Understanding molecules, he's quick to point out, is a prerequisite for understanding just about anything else. "Look at the world," says Zewail. "Everything around us is chemical reactions. Everything--inside you and me, the atmosphere, everything we breathe, we touch. Everything is a chemical reaction. So we have to develop a unified theory of how chemistry takes place. We can't understand this unless we really have a coherent understanding of how atoms and molecules like or dislike each other. That's our ultimate goal."
The catch is that these intimate chemical acts occur on a time scale that is almost unimaginably short: they take place in femtoseconds, to be precise, which are the quadrillionth parts of a second. A quadrillionth equals a thousand-trillionth, or .000000000000001; for perspective, you might contemplate that a femtosecond is to a second as a second is to 32 million years. To capture such acts as they happen requires having a strobe light fast enough to illuminate the action before it's over, which begins to explain the reason for the lasers down in Femtoland.
Until Zewail came along, chemists had been unable to watch the making and breaking of molecules because they didn't have a flashbulb capable of breaking time into femtosecond chunks--the scale of the reactions themselves. They had to settle for studying the before and after; the "during" was a blur. This situation, as one Swedish chemist put it, was analogous to trying to understand Hamlet while seeing only the introduction of the characters followed immediately by the closing scene, which is to say a stage set with "a considerable number of dead bodies and a few survivors." Zewail managed to capture the mysterious acts in between, known in the lingo as transition states. These states mark the adjustments that must be made before two atoms that have lived as separate entities can be joined in a coupled pair that behaves as one--the courtship, if you will, between meeting and marriage--or conversely, the trauma that occupies the space between marriage and divorce. Zewail calls these states "the configurations of no return."
Take, for example, sodium iodide, which is a simple salt not unlike table salt. "Fascinating problem," says Zewail, explaining that sodium iodide is an innocuous substance, but break a single chemical bond and you have sodium and iodine, both of which can be extremely hazardous to your health. "It's a beautiful example of how one chemical bond can change behavior," he says.
Now imagine trying to watch sodium iodide disintegrate into its component atoms with a strobe that can freeze action only every trillionth of a second. First you see the sodium and iodine atoms nuzzle together in a molecule. And then, a trillionth of a second later, they're apart. All you see is an iodine atom and a sodium atom going their separate ways, moving through space as though they'd never been one. What happened in between has happened in less than a trillionth of a second and so remains a mystery.
Now do the experiment using the requisite tool of the practicing femtochemist, which is a laser that fires femtosecond pulses every ten femtoseconds or so. The experiment starts with a laser pulse--zap!--and the iodine and sodium start splitting up. Ten femtoseconds later--zap!--another laser pulse comes in and the scene is illuminated, revealing the iodine and sodium moving back together. Another ten femtoseconds--zap!--they're drifting apart. Ten more--zap!--they're drifting back together. And so they go, says Zewail, "back and forth 12 times. Exactly."
As it turns out, the sodium and iodine atoms look like two balls on either end of a faulty spring, moving together and apart, together and apart, until finally the spring breaks and the balls part for good. As Zewail puts it, "they love each other for 12 cycles"--the number is determined by the force between the atoms--"before they can finally break the bond." But without the femtosecond laser, you'd never see it. The fundamentals of the chemical reaction would be the stuff of theory and little more.
The femtosecond laser was the advance that made femtochemistry possible. As physicist Mark Rosker, who as a postdoc helped Zewail assemble his first femtochemical experiment, puts it, "Once you had the light, everything else was simple."
The light, however, was anything but simple. For starters, Zewail explains, electronic circuits simply could not be built that would turn anything on and off faster than every hundredth of a billionth of a second or so; they were limited by the time it would take an electric current to travel through even the shortest wire possible, as well as by the response time of the semiconductor materials from which switches are made. That meant the creation of a femtosecond pulse had to somehow exploit the nature of something faster than electronics--namely, light itself. The solution involved a technique called mode-locking, which was put to work at the femtosecond level by physicists at Bell Laboratories in the early 1980s.
