27 October 2017 in Physical Review B
For decades, the three-phonon scattering process has been considered to govern thermal transport in solids, while the role of higher-order four-phonon scattering has been persistently unclear and so ignored. However, recent quantitative calculations of three-phonon scattering have often shown a significant overestimation of thermal conductivity as compared to experimental values. In this Rapid Communication we show that four-phonon scattering is generally important in solids and can remedy such discrepancies. For silicon and diamond, the predicted thermal conductivity is reduced by 30% at 1000 K after including four-phonon scattering, bringing predictions in excellent agreement with measurements. For the projected ultrahigh-thermal conductivity material, zinc-blende BAs, a competitor of diamond as a heat sink material, four-phonon scattering is found to be strikingly strong as three-phonon processes have an extremely limited phase space for scattering. The four-phonon scattering reduces the predicted thermal conductivity from 2200 to 1400 W/m K at room temperature. The reduction at 1000 K is 60%. We also find that optical phonon scattering rates are largely affected, being important in applications such as phonon bottlenecks in equilibrating electronic excitations. Recognizing that four-phonon scattering is expensive to calculate, in the end we provide some guidelines on how to quickly assess the significance of four-phonon scattering, based on energy surface anharmonicity and the scattering phase space. Our work clears the decades-long fundamental question of the significance of higher-order scattering, and points out ways to improve thermoelectrics, thermal barrier coatings, nuclear materials, and radiative heat transfer.
5 June 2009 in Nano Letters
15 April 2021 in ACS Applied Materials & Interfaces
1 January 2022 in Computer Physics Communications
1 August 2017 in Solar Energy Materials and Solar Cells
6 January 2016 in Physical Review B
Recently, first principle-based predictions of lattice thermal conductivity from perturbation theory have achieved significant success. However, it only includes three-phonon scattering due to the assumption that four-phonon and higher-order processes are generally unimportant. Also, directly evaluating the scattering rates of four-phonon and higher-order processes has been a long-standing challenge. In this work, however, we have developed a formalism to explicitly determine quantum mechanical scattering probability matrices for four-phonon scattering in the full Brillouin zone, and by mitigating the computational challenge we have directly calculated four-phonon scattering rates. We find that four-phonon scattering rates are comparable to three-phonon scattering rates at medium and high temperatures, and they increase quadratically with temperature. As a consequence, of Lennard-Jones argon is reduced by more than 60% at 80 K when four-phonon scattering is included. Also, in less anharmonic materials—diamond, silicon, and germanium— is still reduced considerably at high temperature by four-phonon scattering by using the classical Tersoff potentials. Also, the thermal conductivity of optical phonons is dominated by the fourth- and higher-orders phonon scattering even at low temperature.