details (size, dimensionality) and defects (vacancies, Stone-Wales defects, edge roughness, isotopic disorder) on the thermal conductivity of C and BN nanostructures are explored. It is observed that single vacancies scatter phonons stronger than other type of defects due to unsatisfied bonds in their structure. In pristine states, BN nanostructures have 4-6 times lower thermal conductivity compared to C counterparts. The reason of this observation is investigated on the basis of phonon group velocities, life times and heat capacities. The calculations show that both phonon group velocities and life times are smaller in BN systems. Quantum corrections are also discussed for these classical simulations. The chemical and structural diversity that could be attained by mixing hexagonal boron nitride and graphene provide further avenues for tuning thermal and electronic properties. In this work, the thermal conductivity of hybrid graphene/hexagonal-BN structures: stripe superlattices and BN (graphene) dots embedded in graphene (BN) are studied. The largest reduction in thermal conductivity is observed at 50% chemical mixture in dot superlattices. The dot radius appears to have little effect on the magnitude of reduction around large concentrations while smaller dots are more influential at dilute systems.