Phoebe: a collection of Phonon and Electron Boltzmann Equation solvers

Understanding the electrical and thermal transport properties of materials is critical to the design of all kinds of devices. Computational modeling nowadays allows for accurate in-silico predictions of materials properties, providing insights and guidelines for materials design. In order to accurately simulate materials transport properties, it is necessary to achieve an accurate description of electron and phonon properties, including the effect of interactions between carriers. While first-principles methods are capable of describing the microscopic properties of each carrier, using them to compute transport properties is still a formidable task, computationally demanding and memory intensive, as it requires carefully assembling all pieces of microscopic information to provide a picture of a macroscopic transport property. To address this challenge, we present a newly developed software package, Phoebe (a collection of PHOnon and Electron Boltzmann Equation solvers), which evaluates the effect of electron-phonon, phonon-phonon, boundary, and isotope scattering to predict the electron and phonon transport properties of materials. This open source C++ code combines MPI-OpenMP hybrid parallelization with GPU acceleration and distributed memory structures to manage computational cost, allowing Phoebe to effectively take advantage of contemporary computing infrastructures. Using this new framework, Phoebe can accurately and efficiently predict a wide range of transport properties such as the electrical, thermal conductivity, and thermoelectric performance, opening the way for the computational analysis of large and complex crystals.