To understand mode-locking, suggests Zewail, start with a light bulb. The light emitted is composed of electromagnetic waves, all with random wavelengths and out of phase, meaning that they're oscillating up and down out of step with one another. The result is the familiar beam of white light. A laser beam, in contrast, generates all the light waves at virtually a single wavelength, or color, and all in phase, which is to say they're "coherent," all oscillating in more or less perfect step.
But an ordinary laser still produces a continuous electromagnetic wave. What Zewail needed was one that could shoot coherent chunks of light, each one passing by in the blink of a femtosecond. Think of it as the pursuit of the ultimate strobe. To achieve its goal, the femtosecond laser produces coherence in a somewhat backward manner. Rather than generating light at just one wavelength, it generates as many as 10,000 different wavelengths (although the resulting beam might still look red or green to the human eye). These 10,000 wavelengths are equally spaced, like steps on a ladder. They are out of phase 99.9996 percent of the time, their peaks and troughs going up and down at slightly different times and at slightly different places, effectively canceling one another out. But that other .0004 percent of the time they line up exactly in phase, with all the different peaks and troughs moving up and down together. And for that brief instant a pulse of coherent laser light is created. It is as if a superfast shutter inside the laser opens to allow that single pulse to pass through, then opens again a short time later to spit out another one.
What comes out is the femtosecond pulse, a packet of light shaped like an exquisitely thin pancake, a vertical slice of light. If you could freeze the pulse in midair, says Rosker, and hold a tape measure up to it, you'd find it to be 15 microns (15 millionths of a meter) thick and a couple of thousand microns (a few millimeters) in diameter.
For the purpose of femtochemistry, this pulse is now split in two. The first, or "pump," pulse is used to initiate whatever chemical reaction is being studied. It wouldn't be practical to sit around waiting for the reaction to happen; how would you ever catch it in action? Instead the femtochemist deliberately starts (or pumps) the reaction with the first pulse. The second, or "probe," pulse is delayed a few femtoseconds by bending it around with mirrors and prisms so that it has to travel a micron farther than the pump pulse. (Adding an extra micron to the path of this second pulse delays it 3.3 femtoseconds. Adding more microns delays the pulse further.) The probe is the strobe light that illuminates the reaction at any desired time after the reaction begins.
What can you actually see with such a finely honed sliver of light? The secret to femtochemistry turns on the fact of nature that each species of molecule or atom absorbs and radiates light at precise wavelengths--a sodium atom, for example, absorbs and emits one particular wavelength of yellowish light. If atoms happen to be in the process of reacting--cleaving together into a molecule or breaking away from the molecule--their effect upon one another at each step along the way will make the molecule absorb infinitesimally different wavelengths. The iodine atom approaching the sodium, for example, will disturb the sodium's electrons somewhat; how much will depend on the distance between the two atoms.
In other words, not only does each molecule have its own spectral signature, but that signature will vary subtly depending on what the molecule happens to be doing at the time. And after a molecule absorbs its wavelengths of choice, it then reradiates, or fluoresces, its characteristic wavelengths in all directions. A set of lenses placed out of the way of the incoming laser can now be arranged to detect this radiation.
Thus for each of the two pulses, an experimenter has to determine the precise wavelengths of light needed--first for the pump, to start the reaction, and then for the probe, to capture what's happening a few femtoseconds later. To catch a chemical reaction in the act, the femtochemist tunes the probe pulse to a wavelength that will be absorbed only by a given molecule at one particular stage of a reaction, and not by any other molecule or by the same molecule at any other stage in the reaction. By shooting in probe pulses at incremental time delays after the pump pulse starts the reaction, and then watching to see when the target molecule reradiates that energy, the femtochemist can clock exactly how long the reaction takes to reach that particular stage.
Zewail's very first set of experiments was a nifty example of how the whole system works. Zewail wanted to analyze the disintegration of the simple molecule cyanogen iodide, which is an iodine atom bonded to a cyanide molecule. He tuned his pump pulse until it had a wavelength of exactly 306 nanometers (or 306 billionths of a meter). This happens to be in the ultraviolet range of the spectrum, and it is the energy that will readily break the bond between the iodine and the cyanide.
