Georgios Varnavides⁺, Adam S. Jermyn,⁺ Polina Anikeeva, Prineha Narang
Posted on Publisher's version
Electron hydrodynamics arises when momentum-relaxing scattering processes are slow compared to momentum-conserving ones. While the microscopic details necessary to satisfy this condition are material-specific, experimentally accessible current densities share remarkable similarities. We study the dependence of electron hydrodynamic flows on the rates of momentum-relaxing and momentum-conserving scattering processes in a microscopics-agnostic way. We develop a framework for generating random collision operators which respect crystal symmetries and conservation laws and which have a tunable ratio between the momentum-conserving and momentum-relaxing lifetimes. Using various random instances of these collision operators, we calculate macroscopic electron viscosity tensors and solve the Boltzmann transport equation (BTE) in a channel geometry over a grid of momentum-conserving and momentum-relaxing lifetimes, and for different crystal symmetry groups. We find that different random collision operators using the same lifetimes produce very similar current density profiles, meaning that the current density is primarily a probe of the overall rates of momentum conservation and relaxation. By contrast, the viscosity tensor varies substantially at fixed lifetimes, meaning that properties like channel resistance provide detailed probes of the underlying scattering processes. This suggests that, while details of the scattering process are imprinted in the electronic viscosity tensor, for many applications theoretical calculations of hydrodynamic electron flows can use experimentally-available lifetimes within a spatially-resolved BTE framework rather than requiring the costly computation of ab initio collision operators.