The Children of the STM: The Nobel Prize-winning scanning tunneling microscope has inspired a whole generation of imaging devices that use everything from magnetic forces to sound waves to examine samples.

GARY MCCLELLAND drags a sharp probe across a graphite surface and watches as it catches briefly, slips, and then catches again in a regular pattern. It seems like a straightforward-perhaps even dull-demonstration until McClelland, a physicist at IBM's Almaden Research Center in San Jose, California, explains that the slip-stick cycle is not created by some irregularity in the surface. The surface is as smooth as a surface gets. Instead, the probe is responding to a landscape of microscopic hills and valleys created by the surface atoms themselves. It is observing friction at the level of individual atoms. McClelland's friction force microscope is one of a growing family of devices that are changing the way scientists look at matter. Instead of focusing light, x-rays, or electrons on a sample, these scanning probe microscopes map out a surface by "feel," much as a blind man taps his cane to explore the ground in front of him. In many cases, the new microscopes offer better resolution than was previously available, but perhaps more important is their ability to see a sample in new ways. One member of the family, for instance, is sensitive to magnetic fields and can be used to examine the magnetic patterns on a computer hard disk. A second was designed to monitor ion flow into and out of living cells. Others measure electric charge, temperature, optical spectra, and residual stress. The devices have opened a new window on the microscopic world. The patriarch of the family is the scanning tunneling microscope (STM), invented in 1981 by Gerd Binnig and Heinrich Rohrer of IBM's Zurich Research Laboratory. The STM, which won the 1986 Nobel Prize in Physics for its creators, examines a sample by means of a tiny probe that is brought to within a nanometer (a billionth of a meter) of the surface. A voltage applied between the tip of the probe and the sample creates a small tunneling current-a flow of electrons that "leaks" or "tunnels" through the gap between tip and sample-whose magnitude depends sensitively on the width of the gap. As the probe is scanned across the sample, the tip is moved up and down to keep the tunneling current constant. By monitoring