Crucial to making computer chips and other products of the microelectronics industry is a class of gases known as perfluorocompounds (PFCs). PFCs, however, are suspected of contributing to global warming due in part to their stability in the atmosphere. As a result, scientists at MIT and elsewhere are exploring options toward reducing their use.
MIT researchers and colleagues from Stanford and the University of California at Berkeley are tackling one approach. They are looking for alternative gases that could do the same job but are environmentally more benign.
The MIT team, which is currently screening about 20 candidates, is led by L. Rafael Reif, a professor in the Department of Electrical Engineering and Computer Science (EECS) and director of the Microsystems Technology Laboratories. He is working with EECS graduate students Simon M. Karecki and Laura C. Pruette.
After the MIT researchers complete the screening process, the Stanford and Berkeley groups will conduct modeling and simulation studies "to improve our fundamental understanding of promising candidates," Mr. Karecki said. Chemistries that continue to look promising will be beta-tested at companies.
PFCs are used for two main processes in the microelectronics industry: wafer patterning and chamber cleaning. In the first, they are used to create patterns on chips by selectively removing parts of the film that coats each chip.
In the second, PFCs remove the residue that clings to the walls of the chamber in which a thin film is deposited on the chips. This is important because "once this residue builds up to a certain thickness, it starts to peel off, creating particles," Ms. Pruette said. "Those particles are really destructive to the chip-making process because they can be of the same dimension as features on the chip itself."
Successful replacements for PFCs must have low global-warming potential and little or no ozone-depleting potential, perform about as well as PFCs, include fluorine in their chemical makeup, and be relatively safe to use.
With respect to safety, Mr. Karecki noted, "we're working with a tradeoff." Generally speaking, PFCs are relatively safe and have desirable handling properties (for example, they are not flammable or corrosive). The stability of these molecules, however, is what leads to their long atmospheric lifetimes. The team is therefore looking for chemicals that are less stable so they will break down in the atmosphere. "But in most cases," Mr. Karecki said, "that instability also implies they will be nastier to work with."
The MIT researchers' list of potential replacements for PFCs was compiled from a literature search and suggestions from gas companies. In the screening process, they are first determining if a candidate chemical can indeed remove a thin film of silicon-based insulating material (silicon dioxide or silicon nitride) from the surface of a silicon wafer. If the answer is yes, they determine how fast it does so as compared to PFCs. (The team is focusing on silicon dioxide and silicon nitride because their removal accounts for the largest share of PFC usage in the microelectronics industry.)
The team has found a few chemicals that look promising. These include three iodofluorocarbons--gases that contain iodine as well as fluorine. In the next stage of the screening process, the researchers hope to apply these promising candidates to chamber cleaning (tests could begin this fall), with wafer patterning to follow. Chemicals that continue to appear promising will be analyzed by the Stanford and Berkeley groups and ultimately beta-tested.
Professor Reif notes that industry is very involved in the research. "We have direct interaction with equipment manufacturers and gas suppliers so that they will know our results very early on and will be able to incorporate them into next-generation equipment," he said. Both Mr. Karecki and Ms. Pruette said that this involvement with industry was a key reason for their interest in the research.
The PFC project, which has been underway since 1995, was initially sponsored by Sematech. It is now one of several organized and funded through the NSF-SRC Engineering Research Center for Environmentally Benign Semiconductor Manufacturing. The Center is supported by the National Science Foundation and the Semiconductor Research Corp., a consortium that includes all the major US semiconductor companies.
MIT, the University of Arizona, Stanford and Berkeley are senior partners in the Center. "The goal is to develop environmentally benign manufacturing techniques for the microelectronics industry," said Professor Reif, who is an associate director of the Center.
In addition to the PFC work, there are three additional projects at MIT that are organized through the Center. They are led by Associate Professor Karen K. Gleason of chemical engineering, Associate Professor Duane S. Boning of EECS and Professor Lionel C. Kimerling of materials science and engineering.
A version of this article appeared in MIT Tech Talk on April 30, 1997.