We still don't understand icicles
A classic symbol of winter's chill, rows of icicles hanging dramatically off roofs and trees show off the simple artistry of nature. But there's more to them than their fleeting beauty: Icicles are one of the unsolved mysteries of physics.
"Despite seeing them all the time, icicles are actually poorly understood," said Stephen W Morris, a physicist at the University of Toronto who has been studying their shapes and ripples since 2007.
His recent research focuses solely on the ripples: No matter how big an icicle is, the hallmark ripples or ribs that form along its sides always have the same wavelength, or distance between one peak to the next - about a centimetre between neighbouring bumps. But no one knows why.
Some theorists have linked this regularity to surface tension between the thin film of water flowing on the surface of the icicle and the surrounding air. But Morris has found that a key factor is something much simpler: salt.
"What we discovered is an extremely strange fact: You need a small concentration of salt to produce ripples," he said. For some reason, the periodic ribbing has to do with impurities in the water. In his experiments, icicles made with extremely pure water lacked ripples. Even tap water contains enough salt to create the pattern.
Runoff from melted snow contains salts such as calcium or sodium - much less than is found in tap water, but enough to create ripples - picked up from rooftops or air pollution. Morris' lab melted free-growing icicles taken off a garage to test their salt levels and found that they were within the right range of saltiness.
In nature, icy spikes form when accumulated snow or ice melts in direct sunlight or through contact with a warmer surface, such as the roof of a heated house. The resulting water drips off, refreezing when it reaches a pocket of cooler air and forming an icy column that builds up over time.
Also, there's a temperature balance involved: The weather needs to be just right for icicles to form. Too cold, and everything turns to solid ice; too hot, and the dripping water won't have the chance to refreeze.
With the help of his graduate student Antony Szu-Han Chen, Morris has grown hundreds of icy spikes with a homemade icicle maker. Water drips slowly onto a sharpened wooden support suspended inside an insulated, refrigerated box; the support is slowly rotated to encourage symmetry. The scientists carefully control the water composition and air temperature, and even mimic wind with little fans.
Growing your own icicles isn't speedy: With the box set at 14 degrees, it takes about eight hours to make a 50cm-long tapered spike.
A digital camera takes photos of the icicle as it spins so that the topography of its silhouette can be analysed using special computer software.
Morris' first foray into icicle research focused on the overall tapered cone shape rather than the detailed ribbed features. He took cues from physicist Raymond Goldstein's lab at the University of Arizona. Goldstein and his colleagues came up with a theoretical model that explains the shape of a growing icicle based on his previous work on stalactites, the spiky mineral deposits that hang from cave ceilings.
"The physics of stalactite formation is very different from icicle formation," said Goldstein, now at the University of Cambridge. Stalactites grow through calcium carbonate deposition, while icicles bulk up in areas where their thin film of water freezes.
Nonetheless, the same mathematics applies. Ripples on stalactites have the same wavelength as their icy counterparts.
"An individual icicle has bumps and wiggles and imperfections, but if you average over many, it is consistent with this theory," he said. However, Goldstein's model of icicles does not take ripples or salt content into account.
Even though Morris and Chen have found a crucial piece of the puzzle - the role of salt in icicle ribbing - mysteries abound. They still don't know why, for example, changing air temperature or salt concentration has no effect on the ripple wavelength or why the wavelength size is always about a centimetre.
One thing that does depend on the saltiness is how the ripples morph as the icicle grows. For small amounts of salt, ice buildup tends to favour the top curve of the ripple closer to the icicle's stem. No ice actually moves, but the ripple appears to shift upward as the icicles get bigger. So over time, the ripples seem to move slowly up the length of the icicle.
Saltiness is certainly a factor with ripples, but why? No one knows. Morris speculates that it could have to do with a layer of "spongy ice" between the thin film of water and the solid ice. Spongy ice is a mix of ice and water that makes the surface microscopically rough, and sponginess very much depends on how salty the water is.
For now, Morris has looked only into the effects of table salt, but additional materials - other kinds of salt, different minerals, even soap - are on deck for testing.
"Even a very simple-looking thing is kind of the thread you pull on, and it unravels a whole lot of complex things," Morris said. "It's quite a chase."
Goldstein agrees: "The only thing that mattered to me was the aesthetics of it; I was simply fascinated by the beautiful shapes.
"The only driving force was the beauty of what we found in nature."
Kim is a freelance science journalist based in Philadelphia.
- Washington Post