Tuning lattice thermal conductivity in NbMoTaW refractory high-entropy alloys: Insights from molecular dynamics using machine learning potential

Refractory high-entropy alloys (RHEAs) have attracted increasing interest due to their excellent mechanical properties under extreme conditions. However, the lattice thermal conductivity is still not well studied. In this paper, we calculate the lattice thermal conductivity of the NbMoTaW RHEA using the equilibrium molecular dynamics method with a machine learning-based interatomic potential. We find that with increasing Mo concentration, the lattice thermal conductivity increased from 1.72 to 2.16 W/mK, an increase of 25.6%. The underlying mechanism is explained by the phonon density of states and phonon mode participation. Increasing the Mo concentration can induce a blueshift in both the low-frequency and high-frequency phonons. Moreover, we find that at the frequency corresponding to the phonon density of states peak, the NbMo1.5TaW RHEA has the largest mode participation rate, which is the main reason for the anomalous lattice thermal conductivity. In addition, we investigate the effect of temperature on the lattice thermal conductivity and the results show that anharmonicity has a dominant effect. Finally, the effect of compressive strain on the lattice thermal conductivity is explored. Our work discloses that the phonon density of states associated with phonon mode participation plays critical roles in the thermal conductivity of the RHEA, rather than the previously recognized conformational entropy. This contributes to the understanding of the thermal behavior of RHEA and provides an effective route to tune its thermal conductivity.