Polymer blend takes on printed pattern.

Computer chips can be fabricated so easily now, they practically manufacture themselves. In a new study, researchers show they're a step closer to making that a literal statement.

A group at Konstanz University in Germany has developed a quick and simple way to deposit two polymers in an intricate pattern onto a gold wafer. In this technique, described in the Feb. 26 Nature, the two polymers arrange themselves on the gold's surface.

The researchers first stamp a pattern onto the wafer using microcontact printing, a technique developed by George Whitesides of Harvard University and his colleagues. The method works "like a rubber office stamp," imprinting a patterned layer of organic ink just one molecule thick onto the gold, says study coauthor Ullrich Steiner.

Steiner and his coworkers then place onto the stamped gold a blend of two polymers--polystyrene and polyvinylpyridine--in a small amount of solvent. The polymers don't mix readily, and they separate as the solvent evaporates. The polyvinylpyridine sticks preferentially to he bare gold, forcing the polystyrene to collect on top of the stamped parts.

The resulting polymer films could serve as masks for lithography--a chip-manufacturing process that requires some parts of a chip to be protected while other portions are etched away. Microcontact printing alone can't produce useful lithographic masks because the ink films are too thin, Steiner says. He believes that the combination of microcontact printing and polymer patterning--both of which are cheap, simple, and fast--will provide an alternative to standard lithography methods.

He suspects, however, that his technique won't be capable of making chips with very small features (SN: 11/8/97, p. 302). So far, it seems that the polymer lines must be 1 to 2 micrometers wide. "If you make narrower lines, is the polymer pattern able to follow that?" asks Alamgir Karim of the National Institute of Standards and Technology in Gaithersburg, Md., who is studying the physics of how polymer blends separate.

The technique may work very well for producing low-cost, all-plastic devices--"throwaway electronics," Steiner says. To assess that potential, he and his group are currently testing the technique with blends of conducting and light-emitting polymers.