Comparing first-principles computational methods for electrical conductivity: time-dependent density functional theory versus Kubo-Greenwood

Hydrodynamic simulations of high-energy density experiments rely on accurate electrical conductivity models. The Kubo-Greenwood method with density functional theory (DFT-KG) is the gold standard for conductivity calculations but faces challenges like the need to extrapolate the DC limit and inaccurate treatment of electron-electron scattering. In comparison, real-time time-dependent DFT (TDDFT) could describe more many-body behavior by calculating conductivities through the aggregate current density and provides direct access to DC limits. However, TDDFT can also face numerical issues, including spurious long-time behavior from finite size effects and insufficient k-point sampling. This work systematically characterizes the numerical sensitivities of TDDFT and compares them to DFT-KG for calculating conductivities in aluminum and beryllium under warm dense conditions. Preliminary results show good agreement between TDDFT and DFT-KG for optical conductivities. Further comparisons at extreme conditions will assess TDDFT’s description of electron-electron scattering. These findings highlight the need to evaluate the numerical sensitivities for the effective application of these methods and quantifying uncertainties that propagate into hydrodynamic simulations.

SNL is managed and operated by NTESS under DOE NNSA contract DE-NA0003525.