News from
Office of Communications
Stanhope Hall, Princeton, New Jersey 08544-5264
Telephone 609-258-3601; Fax 609-258-1301

Contact: Steven Schultz (609) 258-5729
Date: November 18, 1999

Scientists Discover How to Make Nanostructures Assemble Themselves

Technique Could Yield New Generation of Miniature Electronics

PRINCETON, N.J. -- Princeton researchers have created ultrasmall plastic structures with a method that is cheaper and more versatile than previous techniques. The discovery has yielded surprising insights into the behavior of materials at very small scales, while spawning many basic research questions. It also could pave the way to a new generation of miniature products, from computer memory chips and video components to devices for sorting DNA molecules.

Professor of electrical engineering Stephen Chou and graduate student Larry Zhuang found that they could coax a flat sheet of plastic resin to assemble itself into a minute, perfectly ordered array of pillars -- with remarkably little specialized equipment. The pillars are a little more than half a micron (a millionth of a meter) in height and width. Viewing one of these pillars next to the head of a pin is like looking at a stack of about 15 quarters next to the dome of the U.S. Capitol. Chou expects that refinements of the technique will yield even smaller structures.

The researchers discovered the technique accidentally while working on another nanofabrication process called imprinting. In that process, also invented by Chou, a pattern, or mask, is pressed into soft plastic polymer, like pressing minute fingers into wet plaster. The researchers were pressing a mask into polymer when dust prevented the two pieces from coming together. Afterwards, when they examined the polymer, they found that it contained a pattern of pillars even though the mask never touched it. Not only had the pillars grown by themselves, they had arranged themselves into a perfectly ordered array.

"It was a very surprising discovery," says Chou. "No one had ever seen such a thing."

Suddenly Chou had an entirely new production technique: simply bring the two pieces close together and let the pattern assemble itself. He dubbed the new approach LISA, for Lithographically Induced Self Assembly. Princeton University has filed for U.S. and international patents on the process, which Chou first described at the International Symposium on Cluster and Nanostructure Interfaces in October. He expects to publish the results in the December issue of the Journal of Vacuum Science and Technology.

The pillar pattern could be perfect for many applications, says Chou. The first use, he says, is likely to be in accelerating the development of organic light-emitting devices, which are the basis for the next generation of flat-panel displays and other video equipment. LISA would allow each dot of light, or pixel, to consist of a cluster of much smaller dots, instead of just being a single block of polymer material. That change would give the displays increased life spans and improved color reproduction, while making construction simpler and cheaper, says Chou.

A longer-term, but potentially further-reaching application is in the design of ultrasmall circuits. LISA could solve the increasingly difficult problem of attaching wires to ever-shrinking electronic devices. "Using the LISA process, you can fabricate your wires first, then it will assemble your devices between the wires on its own," says Chou. One example is the construction of DRAM memory chips for computers. "It's basically a paradigm shift in how you make memory."

Chou also believes LISA could be valuable to biophysicists at Princeton and elsewhere, who have found that passing DNA strands through a minute array of posts provides a cheap and simple way of sorting the molecules by size.

LISA is much better suited to mass production than the most common nanofabrication technique, a slow and expensive process called photolithography, says Chou. So far LISA cannot make features as small as those produced by photolithography, but that may change. One hope is to set up a repeating process where a relatively large mask makes many pillars, which would then be used as masks to make a new set of even smaller pillars. "In principle, you can get smaller and smaller and smaller things," says Chou.

The LISA method complements Chou's earlier discovery of the imprinting technique. Imprinting has become a valuable tool because the arduous photolithography process only needs to be done once in making the mask -- after that, nanostructures can just be stamped out. As a result, the cost is reduced more than 1,000-fold and opportunities arise for mass-produced products. LISA is even simpler: a carefully engineered mask is not necessary. The mask merely defines the outline of the pillar array.

"Here the polymer will form itself, without a mask. It's really wonderful," says Chou. Another advantage is that LISA could work with a wide variety of materials. Currently, Chou is testing LISA in the same material used in Plexiglas, but he believes the technique will work with metals and other non-polymer materials. Imprinting and photolithography work only in plastic polymers.

Despite many follow-up experiments, Chou is not entirely certain of the physics at work in making the pillars rise up toward the mask. In general, he believes it arises from interplay between the electrostatic attraction of the mask and the hydrodynamic instability of the polymer. "There are just tons of new questions you need to ask," says Chou, "But it doesn't surprise us because we know we are working in a regime that no one has tried."

Please Note: Pictures are available at: