News Release

Researchers expanding ability of computers to simulate real-world chemical reactions

Peer-Reviewed Publication

Virginia Tech

Blacksburg, Va., April 7, 2002 -- Computers, it turns out, can't handle every problem at the speed of electrons. Ask them to provide a simulation of a complex molecule -- more than about 15 atoms -- and you could wait years for the outcome of an experiment. But researchers from Virginia Tech and Bethel College of St. Paul, Minn., may have an approach to expanding the capability of quantum chemistry.

The research will be presented at the 223rd national meeting of the American Chemical Society, April 7-11 in Orlando.

Virginia Tech chemistry professor T. Daniel Crawford uses the power of the computer to study chemistry. "There are many chemical reactions that you don't want to do in the lab. Perhaps they are too slow or too toxic or too explosive," he explains. "Or there may be micro- or nanoscopic components of reactions you can't probe experimentally.

"Presently, quantum chemistry is an important tool for studying small-molecule reactions, such as the depletion of ozone in atmospheric chemistry," says Crawford. "It is also used to calculate the properties of molecules that make up the clouds between and around stars."

"We are trying to expand computational modeling to molecules between 30 and 50 atoms to provide a much broader view of chemistry, so the power of quantum chemistry can be used for biochemistry, medical chemistry, or materials science, for instance," says Crawford.

The goal of quantum chemistry is to simulate real-world chemical reactions on the computer. "If a reaction involves only small molecules, then it is possible to carry out a series of calculations of greater and greater complexity, until the properties in which one is interested, such as the structure of a product molecule, are determined to extremely high accuracy -- sometimes even greater than could be measured experimentally."

For computer computations to be accurate -- to really simulate experiments -- the models tend to be very time-consuming. For example, to calculate the energy of a simple amino acid called valine would take five days on state-of-the-art computers, Crawford says. The same calculation on two valine molecules together (a valine-valine dimer) would take 10 years. As if the time frame weren't daunting enough, replicating such an experiment is discouraging.

So, Crawford, and Rollin A. King, professor of chemistry at Bethel College, are trying to reduce the scaling without losing accuracy. "We are using a technique called 'local correlation'. The basic idea is to describe the electronic interactions in a molecule piece by piece. You shouldn't have to consider interactions of electrons on parts of the molecule that are far apart, in general, so by breaking the molecule into fragments, we can reduce the cost (time) of the calculation."

###

Crawford and King will present "A locally correlated equation-of-motion coupled cluster approach for excited states of large molecules" (Comp 2) at 9:20 a.m. Sunday, April 7, at Convention Center room 108A, level one.

Crawford and King have worked together since being Ph.D. students at the University of Georgia. The research is funded by Crawford's NSF Career award, along with grants from the U.S. Department of Energy and the Jeffress Memorial Trust.

Contact for more information:
Dr. Crawford, 540-231-7760, crawdad@vt.edu
www.chem.vt.edu/chem-dept/crawford/homepage.htm

PR CONTACT:Sally Harris (540) 231-6759 slharris@vt.edu


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.