Spin-selective photocatalysis and quantum transport using ab initio density-matrix dynamics

Spin-selective photocatalysis and quantum transport using ab initio density-matrix dynamics Yuan Ping, University of California, Santa Cruz Ravishankar Sundararaman, Rensselaer Polytechnic Institute The capability to control chemical reactions using the spin of electrons, in the rapidly emerging field of “Spin Chemistry”, provides exciting opportunities to produce desired molecules and solids with high selectivity and enhance the energy efficiency of chemical synthesis. Simultaneously, the increased focus on manipulating spin in materials for quantum information is leading to advanced techniques for precise spin control, which can further advance the potential of spin-dependent catalysis. Realizing the promise of spin chemistry now requires a mechanistic understanding of spin effects on chemical reactivity, in turn necessitating computational prediction of the generation, dynamics and chemical impact of spin in catalytic materials. The overarching goal of this research program is to develop tools for simulating spin dynamics and the impact of spin on catalysis and photochemistry, including the effects of optical orientation and material chirality typically used to manipulate the spin states. To this end, we will quantitatively predict far-from-equilibrium quantum dynamics and transport, with electrons and spin degrees of freedom coupled to phonons, photons, defects and liquid environments. We will develop a fully general, massively parallel computational platform for first-principles quantum dynamics and transport. In particular, this will leverage our recently-developed ab initio density-matrix formalism that accounts for both coherent and incoherent dynamics, and extend it beyond the Markovian approximation to include memory effects for spin-selective photocatalysis, and to include spatial inhomogeneity for quantum transport. To directly connect with experimental probes of spin dynamics, we will integrate predictive capability for optical and X-ray probes in our ultrafast dynamics simulations. We will use this framework to simulate spin-selective electron excitation, relaxation and decoherence, as well as transport in molecular and hybrid materials – especially chiral systems, elucidating fundamental mechanisms of chiral-induced spin selectivity and spin-dependent photocatalysis. We will develop this predictive capability targeting spin chemistry in four thrusts: 1. Ab-initio quantum dynamics of spin coherence, including coupling with environments such as phonons and solvents, extended beyond the Markovian approximation to include memory effects. 2. Quantum spin transport with ab initio scattering and many-body interactions, going beyond predictions of transport coefficients such as mobility and spin diffusion lengths, towards direct spatial evolution of entangled quantum states. 3. Spectroscopic signatures of spin and chirality, including time-dependent circular dichroism and optical rotation signals for both finite and extended systems, spanning optical and X-ray probes to connect with a wide range of experimental measurements. 4. Quantum dynamics in solvated environments including incoherent evolution of solute electronic states due to dissipation in the solvent response, allowing treatment of solvent damping effects without explicit solvent molecules at the electronic structure level. We will distribute these techniques in optimized exascale-ready open-source code, leveraging our existing density-matrix Lindbladian dynamics framework. We will thereby arm the computational chemistry community with predictive modeling of coherent and incoherent spin dynamics, paving the way towards detailed mechanistic understanding of Spin Chemistry.