Before the advent of his "scanning-polarization-force microscope," or SPFM, images of liquid surfaces were limited to what one could see with an optical microscope. Using the SPFM, Salmeron and his colleagues have seen strings of potassium hydroxide droplets persist along graphite ledges only a few atoms high, even after the graphite has apparently been blown "dry." They have seen fresh-cleaved sheets of mica, exposed to air, exhibit increasingly branching patterns of wetting. They have observed ice crystals coated with a layer of liquid water, even at temperatures below freezing.

In conventional atomic-force microscopes, the tip of a sharp probe scans across a solid substrate in contact with the surface. "If you try that with liquids, the tip sinks in, or the liquid wets it and climbs up, distorting the surface you're trying to study," Salmeron says.

His trick was to lift the probe several tens of angstroms. A small voltage difference is established between the probe and the substrate so that electrostatic forces can be detected and controlled to a constant value; this produces a profile of constant force as the tip scans over the surface.

"There is a penalty for this—the lateral resolution is not better than the distance between the probe and the surface. Still, that's only a few hundred angstroms, far better resolution than an optical microscope, which is typically a few micrometers. And the height resolution, under an angstrom, is as good as any other scanning-probe microscope."

The probe of the SPFM is a fiber of silicon nitride whose tip is thinly coated with platinum. A laser beam, focused on the back of the flexible fiber, is reflected to a detector diode, which records the cantilever's dips and bends as the tip flies over the surface of the liquid.

By using insulating samples, the counterelectrode can be located far from the probe, for example under a glass slide. Two factors influence the image: one is the distance of the tip from the surface of the liquid. The other is the polarizability of the liquid itself.

Salmeron and his coworkers have used the SPFM to make spectacular images of sulfuric acid corroding an aluminum surface. The reaction of acid and metal creates a salt, which in dry air precipitates on the uncorroded aluminum, resembling a basaltic flow on the moon; within this moonlike mare (sea), liquid sulfuric acid fills deep craters. Both the liquid drops and the underlying craters are visible in successive, registered images.

To make these images the microscope was first used in scanning-polarization-force mode, showing liquid domes over flat "pancakes" of liquid. Then the probe tip was allowed to descend to the surface in atomic-force mode, revealing that under the standing drops, deep pits penetrated the salt to uncorroded aluminum below. Such unsuspected details of corrosive reactions may shed light on the acid-rain corrosion of airplane parts and other aluminum structures.