Electronic Structure Theory for Photo-Induced Spin-Forbidden Dynamics

  • Shiozaki, Toru (PD/PI)

Project: Research project

Project Details


The objective of this research is to accurately model photo-induced spin-forbidden dynamics in spin-crossover processes and photoredox catalysis cycles by developing relativistic electronic structure methods that account for spin–orbit coupling to infinite order. We will interface these methods to the state-of-the-art dynamics models to allow for accurate on-the-fly spin-forbidden dynamics simulation. Our working hypothesis is that, to accurately model of spin-forbidden reactions (even those of lighter molecules), the molecular electronic structure must be modeled by theories that take into account spin–orbit coupling exactly. Our hypothesis is based on the fact that fully relativistic methods give qualitatively more accurate descriptions of intersystem crossing: Relativistic potential energy (PES) surfaces correspond to the ‘adiabatic’ picture, while conventional non-relativistic PES can be seen as the ‘diabatic picture. We will test this hypothesis by deriving and implementing analytical nuclear gradients of the two-component relativistic approaches, such as the zeroth-order approximation (ZORA), the effective core potential method (ECP), and the recently developed exact-decoupling method (X2C). Application to spin-crossover processes and activation steps of photoredox catalysts will be performed by interfacing the electronic structure methods to modern dynamics software, with which we aim to fundamentally change the way we understand spin-forbidden transitions. The electronic structure programs will be implemented in the BAGEL program package developed in our research group. BAGEL is open-source and freely available to end users, promising broader impacts of the proposed research.
Effective start/end date12/1/145/31/18


  • Air Force Office of Scientific Research (FA9550-15-1-0031/P00001)


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.