The New Yorker called on Dartmouth engineering professors Erland Schulson and Donald Perovich to help explain the role of supercooled rain in creating dangerous patches of black ice:

“Warm ice” is the term used by Professor Erland Schulson, of Dartmouth College’s Thayer School of Engineering (Exit 13, Hanover, New Hampshire), for ice that is close to melting. That is the state of “most ice we encounter in the world,” he explained, when I visited his office. Melting ice with a thin layer of water on top of it is as slippery as the natural world gets—“nothing slipperier than that,” the professor said; hard frozen ice is much less slick.

Schulson teaches Thermodynamics and Kinetics in Condensed Phases, which he described as “a course about transformations between states.” Because supercooled rain is a liquid that really should be a solid, it is of particular interest in the classroom. What’s going on at the molecular level that prevents the transformation from occurring as it should?

Schulson, whose research focusses on microstructures in ice, drew a big circle, signifying a raindrop, on a pad of paper he found amid the clutter on his desk. Then he drew little squiggles inside the circle, indicating a “crystalline array” of ice starting to form within.

“Freezing rain exists in liquid form at below zero degrees Celsius,” he said. “So zero can’t be a real freezing point. Better to turn it around and say ice has a melting point, which is zero degrees, no question.”

Schulson explained that as soon as H₂O molecules begin to organize into a nucleus they form a surface between the crystalline structure and the liquid. That surface, like all surfaces, has an energy. But as long as the cluster is too small to be stable and the liquid area around it is large, then, according to thermodynamic law, the area-per-unit energy of the water (that is, the ratio of the surface area around the array to the energy density of the array itself) will be lower than that of the array. And, because inertia favors a lower energy state, the array quickly disappears.

“It’s all about minimizing energy,” added Don Perovich, another Thayer professor, who had joined the conversation. Before coming to Dartmouth, Perovich, who specializes in sea ice, worked at the Army Corps of Engineers’ Cold Regions Research and Engineering Laboratory, also located in Hanover; a former colleague there, Kathleen Jones, researches the problem of freezing rain on power wires, among other things.

Schulson went on, “Even though it seems that with this region”—he pointed to the nucleus—“the energy should be lower than in the randomly organized water molecules around it,” that is not the case until the array reaches a certain size, at which point, because “the surface-energy barrier to nucleation is no longer important,” the liquid becomes solid.

I asked, “Can we say the apparently clear boundary between solid and liquid is not quite as clear when we look closely at what’s going on?” I was remembering “Hamlet”: “O! that this too too solid flesh would melt.”

The scientist corrected my poetic thinking.

“You can’t be half solid and half liquid,” Schulson replied. “It’s one or the other.”

Did the professors have a solution for the problem of black ice on roads?

“To tell you the truth, I try not to be the first person on the road in the morning,” Perovich said. “Because you really can’t see it.”