Non-adiabatic dynamics
The Born-Oppenheimer (BO) approximation, which assumes electrons rearrange instantly on the timescale of nuclear motions is the basis for the concept of Potential Energy Surfaces and therefore underlies much of our understanding of chemistry. However, the BO approximation breaks down when the splittings between occupied electronic states become comparable to the kinetic energy of the nuclei, and then one must consider the dynamics of both the electrons and nuclei simultaneously. These so-called non-adiabatic effects are especially important in excited-state reactions, which occur in many systems of biological and technological relevance, such as fluorescent proteins, photosynthesis, DNA damage, the primary event of vision, and organic photovoltaics.
We are developing ab initio methods to perform excited-state dynamics simulations in order to learn how non-adiabatic effects influence a reaction, and how they might be controlled to give a desired reaction product. One research effort is in developing robust electronic structure methods that yield continuous excited-state potential energy surfaces over the entire course of a reaction- a prerequisite to any dynamics simulation. In addition, we are developing efficient multiscale approaches to extend the reach of ab initio excited-state dynamics simulations from isolated molecules in the gas phase to solvated molecules in the condensed phase, and all the way up to aggregates of chromophores in complex environments, such as photosynthetic reaction center complexes. A focus in our work is in making direct connection to experiment by simulating experimental observables.
Selected relevant publications:
X. Liu, A. Humeniuk, W. J. Glover, “Conical intersections in solution with polarizable embedding: Integral-exact direct reaction field”, J. Chem. Theory Comput. 18, 6826-6839 (2022) doi.org/10.1021/acs.jctc.2c00662
A. S. P. Paz, W. J. Glover, “Diabatic Many-Body Expansion: Development and Application to Charge-Transfer Reactions”, J. Chem. Theory Comput. 17, 1497 (2021) doi.org/10.1021/acs.jctc.0c01231
W. J. Glover, A. S. P. Paz, W. Thongyod, C. Punwong, “Analytical Gradients and Derivative Couplings for Dynamically Weighted Complete Active Space Self-Consistent Field”, J. Chem. Phys. 151, 201101 (2019) doi.org/10.1063/1.5130997
W. J. Glover, “Smoothing out excited-state dynamics: Analytical gradients for dynamically weighted complete active space self-consistent field”, J. Chem. Phys. 141, 171102 (2014) doi.org/10.1063/1.4901328
Quantum-Biased Molecular Optimizations and Dynamics
Traditional quantum chemistry calculations proceed by choosing a particular arrangement of atoms and solving the electronic structure problem to find where the electrons go. However, sometimes we would like to do the reverse and find the optimal arrangement of atoms for a given desired electronic structure. Recently, we developed a class of methods that achieve this “quantum biasing”, both for molecular dynamics and geometry optimizations. The methods are particularly suited to reactions where the motions of electrons (or holes) themselves constitute the important reaction coordinate, and we are applying them to biologically relevant systems, such as the damage of DNA from ionizing radiation.
Selected relevant publications:
A. S. P. Paz, W. J Glover, “Efficient analytical gradients of property-based diabatic states: Geometry optimizations for localized holes”, J. Chem. Phys. 158, 204107 (2023) doi.org/10.1063/5.0142590
W. J. Glover, B. J. Schwartz, “The Fluxional Nature of the Hydrated Electron: Energy and Entropy Contributions to Aqueous Electron Free Energies”, J. Chem. Theory Comput. 16, 1263 (2020) doi.org/10.1021/acs.jctc.9b00496
J. R. Casey, B. J. Schwartz, W. J. Glover, “Free Energies of Cavity and Noncavity Hydrated Electrons Near the Instantaneous Air/water Interface”, J. Phys. Chem. Lett. 7, 3192 (2016)
doi.org/10.1021/acs.jpclett.6b01150
W. J. Glover, J. R. Casey, B. J. Schwartz, “Free Energies of Quantum Particles: The Coupled-Perturbed Quantum Umbrella Sampling Method”, J. Chem. Theory Comp. 10, 4661 (2014) doi.org/10.1021/ct500661t
Graphical Processing Unit Quantum Chemistry
To tackle the large systems of interest to our group, we use and contribute to the GPU-based quantum chemistry code Terachem. Furthermore, we have exclusive use of a large GPU cluster at NYU Shanghai with over 1.2 PetaFLOPS of computing power.
Selected relevant publications:
A. Humeniuk, W. J Glover, “Efficient CPU and GPU implementations of multicenter integrals over long-range operators using Cartesian Gaussian functions”, Comput. Phys. Commun. 280, 108467 (2022) doi.org/10.1016/j.cpc.2022.108467