Lattice Dynamics and Thermal Transport Properties of Nanophononic Materials

The phenomenon of thermoelectric energy conversion holds great promise in harvesting wasted heat and improving thermal energy management. This technology, however, is not widely used due to its generally poor efficiency stemming from intrinsic material-level limitations. This dissertation proposes to utilize the concepts of phononic crystals and metamaterials at the nanoscale in order to manipulate phonon lattice vibrations in a manner that qualitatively alters the thermal transport mechanisms and improves the thermoelectric energy conversion figure-of-merit. Phononic crystals utilize Bragg scattering while metamaterials use subwavelength properties to manipulate wave propagation in an elastic medium. With the advent of the nanotechnology revolution, the ability to fabricate material systems with nanostructured geometric features renders the concepts promised practically feasible. First, a Lagrangian formulation is derived to obtain the phonon dispersion spectrum of nanophononic crystals (NPCs) based on a simple three-dimensional mass-spring model. The formulation is then used to examine the opening of frequency band gaps due to the introduction of point-mass lattice defects. Next, models of silicon utilizing the Tersoff inter-atomic potential are then developed with a focus on investigating the effects of incorporating the full dispersion characteristics of 3D NPCs. The role that dispersion plays in shaping the nonlinear scattering properties as well as the thermal conductivity of the nanostructured material as a whole is thoroughly investigated. The results show that for relatively small voids and void spacing-where boundary scattering is dominant-dispersion at the NPC unit cell level plays a noticeable role in determining the thermal conductivity. Finally, the focus shifts to 2D thin-films which has significant differences in the phonon band structure and exhibits lower values of thermal conductivity. A thorough modeling scheme is proposed that provides substantially more accurate results compared to the conventional formulation, which uses bulk dispersion in the prediction of the phonon thermal conductivity. The results show that the thin-film full dispersion model better fits with the experimental data over a large temperature range. Finally, features are added to the thin-film forming a nanoscale phononic metamaterial-a novel concept that yields further reduction in thermal conductivity and potentially a substantial improvement in the thermoelectric figure-of-merit