Lattice thermal conductivity of Si/Ge composite thermoelectric material: Effect of Si particle distribution

Embedding nanoparticles (NPs) in a matrix can effectively enhance the phonon scattering by the interface, reduce the lattice thermal conductivity, and improve the thermoelectric properties of the material. However, the understanding of how the distribution of embedded NPs affects the thermal conductivity is still not clear. To explore the underlying mechanism, frequency-dependent Monte Carlo simulation and the effective medium method are applied to study the lattice thermal conductivity of Si/Ge composite (Si NPs embedded in Ge matrix). The effect of the free path distribution (FPD) of Ge phonon induced by the heterogeneous distribution of Si NPs is introduced into the effective medium method, and then, this method is used to calculate the lattice thermal conductivity of Si/Ge composite when Si NPs are unevenly distributed. Results show that decreasing the separation distance of adjacent NPs can slightly decrease the lattice thermal conductivity. Assuming that the FPD of Ge phonon induced by Si-Ge interface scattering obeying lognormal distribution and that the deviation sigma indicates the degree of inhomogeneity of Si NPs distribution, lattice thermal conductivity of composites with different sigma is obtained. It is found that lattice thermal conductivity significantly decrease by more than 40%, with the increase of sigma, especially for high-Si concentrations. The present study indicates that the particle distribution in a composite can markedly affect the lattice thermal conductivity.