The system is rigged
Study shows how quantum world works in our favor
By Steven Schultz
Princeton NJ -- For scientists who work in the realm of the very, very small, Princeton chemist Herschel Rabitz is a bearer of good news.
First he invented a method for manipulating individual atoms and molecules in ways that had previously seemed impossible. Now Rabitz and colleagues have discovered a surprising principle of controlled quantum mechanics that explains their success and opens the door for many more advances, from the creation of new chemicals to the remote detection of bioterrorism compounds.
Scientists have tried since the 1960s to use lasers to control the behavior of individual atoms and molecules -- attempting to cut and join them and make them emit particular frequencies of light. Until recently, that quest seemed too complex; not even the fastest computer could calculate the movement of each electron, the jiggle of each atomic bond. In the last 12 years, Rabitz and colleagues have demonstrated what amounts to a shortcut in which the laser does its job without complex calculations. Logic still suggested, however, that the shortcut was too good to be true and must work only in special cases.
In the March 26 issue of Science, Rabitz and graduate student Michael Hsieh along with Drexel University Professor Carey Rosenthal have shown that the shortcut is not a fluke. It draws on a fundamental property of quantum mechanics and assures that any control procedure meeting a few basic criteria is mathematically guaranteed to succeed.
"This is really looking very exciting," said Marlan Scully, a physicist at Texas A&M University, who is using Rabitz's technique to develop a laser that could detect anthrax spores in a cloud a mile away. Because of this work, Scully said, many more scientists are likely to pursue projects that previously seemed impossible.
At very small scales, the classical rules of physics give way to quantum mechanics. The reason is that every object, whether a person or electron, can be viewed not just as a fixed particle, but also as a wave that oscillates at a certain frequency. The wavelength of this oscillation is extremely small, so for large objects the wave properties are meaningless. For small particles, however, the wavelength can be bigger than the particle itself. Electrons, for example, cannot be pinned down to an exact location; they exist as a blur somewhere within the range of their wavelength. This small-scale behavior, known as quantum mechanics, is notoriously hard to control.
Rabitz found a way to control quantum behavior by tuning ultra-short laser pulses to contain just the right frequencies of light and locking the pulses in the proper sequence so they resonate with the natural wavelengths of atoms and guide them to a new configuration. Calculating the best sequence of frequencies would be too complicated, so Rabitz starts with random mixes of frequencies and checks whether one works better than another. A computer then suggests many thousands of variations, letting the molecule and laser work out the precise sequence of frequencies that will achieve the desired result. The whole process may take just a few minutes or even seconds.
In recent years, researchers have used the technique to make a more efficient source of X-rays and to manipulate photosynthesis in bacteria. In 2001, Rabitz reported that he used a laser to snip a piece out of a complex molecule and stitch the remaining parts back together. The method worked so well that Rabitz began to wonder why quantum behavior that appeared so daunting was so readily tamed. It seemed improbable that the computer guiding the experiments would always find the optimum laser pulses when the number of possible adjustments is so great it would take more than a lifetime to run through them all.
"If people had sat down and thought about it 15 years ago, the rational conclusion would have been, 'Don't try this,'" Rabitz said.
In their new paper, Rabitz and colleagues showed that all quantum systems have a general property that essentially stacks the odds in the researchers' favor. They found that from any starting point, there is a guaranteed route to the best possible solution. Rabitz likened the computer's search for the right laser pulse to a lost hiker trying to find the peak of a mountain; so long as the terrain permits the hiker to keep heading uphill, the hiker is bound to reach the top. Rabitz showed that the "quantum landscape" contains no false peaks -- all slopes lead to the top, or best solution. So long as the computer keeps guiding the experiments toward improved laser pulses, it is guaranteed to find an optimum series of pulses that achieve the desired result.
"You are always on a slope and any reasonable algorithm will guide you to the top," Rabitz said.
Scully said the result represents a major shift in thinking about what scientists can expect to achieve with controlling quantum behavior. "The explanation is beautiful, a wonderful advance of the theory," he said.
Oxford University physicist Ian Walmsley said the assurance that good solutions are attainable is likely to prompt many more researchers to attempt what had seemed like a needle-in-the-haystack search. "You might be hesitant to go rummaging around in the haystack if you thought it was only a rusty old pin in there," said Walmsley. "To know the golden needle exists is an important enticement."
Rabitz is following up the research with further analysis of the quantum landscape as well as studies of particular applications, such as the develop- ment of computers that harness quan-tum mechanics to perform faster calculations.