You are here

Parallel Two-Electron Reduced Density Matrix Based Electronic Structure Software for Highly Correlated Molecules and Materials

Award Information
Agency: Department of Defense
Branch: Army
Contract: W911NF-16-C-0124
Agency Tracking Number: A2-6270
Amount: $495,770.00
Phase: Phase II
Program: STTR
Solicitation Topic Code: A14A-T013
Solicitation Number: 2014.0
Timeline
Solicitation Year: 2014
Award Year: 2016
Award Start Date (Proposal Award Date): 2016-09-21
Award End Date (Contract End Date): 2017-09-25
Small Business Information
6601 Owens Drive
Pleasanton, CA 94588
United States
DUNS: 837635556
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Eugene DePrince
 (850) 645-1291
 deprince@chem.fsu.edu
Business Contact
 Hilary Pople
Phone: (773) 343-6072
Email: hilary.pople@gmail.com
Research Institution
 Florida State University
 Gary Ostrander
 
Sponsored Research Administration 874 Traditions Way, Third Floor
Tallahassee, FL 32306-4166
United States

 (850) 644-1464
 Nonprofit College or University
Abstract

Variational two-electron reduced-density-matrix (v2RDM) methods can provide a reference-independent description of the electronic structure of many-electron systems that naturally captures multireference correlation effects. These methods offer one of the few possible routes to performing the large-active-space computations that are necessary for the qualitative description of strongly-correlated molecules and materials. However, existing implementations of the v2RDM method exhibit a time-to-solution that is often prohibitively large for the general description of extended molecules. We are developing a parallel implementation of the v2RDM approach within the Q-Chem quantum chemistry package. We have implemented two- and three-particle N-representability conditions that are generalized to treat both closed- and open-shell systems, and we have interfaced our v2RDM solver with an orbital optimization procedure, thereby achieving a v2RDM-driven complete active space self-consistent field procedure. We will improve the parallel performance of our algorithms and expand the functionality of our software, fully integrating it with the useful features available in Q-Chem, including effective core potentials, methods for incorporating solvent effects, and analytic gradient evaluation. Our final software product will also include v2RDM-based approaches for the treatment of dynamical electron correlation and electronically excited states.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government