Skip to main content

MN15 and GAM+U: New Kohn-Sham Density Functionals with Broad
Accuracy for Multi-Reference and Single-Reference Systems, Noncovalent
Interactions, and Solid-state Properties


EMSL Project ID
49213

Abstract

Kohn-Sham density functional theory is widely used; however, no currently
available exchange-correlation functional can predict all chemical properties with chemical accuracy. We plan to develop new density functionals that can be broadly applied for molecular and solid-state properties. A hybrid density functional, called MN15, is being designed with the purpose of having broader accuracy for molecular systems than any currently available one. The properties considered in the parameterization include main-group bond energies, transition metal bond energies, ionization potentials, electron affinities, proton affinities, reaction barrier heights, non covalent interactions, hydrocarbon thermochemistry, excitation energies, isomerization energies, sulfur molecule atomization energies, and molecular structures. Our training set also includes 87 data for noncovalent interactions and 17 data for absolute atomic energies. We will further test MN15 for 10 transition-metal coordination energies, the entire S66x8 database of noncovalent interactions, 21 transition metal reaction barrier heights, 69 electronic excitation energies of organic molecules, 31 semiconductor band gaps, and seven transition metal dimer bond lengths. Another density functional, which is a gradient approximation to the exchange-correlation functional, called GAM+U, is being developed with the goal of correctly predicting solid-state properties such as lattice constants, band gaps, cohesive energies, bulk moduli, and adsorption energies. It will have the advantage of having the same accuracy as hybrid functionals (containing nonlocal Hartree-Fock exchange) but at much lower cost. The Hubbard U correction in the formalism of Dudarev et al. will be adjusted to find values that correctly predict solid-state properties. A first step in this direction is to find the value of U that gives accurate values for both band gaps and lattice constants of our database of 31 solids.

Project Details

Start Date
2016-02-19
End Date
2016-09-30
Status
Closed

Team

Principal Investigator

Donald Truhlar
Institution
University of Minnesota

Team Members

Kelsey Parker
Institution
University of Minnesota

Xiaoyu Li
Institution
University of Minnesota

Bo Long
Institution
University of Minnesota

Laura Fernandez
Institution
University of Minnesota

Pragya Verma
Institution
University of Minnesota

Haoyu Yu
Institution
University of Minnesota

Kaining Duanmu
Institution
Pacific Northwest National Laboratory

Related Publications

Duanmu K ,Roberto-Neto O ,Machado F B,Hansen J A,Shen J ,Piecuch P ,Truhlar D G 2016. "Geometries, Binding Energies, Ionization Potentials, and Electron Affinities of Metal Clusters: Mgn 0, ± 1, n = 1?7" Journal of Physical Chemistry C 120(24):13275–13286. 10.1021/acs.jpcc.6b03080
Yu H ,He X ,Truhlar D G 2016. "MN15-L: A New Local Exchange-Correlation Functional for Kohn–Sham Density Functional Theory with Broad Accuracy for Atoms, Molecules, and Solids" Journal of Chemical Theory and Computation 12(3):1280-1293. 10.1021/acs.jctc.5b01082
Yu H ,Zhang W ,Verma P ,He X ,Truhlar D G 2016. "Nonseparable Exchange–Correlation Functional for Molecules, Including Homogeneous Catalysis Involving Transition Metals" Physical Chemistry Chemical Physics. PCCP 17():12146. 10.1039/c5cp01425e