If there were a Nobel Prize for the best scientific party trick, this would be a contender. Take two tin cans, one with the top removed and the other with both ends cut out to form a tube. Turn the closed-bottomed can upside down and push both cans into dense sand. The closed-bottomed can will sink more easily than the open tube. That's weird because if you push the two cans into water, exactly the opposite will happen.
Raphaël Clément and another graduate student at the Université Paris Diderot in France discovered the effect while playing around in the lab with empty tea canisters. "It works quite well with an upside-down glass, too," Clément says. "You can try it at home." The effect highlights the subtle and often surprising physics of granular materials. Physicists have fundamental theories of solids, liquids, gases, and even ionized gases called plasmas, but they have no such theory to predict the behavior of a common dirt pile.
To see why the effect is surprising, first imagine pushing the inverted closed-bottomed can and the tube into water. Naturally, the open-ended tube slips right in. The closed-ended can, however, traps air that creates a huge upward buoyant force as the air is forced down into the water. Trying to submerge the closed can is like trying to dunk a balloon.
Press the tube and can into dense sand, and the picture is completely different. To prove it, Clément, his adviser Stéphane Douady, and colleagues suspended aluminum cylinders 7 centimeters wide, 12 centimeters tall, and 2.1 millimeters thick—one with a lid at one end—just above a bed of building sand. They then weighted the cylinders with "loads" ranging from 200 grams to more than 4 kilograms and dropped them into the sand. For the same weight, the close-ended cylinder descended significantly farther than the open-ended one did, the researchers report in a paper in press at Physical Review Letters.
That's because the air trapped in the closed-ended cylinder affects the sand in a dramatic way, the researchers argue. Ordinarily, still sand acts somewhat like a solid. But if the air around the sand grains moves quickly enough, the sand will flow like a fluid, as physicists and engineers have long known. And when the closed-ended cylinder sinks into the sand, the air trapped in it rushes out from beneath the cylinder's rim, fluidizing the sand at that point and making it easier for the stuff to get out of the can's way. The sand has to be relatively densely packed, and the can must drop quickly. If the researchers slowly added weight to the can, the air seeped away gradually and the can descended no farther than the tube.
The researchers were probably lucky to stumble onto the effect, as it likely works only for certain grain sizes and cylinder sizes, says Stephan Koehler, a physicist at Worcester Polytechnic Institute in Massachusetts. "I think they really hit the sweet spot in a huge parameter space," he says. Troy Shinbrot, a physicist at Rutgers University in New Brunswick, New Jersey, says he wishes Douady and colleagues had explored exactly how the effect depends on such factors. Still, he says: "It's a cool effect. I might use it in one of my classes to show the difference between granular materials and fluids or solids."