The pump pulse was fired into a gas chamber holding the cyanogen iodide molecules--zap!--and the molecules started to come apart at the seams. The pump started the clock on the reaction; t was now equal to zero, and counting.
Then came the probe. The first thing Zewail wanted to do was measure how long the reaction took in its entirety, which he did by choosing a wavelength for the probe--exactly 388.9 nanometers, also in the ultraviolet, although not so far ultraviolet as the pump pulse--that is readily absorbed by cyanide only when it is sitting alone in space, isolated from any other molecules or free-floating atoms.
Zewail started the reaction, then snapped femtophotos at 10- femtosecond intervals simply by delaying the probe pulse. Every 10 femtoseconds he checked for fluorescence at 388.9 nanometers. When he got it he knew how long the reaction took from beginning (cyanogen iodide) to end (iodine and cyanide). The answer turned out to be 200 femtoseconds. And Zewail happened to be the first person ever to measure this fundamental bond-breaking interval directly.
After establishing the lifetime of the entire reaction, his next step was to change the wavelength of the probe to time various steps in the process of the molecule's disintegration. From there it was almost mundane: shoot in the pulses--now, say, at 390.5 nanometers, a slightly redder ultraviolet, corresponding to an energy cyanide will absorb when it's still just close enough to the iodine to feel its presence--and watch for the fluorescence. Retune the probe pulse, make it a hair redder still, shoot in the pulses, time the fluorescence, and so on. By changing the wavelength and the time delay of the probe pulse, Zewail was actually watching the unraveling of the molecule, distinguishing motions as small as a tenth of a billionth of a meter or less. For the first time ever, a chemist had watched the breaking of an elementary chemical bond as it happened.
Since Zewail first captured the dynamics of a chemical reaction on its femtosecond time scale, he and his colleagues have studied some 50 molecular reactions and are branching out into different fields. He now has five laboratories in his subbasement Femtoland, each devoted to a different line of femtochemistry. In these labs researchers are studying everything from how elementary chemical reactions are altered in complex environments- -by creating a shell of water, for instance, around their molecules before zapping the systems with laser pulses--to how the tangle of forces at play between the many atoms of a complex molecule affects the dynamics of the otherwise simple breaking and bonding reactions.
Meanwhile femtochemistry has exploded in academic research labs across the country, where hundreds of chemists are now using the technique to study such problems as the molecular mechanics of human vision and the dynamics of proteins. Much of this has been made possible by the recent availability of relatively inexpensive titanium-sapphire lasers that researchers can buy nearly off the shelf, and which have made using femtosecond lasers "as easy as turning on a light switch," according to chemist Nancy Levinger of Colorado State University. Titanium-sapphire lasers produce pulses in the 800-nanometer-wavelength range that matches the near infrared so readily absorbed by biological systems. For the first time, researchers are getting a real-time glimpse of such fundamental processes as photosynthesis. "Nature can turn light energy into chemical energy at almost 100 percent efficiency," says Levinger. "It's easy to see why so many people are interested in understanding how that works."
And now chemists are learning how to use femtochemistry to control the yield of molecular reactions by getting their femtosecond lasers to turn reactions on and off at will. In 1992, Zewail and his team announced that they had managed for the first time to pull off the femtosecond laser control of a chemical reaction, this time involving iodine and xenon--put the two together and the result is xenon iodide. Once again Zewail used two pulses: the pump pulse to initiate the reaction, and a control pulse, as he calls it, to give the molecules either a final nudge to complete the reaction or a rude elbow to stop the reaction entirely. "The motion of the nuclei is taking place," says Zewail, "such that we can say if we delay the pulses by so and so a time, we have 100 percent yield of xenon iodide, and if we delay by such and such, we get 5 percent yield.
"This is just the beginning," Zewail adds. "If we can show that we can do this with the kind of complex molecules the chemical industry uses, then the impact will be enormous." And femtochemistry will, in less than an eye blink, come out of the subbasement and into the light.