The scientist put what looked like black dust into a dish of water. Instead of dust, however, the particles were actually diodes, capable of emitting light under the right conditions. In the dish sat a cylinder, patterned with tiny dots of solder connected by threadlike lines of solder. The goal of the experiment was to get the tiny electronic devices to land on the solder dots.
“The devices, about half the width of a human hair, were obviously too small to position with our big grubby fingers,” said George Whitesides, the Mallinckrodt Professor of Chemistry at Harvard, who led the experiment. “It might be possible to move them one by one with a microscope and robot, but that would be exceedingly tedious. What we actually did was agitate the water gently, and let nature do its thing. The diodes stuck to the dots in the right position without being touched. The solder lines provided an electric connection between the particles, thus forming a device consisting of 113 light-emitting diodes. It actually worked well. It was amazing.”
This process could find application in new types of electronic displays, but that’s only a passing thought for Whitesides. He and his colleagues are more interested in how nature puts together everything from a living cell to weather patterns without any help from human fingers, brains, and machines.
“It’s a case of learning from nature,” says the Mallinckrodt Professor of Chemistry. “We look at living things and ask ourselves how they do what they do. We try to extract basic principles from this, then use them to assemble useful structures without human intervention. That kind of inquiry works both ways; it yields a better understanding of how to cross the border between chemistry and biology, between life and nonlife.”
Crossing the border
At present, researchers who do self-assembly are working with nonliving or static devices. Whitesides’ team oversaw the autonomous coming together of 1,500 tiny cubes of silicon on a surface smaller than 1 square inch in less than three minutes. In the same building at Harvard, Charles Lieber, Hyman Professor of Chemistry, uses similar techniques to put together devices measured in millionths of an inch, which may find application in tomorrow’s computers and as detectors of disease or bioterrorist toxins.
These static devices, however, have already begun to evolve into structures that closely mimic living things, including proteins, DNA, viruses, and even a somewhat humanlike brain.
Proteins make all of life possible, and they can be made in a laboratory by organic chemistry. Once made, what a protein actually does in, say, heart cells, depends on how these long, stringy molecules fold themselves. This folding is rapid and unseen, out of direct sight and reach of both chemists and biologists, but Whitesides’ group has managed to create a self-assembled electronic device that mimics protein folding. The feat, to be described in an upcoming issue of the Proceedings of the National Academy of Sciences, opens the way to fabricating much more dynamic self-assemblies by doing things the way proteins and living cells do them.
Could scientists actually make a living cell this way? “That’s a long way off,” Whitesides answers. “It might, however, be possible to make a virus.” (Viruses do not live on their own, they become active only when inside cells.) “Biologists have taken pieces of viruses and reassembled them. But chemistry is now sufficiently sophisticated to make all the pieces from scratch.” Watching a virus self-assemble itself should reveal a great deal of basic information about how these disease-causing organisms might be neutralized.
Operating on the border between the living and nonliving has Whitesides thinking about finding ways to connect the two systems. “At present, in our lab and in other labs we are beginning to work on making artificial nerve circuits by self-assembly and finding ways to glue these circuits to living cells,” he says. Such hybrid structures might eventually be used to repair severed or damaged nerves that have left victims without use of their arms or legs.
Building a brain
Self-assembly starts with considering the characteristics of the pieces you want to come together – their size, mass, charge, etc. – and the forces involved in their interaction – electrostatic, electromagnetic, gravitational, capillary, etc. The assembly often takes place on a pattern or template etched into a surface. To make a self-assembled sheet or monolayer, for example, thin, patterned gold films are dipped into a chemical solution. Molecules in the solution then assemble themselves in highly ordered sheets on the patterned metal.
In the case of solder and diodes, where the pieces involved are much larger than molecules, the solder template provides both structural strength and electrical connections. These assemblies are not restricted to flat surfaces. They can be done with shaped surfaces and cylinders, thus opening a path to self-constructed, three-dimensional devices. At this point, the most ambitious challenge would be self-assembly of electronic memories, the parts of computers that store information.
A living brain is also a three-dimensional, self-assembled structure, so it should be possible to mimic it with such an electronic array. “People, including us, have already started thinking about how to do this,” Whitesides notes. “We might even employ a technique similar to that used with the diode and solder assembly.”
The big difficulty in making a faux brain is not so much the number of cells involved as it is the rich network of connections between those cells. A human brain boasts billions of cells with hundreds of billions of connections between them. “Self-assembling such a complex structure is a task that would take 20 years or more,” Whitesides comments.
Meanwhile, he says, scientists have “an infinite amount of amusing stuff to do before we get anywhere close to self-assembly of a living system.” These include improving the capabilities of robots, which today are mostly large, dumb, insensate machines with limited mobility and adaptability. Self-assembly is also the logical way to create structures whose size is measured in nanometers (billions of a meter). Applications of nanotechnolgy include everything from computer circuits to medical sensors that can be swallowed or otherwise placed inside a person’s body (see Feb. 28 and March 21 Gazettes.)
For electronic devices, gold templates and solder wires are fine. But living things require self-assembled proteins that recognize and communicate with each other. Unlike diodes and transistors, living organisms are capable of adaptation; they can change to take on different functions in different situations. “Still, the underlying process of self-assembly is the same for living and nonliving things,” Whitesides maintains. “We don’t have to invent new physical phenomena. We already know it can be done because we’re doing it, and because nature has been doing it for billions of years.”
Living cells self-assemble, and understanding life will therefore require understanding self-assembly.
– George Whitesides