Effects of Antisite Defects on Seebeck Coefficient in Fe_2VAl -- Analyses based on Bipolar Random Anderson Model

TL;DR

This paper proposes a microscopic mechanism explaining the sign change of the Seebeck coefficient in Fe_2VAl due to antisite defects, using the bipolar random Anderson model and self-consistent T-matrix approximation.

Contribution

It introduces a novel bipolar random Anderson model approach to explain how antisite defects cause a sign change in the Seebeck coefficient in Fe_2VAl.

Findings

Antisite defects induce new states in the band overlap region.

Higher scattering rate for hole carriers causes negative Seebeck coefficient.

Mechanism offers a new way to control thermoelectric properties.

Abstract

A microscopic mechanism is proposed for a dramatic sign change of the Seebeck coefficient from positive to negative sign by the introduction of antisite defects in Fe$_2$VAl based on bipolar random Anderson model (BPRAM), which incorporates hybridization effects between randomly distributed antisites and host bands, where the valence and conduction bands are treated separately due to their separation in momentum space. Applying a self-consistent T-matrix approximation, we find that antisite defects in Fe$_2$VAl induce new states in the band overlap region, resulting in a scattering rate that is higher for hole carriers in the valence band than that for electron carriers in the conduction band, leading to negative Seebeck coefficient. This mechanism of sign change presents a potential new approach for controlling thermoelectric properties in semimetallic systems without changing carrier concentration.