Computer envisions bio-inspired catalysts.

In their search for ever better industrial catalysts, researchers increasingly seek inspiration from Motor Nature. By studying how enzymes selectively foster and control important chemical reactions, catalyst designers hoep to one day mimic catalytic function with far simpler and more rugged materials.

This week, scientists at Sandia National Laboratories in Albuquerque, N.M., reported a major milestone along the road to such materials: the ability to predict reliabily the precise, three-dimensional shape of theoretical and potentially desirable biologically inspired catalysts.

Catalysts do not actively participate in chemical reactions; instead they offer a temporary docking site for one or more specific compounds. The more precisely a catalyst's pocket-line docking site matches the shape of a desired guest molecule, the more likely such intended molecules will dock and react.

Pharmaceutical chemists have long designed drugs to fit into specific pocket-like sites on selected proteins. "We turned that idea around," says physicist John A. Shelnutt, who leads Sandia's catalyst-design effort. "As far as I know, we're the only people who are using molecular modeling to design catalysts with pockets that fit given target molecules."

At the American Chemical Society's national meeting in San Francisco, Shelnutt delivered a series of papers on his team's work with metalloporphyrins, a class of compounds that serve as the catalytic centers for cytochrome P-450 proteins. These proteins play an important metabollic role in nearly all cells.

From their study of existing porphyrins, the researchers gleaned a series of rules that characterizes how combinations of atoms in this family of molecules typically bend, stretch or become rigin in response to their internal associations -- bonds, repellencies and long-distance, unbounded attractions. They then fed these "force field parameters" into a computer and instructed it to apply this code of conduct to the novel structures they began designing.

Once the team defined a new porphyrin's composition, the computer predicted the hypothetical molecule's preferred configuration. Normally, a porphyrin is flat. But these predictions indicated that as the scientists began replacing hydrogen atoms at the periphery of the molecule with relatively large and bulky subgroups -- such as phenyl rings -- the entire structure would ruffle. Some versions contorted so severely that a deep pocket formed around the metal atom at the center of the catalyst.

One catalyst formed a pocket "just the right size to hold carbon dioxide -- and not something else," notes Shelnutt. By altering electric charges on the pocket's lining, "We can switch the catalyst's potential affinity from carbon dioxide to other small molecules like methane."

How good were those predictions? Shelnutt collaborated with Kevin M. Smith at the University of California, Davis, to synthesize samples of his newly designed porphyrins. X-ray crystallography and Raman spectroscopy have just confirmed that the actual compounds match his predictions. Indeed, Shelnutt's predicted structures are "embarassingly accurate," says Sandia chemist Alan P. Sylwester -- so close, he jokes, that they look as if they were traced from structural diagrams of the synthesized compound.

Shelnutt's team will continue working toward catalysts that might make something useful out of some of the uncontrolled carbon dioxide that poses a serious threat of global warning. The researchers are also designing a solar-driven process to detoxify chemically contaminated water. In preliminary tests, another novel porphyrin catalyst appears more than 100 times faster than the titanium-dioxide catalyst currently used in a similar solar decontamination system now under development.