Electronic structure-inspired multiphysics modeling of complex dynamical processes

This talk will touch upon two intricately connected themes rooted in electronic structure theory arising in the seemingly disparate arenas of disordered plasmas and electrochemical carbon conversion.

A critical quantity for characterizing electron-transport in warm dense plasmas is the average ionization state of an atom (Z_eff), which is ill-defined for common static charge-partitioning schemes owing to an abundance of loosely bound electrons. In the first part of the talk, I will revisit dynamical Born effective charges from mixed stochastic time-dependent density functional theory (TD-DFT) to derive a novel atomic partitioning of a bulk electron response property that we call the “group conductivity.” A physically meaningful Z_eff for disordered metals hinges on the group conductivity, as I will illustrate with mixed liquid metals.

In the second part, I will describe an embedding theory in which a system is replaced by an embedded subsystem interacting with its extended environment via a uniquely defined embedding potential at the DFT level. Using embedded correlated wavefunction methods, we piecewise-refine the electronic structure description of the embedded region of interest to incorporate electron correlation effects inadequately captured by standard DFT functionals. We apply this multiphysics modeling framework to investigate the electrically powered catalytic reduction of CO$_2$ to value-added hydrocarbons on defective copper surfaces, a key step toward de-fossilization in electromanufacturing.