Content area
Full text
"Quantum mechanics," so the saying goes, "is not just a good idea; it's the law!" And among the most famous items in that scientific code is the Heisenberg uncertainty principle, which holds that any measurement of a quantum mechanical system, such as a light wave or an atom, will disturb the system in an unpredictable manner. The more precise the measurement, the greater the disturbance. In other words, says Jeff Kimble, a physicist at the California Institute of Technology, "if you open up a hole to look at the state of a quantum system, the same hole that lets information out, lets fluctuations in."
The result is an intrinsic barrier of fuzziness to our knowledge of, say, the amplitude or phase of a light wave. In the past few years, however, Kimble and other researchers in a quartet of collaborations have managed to show that the Heisenberg uncertainty principle is a little like the tax code: It cannot be broken with impunity, but it has loopholes that--with sufficient ingenuity--can be profitably exploited. The result is a series of experiments that use sophisticated optical techniques to extract information from a quantum mechanical system without disturbing the variable being measured. This quantum sleight of hand opens the way to measurements so precise, says Philippe Grangier of the French Institut d'Optique Theorique et Appliquee, that they can reveal the fundamental "graininess" of light: "the fluctuations in intensity caused by its photon nature."
That ability to sneak around the limit of accuracy set by the uncertainty principle might be valuable for fundamental measurements that require detecting signals so weak and transitory that they push the limits of quantum mechanical precision. And it's the first step toward a more distant goal in fundamental physics: demonstrating the concept of quantum nondemolition, or QND, a term coined by Moscow State University physicist Vladimir Braginsky. In theory, a measurement that exceeds the quantum limit and does so without introducing noise into the signal--leaving it "undemolished"--qualifies as QND.
To prove the concept, however, physicists will have to make a pair of measurements of the same system. The first measures the variable of interest, and the second remeasures it, showing that the variable hasn't been disturbed by the initial measurement. Making these measurements once has...