T h e . N u m b e r s . o f . C h e m i s t r y
Ernest Davidson's laboratory is in the forefront of modern chemistry, housed within one of the more attractive buildings at Indiana University Bloomington. Stone shields with chemical names emblazoned on them deck the walls; glass display cases show both instruments and photographs of instruments from generations past.
But Davidson's laboratory itself does not smell of fermenting chemicals. You will find not one beaker, flask, or test tube. Nothing bubbles, nothing froths. Instead, on a chalk board are written the four subjects that Davidson addresses, alone and in collaboration: spectra, reactivity, structure, and basic theory (though he admits a penchant for the last). Then there is Davidson, himself: an accomplished man, he is a Distinguished Professor, a member of the National Academy of Sciences and the American Academy of Arts and Sciences. He has a desk, some books--and a computer.
Rather than beakers, the computer is Davidson's tool. He uses it to explore the chemistry of numbers--or, more precisely, the numbers of chemistry. Davidson's hours are spent contemplating a purely mathematical description of reality, forging ahead with numbers where observable facts are hard to come by. He is a "quantum chemist," a name that appropriately captures the quasi-chemistry, quasi-physics nature of his work. The balance is tipped finally in favor of chemistry, though, for the issue here is molecules and their interactions. Yet he investigates these conglomerations of atoms by solving equations that are themselves designed to describe the behavior of subatomic particles.
Because the actual motion of atoms in molecules can't be followed in the lab, Ernest Davidson, Distinguished Professor of Chemistry, Indiana University Bloomington, investigates conglomerations of atoms by solving equations that are designed to describe the behavior of subatomic particles. This diagram shows a model of a molecule in the middle of a chemical reaction. The hydrogen atom labeled H8 is in the process of breaking its bond to carbon C4 and reattaching itself to carbon C7. --credit
"The structure of certain molecules can't be measured well in the lab," Davidson says. "So we calculate them mathematically using electron equations to see what the structure should be. Then we compare our answers to the limited data we have and see if it's a good fit."
Davidson was a senior graduate student here at IU in 1961, when all mathematical research was done on an old IBM 650 at the first IU computing center. He logged thousands of hours of computer time then, nearly all of it after dark when the machine was free. Then, after teaching at the University of Washington for more than twenty years, he returned to IU in 1984. "Back then nobody thought to do chemistry purely by the numbers," he says, "because there were no computers to do it on." Nowadays, the Chemistry Building houses more than 200 PCs and fifty UNIX workstations, and no fewer than three professors who do computational chemistry as their primary activity- working, as Davidson does, solely on a numerical approach to their subject.
What's more, more than half the papers he sees published today in The Journal of Chemical Physics (for which he is an associate editor) and elsewhere have "some component" of theoretical work to them; often as much as half the paper is devoted to it. "Twenty or thirty years ago that was simply not the case. Virtually everything back then was experimental," Davidson says. Modern computing has had such an impact on how chemistry is done, he points out, that the last Nobel Prize in chemistry was awarded for work done in density functional theory, that is, describing through mathematics how electrons behave in molecules.
Other recent changes in information technology have altered the parameters of what goes on in this laboratory as well. Perhaps just as influential as commercially available software that can calculate spectra, molecular structure, and reaction paths in one laboratory is the Internet itself. The net has allowed for a boom in collaborative efforts, which have the potential for making headway into the secrets of chemistry far faster and more effectively than the most gifted lone researcher can. "I am collaborating now," Davidson muses, "with co-workers in South Carolina; Vancouver, British Columbia; China; and France." Nor is the situation one that necessarily involves travel, letter writing, or even phone calls. Immediate net access to any team member's results is now possible around the globe; people separated by a continent or more can routinely consult the same up-to-the-minute graphs, altering their theories as new data comes in. "Joint collaboration around the world," he smiles, "is the real information technology revolution."
"There are always more problems to be solved in chemistry," Davidson muses from his beakerless laboratory, where nothing bubbles, nothing froths. "The problems we address at any given time are the ones our computers can handle." It is clear that as the supercomputing age progresses, Davidson and the other quantum chemists will be right there addressing those problems.--William Orem
For more information on the Web:
Return to the Table of Contents