Donald Truhlar of the University of Minnesota- Twin Cities is supported by an award from the Theory, Models and Computational Methods program to develop methods for simulating complex, reactive dynamical systems in chemistry and for applying these methods to diverse systems. A focus of the research is the development of techniques to incorporate quantum mechanical effects in complex system. A key issue that the research addresses is the interface of electronic structure theory and dynamics. The PI and his research group are developing new approaches for efficiently computing potential energy surfaces and coupling terms in a manner that is practical for complex simulations. They are devising many novel methods including new ways to combine quantum mechanics and molecular mechanics, new methods for the diabatic representation of electronic states for use in dynamics calculations on photo-induced processes, new approaches to calculating the nuclear motion in coupled-state dynamics, new approaches to fitting coupled potential energy surfaces, and novel ways to model the participation of enzyme or other environmental coordinates in the reaction coordinate.
The long range objectives of this research involve (i) calculation of branching ratios for photochemical reactions, (ii) simulation of atmospheric nucleation processes and (iii) calculation of the rate constants, kinetic isotope effects, and temperature dependence of the rate constants for reactions catalyzed by enzymes. These research areas are of enormous technological importance especially for health, the environment, the material needs of society, industrial competitiveness and defense
Major activities and objectives This project involved the development of new methods simulating for complex systems in chemistry and included the application of these methods to prototype systems. I accomplished state-of-the-art research in important fundamental and practical areas with broad impact for many areas of chemistry and materials research, and I trained graduate, undergraduate, and postdoctoral students. I incorporated research findings into educational activities. A theme of the work is the development and application of new methods for the incorporation of quantum mechanical effects in complex systems. Systems may be complex because of their size (e.g., nanoparticles), the complexity of their constituents (e.g., enzyme, substrate, coenzyme, and solvent in the same system), or the complexity of the process (e.g., photochemistry involving closely interacting electronic states). Activities proposed included a more accurate treatment of vibrations, developing fragment and subsystem methods, developing more accurate density functionals for multi-reference systems based on new density functional design principles, improved methods for combining quantum mechanics and molecular mechanics, new path integral methods, and new methods for diabatic representations of electronic states for use in dynamics calculations on photo-induced processes. We made significant progress in all of these areas. Key achievements I published 72 journal articles and one book chapter based in whole or in part on research supported by this grant, and several computer codes were distributed in free, portable, well-documented form via our Web site. Here are some highlights of our accomplishments. I presented a general scheme for describing liquid-phase reactions in terms of free-energy surfaces and applied it to base-initiated alpha,beta-eliminations. We proposed a new way to include charge penetration effects in electrostatic modeling that is modest in cost, broadly applicable, and easily incorporated into existing codes. We showed that the interaction of a metal atom with a conjugated hydrocarbon can involve both one- and two-electron charge transfers near sequential conical intersections. We showed that the amount of charge transfer from a boron atom to the cage in an endohedral fullerene is a strong function of the radial distance of the atom from the center, controlled by multistate conical intersections. We presented a successful approach for retinal photochemistry by combining SCF calculations with symmetry-corrected spin-flip Tamm− Dancoff approximation calculations. We presented a new semiempirical molecular orbital method that includes polarization effects much better than previous low-cost methods. We developed a self-consistent state-specific vertical excitation model for electronic excitation in solution. We showed that carbene rotamer energetic effects are responsible for the inverse relation between organophosphine dissociation rate and catalytic activity in olefin metathesis precatalysts. We presented a new generalized gradient approximation (GGA) to the exchange-correlation (xc) functional of density functional theory, called SOGGA11, that is the first functional correct to second order to provide both reliable geometries and small errors for a diverse set of energetic data. We presented a meta-GGA xc functional, called M11-L, that employs a new strategy, dual-range local exchange, to provide broad accuracy for single-configurational and multiconfigurational molecules, solid-state lattice constants, and chemical reaction barrier heights. We proposed a nonseparable xc functional called N12 that is the first xc functional depending only on densities and density gradients that provides good accuracy for solid-state cohesive energies and lattice constants and molecular atomization energies and bond lengths. We reported the MN12-L xc functional with improved across-the-board performance, especially for atomization energies, ionization potentials, barrier heights, noncovalent interactions, isomerization energies of large molecules, and solid-state lattice constants and cohesive energies. We showed that screened charges in the electrostatically embedded many-body (EE-MB) method dramatically improves accuracy for water hexamers. We extended the EE-MB method to dipole moments and charge transfer. We presented new schemes for using diffuse basis functions in electronic structure calculations. We used noncollinear density functional theory to show how the low-spin state in a model of the oxygen-evolving complex of photosystem II avoids frustrated spin coupling. We presented Charge Model 5 for deriving partial atomic charges molecules and solids. We extended the X-Pol method to multilevel representations in which different electronic structure methods are used for different fragments. We presented methods to accelerate the convergence of Feynman path integral calculations of thermodynamic functions and an efficient procedure for calculating ensemble averages, thermodynamic derivatives, and coordinate distributions by effective classical potential methods. We reported electronically nonadiabatic dynamics calculations including spin–orbit coupling for the final-state-specific photodissociation of CH2ClBr. We reported fully converged quantal scattering calculations of state-selected rate constants for the Mu + H2 reaction. We presented a new self-consistent reaction-field implicit solvation model that has several advantages over all other previous generalized Born models. We added higher-order electronic polarization effects to the molecular tailoring approach and applied the improved method to a 20-residue polypeptide. We presented vibrational configuration interaction calculations that represent the most accurate calculation of vibrational energy levels for a molecule with five or more atoms.
