The Physics of Sand

How do you predict the behavior of millions of sand particles? A scientist has come up with a solution.

Pick up a handful of sand, toss it in the air, then try to predict where each grain will land.

Obviously, the task is beyond the ability of most people to compute, but you’d expect a physicist with a state-of-the-art supercomputer to be able to figure it out. Actually, she too may find herself stymied.

Like rocks, snowflakes, coffee grinds and grains of rice, sand is an example of a granular material — discrete solid, macroscopic particles often found together in large numbers.

We know how one unit of a granular material will behave. Throw a ball, for example, and with the right information, a scientist can predict how far it will go and where it will land. It’s basic Newtonian physics.

But when lots of units of a granular material are in motion, they exert so many forces on each other that even our most powerful computer can’t predict the dynamics with sufficient accuracy.

When granular materials move, they often behave like liquids. Think of sand flowing through an hourglass. But at other times, they become jammed and behave like a solid. Think of salt particles coming out of a shaker. If there are enough particles trying to get through a hole at the same time, none will make it through. That’s when they stop moving and become jammed. This can happen as well when marbles try to get through the spout of a funnel. One or two may pass, but when several make the attempt at the same time, they get stuck.

In the late 1980s, the British physicist Samuel Edwards proposed the radical idea that all jammed states — and there might be trillions — were equally possible.

In June, Enid and Nate Ancell Professor of Physics Bulbul Chakraborty, postdoctoral fellow Kabir Ramola and several colleagues at the University of Cambridge in England published a study in Nature Physics that put Edwards’ conjecture to the test through computational modeling. The researchers focused on what happens when, right after the jamming occurs, the granular material is unsettled. They found that when it resettles, there is no particular pattern of the particles that can be predicted, thus proving that Edwards was correct.

It’s the first computational confirmation of Edwards’ conjecture and represents a major breakthrough in our ability to model the arrangement of granular materials, even if that means determining the arrangement can’t be predicted.

The research also opens up avenues for analyzing and understanding granular flows such as the ones we see in avalanches, which are caused when jammed states of rocks and debris become unjammed.

It may also one day prove useful to businesses that use granular materials such as the drug and food industries, agriculture and oil companies.

The other authors on the paper are Stefano Martiniani, K. Julian Schrenk and Daan Frenkel, all at the University of Cambridge.

Categories: Science and Technology

Return to the BrandeisNOW homepage