Effectiveness of nanoinclusions for reducing bipolar effects in thermoelectric materials

TL;DR

This study uses quantum mechanical simulations to evaluate how nanoinclusions can mitigate bipolar effects in thermoelectric materials, potentially enhancing their efficiency by reducing detrimental carrier transport.

Contribution

It provides a theoretical analysis demonstrating that nanoinclusions can effectively reduce bipolar transport effects, improving thermoelectric performance metrics.

Findings

Nanoinclusions reduce minority carrier transport and bipolar effects.

Benefits are greater with higher initial minority band conductivity.

Significant improvement in Seebeck coefficient and power factor observed.

Abstract

Bipolar carrier transport is often a limiting factor in the thermoelectric efficiency of narrow bandgap materials (such as Bi2Te3 and PbTe) at high temperatures due to the introduction of an additional term to the thermal conductivity and a reduction in the Seebeck coefficient. In this work, we present a theoretical investigation into the ability of nanoinclusions to reduce the detrimental effect of bipolar transport. Using the quantum mechanical non equilibrium Greens function (NEGF) transport formalism, we simulate electronic transport through two-dimensional systems containing densely packed nanoinclusions, separated by distances similar to the electron mean free path. Specifically, considering an n type material, where the bipolar effect comes from the valence band, we insert nanoinclusions that impose potential barriers only for the minority holes. We then extract the materials electrical conductivity, Seebeck coefficient, and electronic thermal conductivity including its bipolar contribution. We show that nanoinclusions can indeed have some success in reducing the minority carrier transport and the bipolar effect on both the electronic thermal conductivity and the Seebeck coefficient. The benefits from reducing the bipolar conductivity are larger the more conductive the minority band is to begin with (larger hole mean free path in particular), as expected. Interestingly, however, the benefits on the Seebeck coefficient and the power factor are even more pronounced not only when the minority mean free path is large, but when it is larger compared to the majority conduction band mean free path. Finally, we extract an overall estimate for the benefits that nanoinclusions can have on the ZT figure of merit.