William Blake wrote, "To see a world in a grain of sand … " Now there's an attempt to make those words a reality. At MIT, researchers have developed "smart sand" that's able to self-organize and assemble itself into any object. The idea is that a person could drop a small object into a bucket of smart sand and, after a few minutes, pull up a larger replica made from the particles.

The technique could allow a person to duplicate tools or mechanical parts nearly instantly.

The self-building material, which was developed by Daniela Rus, a professor of computer science and engineering at MIT and her student Kyle Gilpin, isn't the size of sand grains yet. Each "grain" is a cube 10 millimeters (about a third of an inch) long. It has magnets on four sides that can be turned on and off and are used to attract or repel each cube with another. With only a small amount of information, the cubes can arrange themselves into larger objects.

PHOTOS: Extraordinary Beauty of the NanoArt World

WATCH VIDEO: Nanotechnology promises to make our lives better.

The MIT researchers have only built cubes with magnets on four faces, so for now it's only possible to build two-dimensional patterns. (They couldn't fit magnets on the other two.) But eventually, they envision building smaller versions with the magnets on all six faces. One big advantage for this kind of material is that it's reusable, so it lends itself to building things that aren't needed forever — perhaps temporarily replacing machine parts. Once the sculpted object isn't needed anymore, the magnets could be switched off and the grain returned to the "bucket," where they would be used at a later time to build something else.

BLOG: Swarms of Kilobots Are Coming

The computer program, or algorithm, that controls the system is very efficient. In fact, it only needs about 32 kilobytes of code and 2 kilobytes of memory. That's important, since it makes the programming a lot less complicated and reduces the necessary size of the hardware.

Photo: An unfolded smart cube with red wires coiled around the magnets. Credit: MIT/M. Scott Brauer.