Laser chemistry not a snap - but possible.

Laser chemistry not a snap--but possible

When lasers were invented, chemists quickly realized it should be easy to use their special properties to precisely snip apart molecules and drive difficult chemical reactions. It should be easy, but it hasn't been. For the last 25 years the problem has stymied chemists, who have succeeded only in using the laser as a high-tech bunsen burner to heat up molecules rather than cleave them neatly.

Now Princeton (N.J.) University chemist Herschel Rabitz has applied engineering design principles to suggest a way to make the process work. "Once and for all we know the right way of looking at the problem, and it is not impossible--which is what people began to start saying," Rabitz says. The first obstacle to overcome was the presumption that the problem should be straightforward, he says.

A molecule will absorb specific frequencies of light because the chemical bonds between different atoms in it will be excited by specific parts of the electromagnetic spectrum. Chemists quickly grasped the idea of using a laser to hit the molecule with one frequency of light, exciting one specific chemical bond in the molecule until it broke, leaving it open to react with another molecule but leaving the other bonds untouched. "But that bond isn't isolated from the rest of the molecule," Rabatiz says, and other parts of the molecule may bend and vibrate, interfering with the target bond's absorption of the laser's energy.

Rabitz' solution involves using "optimal design theory" to craft a laser pulse to do the job. Rabitz proposes zapping a chemical solution with a short (10.sup.-15 second) pulse of light, and feeding the molecules' reaction to the light into a computer, which would decide how to change the next pulse to do the job better. The computer might change the wavelength, polarity or length of the laser pulse. Once this system settled on the best pulse design, chemists would drive the actual reaction with just the laser and without the computer's help.

Rabitz and others are now attempting to make his approach work in the laboratory, although it may take years to solve the technical problems. Once the technique is developed it could be a boon to synthetic chemists, who currently take great paints to protect one part of a molecule while changing another part in a chemical reaction. The technique also "opens up the prospect of studying molecules in ways that are not possible now," Rabitz says.