Thermoelectric power factor limit of a 1D nanowire

TL;DR

This paper establishes a fundamental upper limit on the thermoelectric power factor of 1D nanowires, experimentally verifies it in InAs nanowires, and predicts achievable power factors in high-quality semiconductor nanowires.

Contribution

It introduces a theoretical limit for the thermoelectric power factor in 1D nanowires and experimentally confirms this limit in quasi-ballistic InAs nanowires.

Findings

Measured maximum power factors match the theoretical limit.

A single 1D electronic channel can achieve power factors of the order of mW/m-K^2.

High-quality, small cross-section nanowires can reach competitive thermoelectric performance.

Abstract

In the past decade, there has been significant interest in the potentially advantageous thermoelectric properties of one-dimensional (1D) nanowires, but it has been challenging to find high thermoelectric power factors based on 1D effect in practice. Here we point out that there is an upper limit to the thermoelectric power factor of non-ballistic 1D nanowires, as a consequence of the recently established quantum bound of thermoelectric power output. We experimentally test this limit in quasi-ballistic InAs nanowires by extracting the maximum power factor of the first 1D subband through I-V characterization, finding that the measured maximum power factors conform to the theoretical limit. The established limit predicts that a competitive power factor, on the order of mW/m-K^2, can be achieved by a single 1D electronic channel in state-of-the-art semiconductor nanowires with small cross-section and high crystal quality.