Arresting Reactions

molecules with orange and red background

Imagine the chemist's frustration. She knows the before and after of a chemical reaction: These static states can be plainly observed. But the really interesting part, the dynamics of what actually happens, remain hidden from view. It's just too quick to follow—fait accompli in a matter of femtoseconds.

Much quicker, that is, than a blink. Welford Castleman, Evan Pugh professor of chemistry at Penn State, has a useful analogy: If you can imagine a second to be 25 football fields full of sand, each field piled 3 feet deep, a femtosecond amounts to only a single grain.

Castleman and his students, however, have devised a way to freeze reactions in place, catching molecules in the act. The apparatus they use, a collection of lasers, mirrors, and light amplifiers all leading to a mass spectrometer (an instrument that measures particles by their atomic weight) fills a long table in a first-floor room in Davey Lab. Dan Folmer, Eric Wisniewski, and Sean Hurley, graduate students in Castleman's group, explain how it works:

By guiding two laser beams to collide in the center of a small jet of dark purple dye, Folmer begins, the researchers create ultrafast pulses of very intense light. Each pulse is only 100 femtoseconds long. To the naked eye this rapid fire looks like a steady stream.

The light is then amplified in four stages—compressed and strengthened—until each pulse has almost unimaginable power: enough, for that fraction of an instant, to equal five hundred nuclear power plants. ("It's such a small area," Folmer explains, "that you get a real high power density.")

A small fragment of this beam is then split away and rerouted, creating a lower energy pulse of a different wavelength, which is trained on a vacuum chamber filled with whatever gaseous molecules the researchers want to get a closer look at. This lesser pulse is used to excite the molecules, stimulating a chemical reaction.

Then, femtoseconds later, "we come in with the second pulse," says Wisniewski, banging his fist into his palm for emphasis. This second, powerful blast ionizes the gaseous particles in mid-reaction, creating what physicists know as a Coulomb explosion. Hurley explains: "The particles are positively charged, so they repel each other—that's Coulomb's law. When it happens this fast, it's an explosion."

The molecule is literally blown to pieces; and in effect the reaction is frozen at that instant in time. The remaining fragments, analyzed according to their molecular weight in the mass spectrometer, provide a lasting "image" of precisely what was going on.

The images pile up: The amplified pulse is repeated ten times per second, with each new explosion capturing the reaction (with different molecules) at an infinitessimally later point in time. When it all finally shuts down and cools off, the researchers are left with a record that is akin to a series of snapshots. "It's like a form of animation," Wisniewski says, "where the sequence of events is summed together."

In an experiment designed to demonstrate the technique, the Castleman team successfully used the Coulomb explosion to "watch" the intermediate stages of a proton transfer in a relatively simple molecule called 7-azaindole.

When formed as a dimer, 7-azaindole is a compound made up of two identical halves that are joined together by two hydrogen bonds. When excited, the dimer undergoes a rather subtle change. The two halves exchange a single proton—one for one, an even swap. After the reaction, the molecule looks almost the same as it did before: Each half has the same atomic mass, 118, that it had to start out. But the structure of the molecule is slightly altered.

That much was known already; 7-azaindole is a well-studied system. What wasn't known, however, was exactly how the proton transfer takes place. Is it step by step, with first one proton shifting position, then the other? Or do the two protons swap sides simultaneously? By submitting the dimer to a series of Coulomb explosions, Castleman and students were able to come up with a decisive answer.

The mass spectrum, Folmer reports, revealed some fragments with mass of 119. "What this means is that the reaction happens step by step. There is a brief period of time when one half of the molecule is heavier than the other half. If it had been a concerted reaction, there would never be a time when the halves weren't equal."

In fact, he adds, they were able to pinpoint the reaction even further, conclusively demonstrating that the two proton transfers occur at different rates. The first, their calculations show, happens only 660 femtoseconds after the molecule is excited; the second, completing the reaction, not until 5,000 femtoseconds have elapsed.

After publishing their results in the journal Chemical Physics Letters, Castleman and his students are eager to see what other secrets their technique can reveal. They write: "We anticipate that this method will emerge into a useful way of studying the dynamics of a wide class of reactions."

Daniel E. Folmer (, Eric S. Wisniewski (, and Sean Hurley ( are Ph.D. students in chemistry in the Eberly College of Science, 152 Davey Lab, University Park, PA 16802; 814-863-3583. Their advisor is A. Welford Castleman Jr., Ph.D., Evan Pugh professor of chemistry, 152 Davey Lab; 865-7242; Also involved in this research was postdoctoral student Lutz Poth, Ph.D. The work was funded by the Air Force Office of Scientific Research.

Last Updated January 01, 1999