Fluid dynamics researchers Jacy C. Bird and William D. Ristenpart were doing routine experiments on water droplets, when the supposedly predictable outcome failed to materialize. Instead of coalescing as expected, the charged water drops they were testing began repelling each other—a result that seemed to fly in the face of elementary science.

Last week, they published their accidental discovery that oppositely charged water drops will not stay together in strong electric fields. And the authors say their finding sheds new light on how oil companies, which must condense water droplets as part of the refining process, can operate more efficiently.

As part of the refining process, oil companies use electric fields to separate the oil and water both present when oil is drilled. That process of electrocoalescence, an amped-up version of what naturally occurs with oil and vinegar in salad dressing, uses electric fields to make oppositely charged water drops coalesce with one other, which helps fully separate them from the oil.

But Bird said the findings suggest this process will only work up to a certain voltage.

“What’s new is that above a critical electric field, the water drops show a direct contact and repel—against what is believed [will happen],” said Bird, a Harvard doctoral student in engineering. Instead of infinitely increasing the coalescing speed of water, an electric field above the critical level causes the drops to deform, combine briefly, and then repel, never actually coalescing.

The explanation for this new find, according to Bird, lies in the geometry of electrically charged water drops. While normal, uncharged water drops have round curves, charged ones have parabolic curves. So while normal drops repel or coalesce upon contact depending on the size of the drop and speed of collision, the unusual shape of charged drops causes them to come together briefly then separate.

“The electric field is bringing the drops together until they contact, and then it’s just geometry from there on—nothing to do with electric fields,” said Bird.

The implications of these findings are being investigated by Ristenpart, one of the original researchers at Harvard who has since become a professor of chemical engineering at the University of California, Davis.

“Right now, we’re trying to understand what is going on fundamentally,” said Ristenpart. “This is a very counterintuitive mechanism taking place. There’s other situations as well, like sometimes a smaller daughter droplet is emitted.”

Ristenpart said he believes that the practical applications will extend beyond the petroleum industry.

“I’m hopeful that this new observation might stem new experimental work on how charge is conducted from storm clouds,” he said. “Also, sources of biofuels have to do with much wetter systems, so potentially you can have more water-based systems too.”