Modulation doping and energy filtering as effective ways to improve the thermoelectric power factor

Abstract

Thermoelectric (TE) materials have undergone revolutionary progress over the last 20 years. The thermoelectric figure of merit ZT, which quantifies the ability of a material to convert heat into electricity has more than doubled compared to traditional values of $$ZT\sim 1$$ZT~1, reaching values even beyond $$ZT\sim 2$$ZT~2 in some instances. These improvements are mostly attributed to drastic reductions of the thermal conductivity in nanostructured materials and nanocomposites. However, as thermal conductivities in these structures approach the amorphous limit, any further benefits to ZT must be achieved through the improvement of the thermoelectric power factor. In this work we review two of the most promising avenues to increase the power factor, namely (i) modulation doping and (ii) electron energy filtering, and present a computational framework for analysis of these mechanisms for two example cases: low-dimensional gated Si nanowires (electrostatically achieved doping), and superlattices (energy filtering over potential barriers). In the first case, we show that a material with high charge density, but free of ionized impurities, can provide up to a five-fold thermoelectric power factors increase compared to the power factor of the doped material, which highlights the benefits of modulation doping, or gating of materials. In the second case, we show that optimized construction of energy barriers within a superlattice material geometry can improve the power factor by up to $$\sim 30\,\%$$~30%. This paper is intended to be a review of our main findings with regards to efforts to improve the thermoelectric power factor through modulation doping and energy filtering.