Trouble with bubbles precedes the popping.

When people talk about their bubbles bursting, they usually mean their dreams have fizzled. But two scientists have new evidence that could revise these frothy cliches and the science of effervescence.

Bubbles do most of their damage not when they burst but when they billow, according to a new study by You Lung Chen and Jacob Israelachvili at the University of California, Santa Barbara. In the May 24 SCIENCE, they conclude that a bubble's birth involves violent stresses that can deform and destroy nearby surfaces. This counters the conventional explanation of how bubbles cause pitting, or cavitation damage, in turbine blades, pumps, propellers and other materials in contact with moving fluids. Such cavitation causes millions of dollars in damage each year, and engineers may now need to reevaluate their anti-cavitation strategies, says Israelachvili.

The two researchers noticed bubble damage while studying forces between moving surfaces. Israelachvili had invented a device with two mica sheets that he could move toward and away from each other. The instrument allowed him to monitor forces affecting the sheets and to view their effects through a microscope (SN: 4/30/88, p.283).

For their experiments, the scientists sandwiched a thick, slow-moving liquid between the mica sheets, compressed it and then started pulling the sheets back. As the sheets separated, the gooey liquid rushed in to fill the space. Israelachvili and Chen found that if the viscous liquid does not fill the gap fast enough, a tension develops in the mica. Then the bubble forms and the mica snaps back, recoiling so forcefully that damage occurs. All this happens within a millionth of a second, Israelachvili explains. "It's like an earthquake."

Many bubble experts express skepticism over the new report. "My first reaction is that it must be wrong," says cavitation expert Robert J. Etter with the U.S. Navy's David Taylor Research Center in Carderock, Md.

At most, says Lawrence A. Crum, an acoustician at the University of Mississippi in Oxford, "it's something that may occur only in a very limited number of cases."

Andrea Prosperetti, a mechanical engineer at Johns Hopkins University in Baltimore, agrees. "In water," he notes, "you don't have to pull very hard to create a bubble." As a result, no big stresses have a chance to develop in water as they do in the liquid used by the California team. Prosperetti adds that the new data appear to contradict the fact that "you usually find the cavitation damage where the bubbles collapse."