Optimizing phonon scattering by nanoprecipitates in lead chalcogenides

TL;DR

This paper investigates how nanoprecipitates can be optimized to significantly lower the thermal conductivity of lead chalcogenides, using ab-initio calculations and exploring size distribution effects.

Contribution

It introduces a detailed ab-initio-based approach to model phonon scattering by nanoprecipitates and demonstrates how size distribution tuning can further reduce thermal conductivity.

Findings

Precipitates with realistic size distributions outperform single-size models.

Thermal conductivity reduction correlates with phonon spectrum and mass density difference.

Optimal nanoprecipitate design can significantly enhance thermoelectric performance.

Abstract

We calculate the thermal conductivity of PbTe and PbS with seven different types of nanoprecipitates using an ab-initio-based Boltzmann transport approach. We find that precipitates with realistic size distributions can reduce the thermal conductivity well below the predictions of theoretical models assuming a single precipitate size.We explore the question of how to tune this distribution to reduce the thermal conductivity even further. The predicted minimum value is strongly correlated with the phonon spectrum of the host material and with the mass density difference between the host and the inclusions.