Metastable electronic states: electronic structure, dynamics, and chemistry

Trustees Of Boston University, Boston

Investigators

Abstract

Electron-molecule interactions often lead to complex chemistry initiated by electron capture into a temporary state that has enough energy to eject an electron, yet, lives long enough to trigger a chemical reaction. The lifetime of this metastable state, therefore, sets the timescale for the chemical conversion. Metastable electronic states are key intermediates in radiation damage of biomolecules and they are also routinely formed in highly energetic environments, e.g. plasmas. This research program proposes development of new models enabling quantitative predictions of the energies and lifetimes of metastable electronic states. The outlined computational studies are aimed at advancing the understanding of the role of metastable states in radiation damage of biological systems, in photovoltaics, and catalysis. The goals of the proposed research are three-fold: (i) enabling robust correlated treatment of Feshbach and multiply-excited resonances; (ii) incorporating nuclear motion via Born-Oppenheimer ab initio dynamics models; (iii) integrating electronic structure methods for resonances with density functional embedding approaches for the description of chemical reactions on metal surfaces. The proposed methods will be applicable to realistic molecular systems (~ 50-100 atoms). Specifically, a new method combining the complex absorbing potential approach and extended multiconfigurational quasidegenerate perturbation theory is proposed to enable calculations of Feshbach and multiply-excited resonances' position and widths. Since resonance decay via electron ejection and nuclear relaxation often occur on the same timescale, taking into account nuclear motion is crucial for understanding electron-molecular interactions. Born-Oppenheimer ab initio dynamics on complex potential energy surfaces will be implemented in the on-the-fly manner to describe interdependent electron ejection and nuclear motion. The simplicity of the model makes it applicable to large molecular systems. Finally, a hybrid approach combining electronic structure methods for resonances with density embedding techniques is proposed to account for the metastable character of temporary states involved in reactions on metal surfaces. A special emphasis is placed on implementing the developed methodologies as efficient codes available through widely used software packages and as open-source modules via the PI's website. The outreach program includes an annual computational chemistry workshop for high-school summer research students hosted by PI at Boston University.