First-Principles Calculations of Optoelectronic and Transport Properties of Materials for Energy Applications.

Abstract

Modern semiconductor technology and nanoengineering techniques enable rapid development of new materials for energy applications such as photovoltaics, solid- state lighting, and thermoelectric devices. Yet as materials engineering capabilities become increasingly refined, the space of controllable properties becomes increasingly large and complex. Selecting the most promising materials and parameters to focus on represents a significant challenge. We approach this challenge by applying state-of-the-art predictive first-principles calculation methods to guide research and development of materials for energy applications. This work describes our first-principles investigations of nanostructured group-III-nitrides for solid-state lighting applications and bulk titanium dioxides for thermoelectric applications. We demonstrate several remarkable properties of nanostructured group-III-nitrides. In InN nanowires with diameters on the order of 1 nm, we predict that quantum confinement shifts optical emission into the visible range at 2.3 to 2.5 eV (green to cyan) and results in a large exciton binding energy of 1.4 eV. These findings offer a new approach to addressing the ”green-gap” problem of low efficiency in solid-state lighting devices emitting in this part of the spectrum. In ultra-thin GaN-AlN quantum wells, we show how to adjust the well and barrier thicknesses for tuning the optical gap in the deep ultraviolet range between 3.85 and 5.23 eV. Furthermore, we predict that quantum confinement in ultra-thin GaN wells results in large exciton binding energies between 80 and 210 meV and enhances radiative recombination by reducing the exciton lifetime to as short as approximately 1 ns at room temperature. These findings highlight the capability of quantum-confined group-III-nitrides to improve the efficiency and utility of visible and ultraviolet solid-state light emitters. Additionally, we calculate the n-type thermoelectric transport properties of the naturally occurring rutile, anatase, and brookite polymorphs of TiO2 and predict optimal temperatures and free-carrier concentrations for thermoelectric energy conversion. We also predict a theoretical limit on the figure of merit ZT of 0.93 in the rutile polymorph, demonstrating that TiO2 can potentially achieve thermoelectric energy conversion efficiency comparable to that of commercialized thermoelectrics.