Virtual Chemistry — or, How To Do Chemistry Without Chemicals

After snorkeling among the parrot-fish, reef sharks and spangled-emperors of the Lord Howe Island lagoon, Tim Lovell rested on the thwarts of the tour boat. As his fellow divers emerged to dry themselves in the South Pacific sun, someone asked this visiting Englishman what he was doing at this Australian national park. Lovell leaned back, closed his eyes and said in his proper British accent, "Right now, I am working out the structure of classic transition metal dimer systems."

Lovell, a doctoral candidate on an Engineering and Physical Sciences Research Council Overseas Studentship to the Australian National University (ANU) in Canberra, does all his work on computer screens, in the world of virtual chemistry. Practitioners of this field, properly known as "computational chemistry," use powerful quantum mechanical programs and desktop workstations to unveil the geometry and chemical properties of molecules. "Once you have the structure down on paper, you can begin to understand the function," Lovell explains. Practical applications include the modeling of everything from catalysts to superconductors.

By stockpiling his computer with data and then programming it to run a series of successive calculations over time, Lovell can crunch numbers for his thesis even while taking a much-needed break to snorkel off a semi-tropical reef. (The meticulous and time-consuming labor of interpretation comes later.)

The computer offers another benefit as well, Lovell says — it keeps him out of the lab, where he had the unfortunate habit of breaking things. "Glassware is a nightmare," he proclaimed, citing the time he accidentally poked a spatula through a beaker while trying to extract the sediment that was his experiment's end product.

Many chemists have similar stories from their undergraduate days; one eminent professor (who wished to remain anonymous) described how he once unintentionally mixed bromine and ammonia together, creating a thick, yellow-brown fog that forced classes to be cancelled. "I was a hero to my classmates for weeks afterward," he recalled.

But for Lovell, such mishaps were more unnerving. A final year experiment at the University of Bath was "disastrous" when he forgot to remove the stopper from a flask that was supposed to be left open. The plug caused pressure to mount inside the glass until it would have exploded but for an observant professor's fast reflexes.

With dry understatement, Lovell reminisced: "That's when I realized wet chemistry was not for me."

Salvation arrived in the form of Robert Deeth, a Cambridge-educated computational chemist from Australia. Deeth was gifted at explaining his field's arcane details in a witty and entertaining way, and Lovell says Deeth's lectures inspired him to turn to chemistry via computer. Among other advice, Deeth told Lovell: "It doesn't matter when you work, as long as you get the work done." This idea appealed to Lovell, whose well-rounded interests included an assault on professional golf. (He was undefeated Player of the Year at his university.) Since computational chemistry meant working anywhere at anytime, Lovell could combine his two loves — the intellectual challenge of chemistry and the physical challenge of sports.

The practice of computational chemistry offers many such benefits, says Robert Stranger, Lovell's current advisor and head of the Inorganic Computational Chemistry Group: it's a new field with room to grow, free from the handling of toxic chemicals, and allows researchers the opportunity to look at transitory complexes that appear and disappear too quickly to be measured by physical means during a reaction.

Computational chemistry also has the advantage of being applicable to the whole of the chemical realm. "Most chemists have to stick to one specialty, but we can indulge our intellectual curiosity and explore all the different domains," Stranger said. "If anything, it has expanded my interest in all areas of chemistry." However, he warns, "Computational chemistry is an additional tool for the experimental chemist, not a substitute for hands-on experience with chemicals, beakers and Bunsen burners."

Much of the field would not have been possible even a decade ago. Constant advances in hardware and software now enable smaller computers to do calculations that were once tackled only by large mainframes and supercomputers. Things that used to take months or weeks now take days or hours, said Lovell. World Wide Web technologies have made it easier for developers to improve existing programs, and an end-user in Australia can download the latest version of a program written in Amsterdam and have it up and running in hours.

Much of the impetus behind these new software programs comes from "density functional theory," which allows properties of molecules — such as structure and bonding — to be expressed in terms of their electron density. This enables computational chemists to better translate raw data on a molecule — the number of atoms and electrons, their bond angles and energies — into a giant numerical equation. Using this, researchers can apply quantum mechanics to try to determine a variety of properties, including the most likely stable structure, or "ground state."

Like a giant Rubik's cube, there is only one correct answer to a problem with an enormous variety of possibilities; a single molecule with just four rotatable bonds searched in 60-degree increments will generate 1,296 separate permutations. And that results from just studying one aspect of a simple molecule; add more atoms or more features and the problem increases exponentially. With the help of a few approximations based on prior experience about the behavior of molecules, software developers (and computational chemists) can reduce the number of variables to get answers in a reasonable time frame.

Lovell's interest is in examining the nature of the bonds that occur in molecules containing two or more metals. The information that he is able to glean from models of relatively simple metallic structures — known as dimers — can be applied to the workings of more complex bioinorganic ones such as those found in photosystem II, essential for the
conversion of water into oxygen in plants. "Using computational chemistry, we can attempt to answer questions like ‘What is the the purpose of this geometric arrangement? Why did nature choose this and not something else?'" he said.

The fast hardware and sophisticated programs will help to answer some of these questions — "with no chance of people like me blowing themselves up," Lovell adds.

— Dan Drollette
REACTION TIMES, American Chemical Society

Λ back to top