Famed semiconductor conference renamed for founder Lester Eastman

By Bill Steele

In the summer of 1967, Lester Eastman, then a junior faculty member in electrical engineering at Cornell, decided to get together with all the other people working on high-frequency, high-power transistors -- there weren't many -- and invited them to a conference. The conference has continued as a regular event every two years, while the once obscure technology has become an industry worth about $15 to $20 billion a year. Some technology Eastman pioneered is probably in the cell phone in your pocket and in the lasers that relay your call over fiber-optic networks.

Professor Lester Eastman, foreground, in the Microwave Measurements Laboratory in Phillips Hall, poses with Vinayak Tilak, who received his doctorate on July 17 to become the 111th Ph.D. trained by Eastman during his 37 years on the Cornell faculty. Nicola Kontoupes/University Photography

This year, what had been known as the Biennial IEEE (Institute of Electrical and Electronic Engineers) /Cornell University Conference on High-Performance Devices has been renamed the IEEE Lester Eastman Conference on High-Performance Devices. The conference will be held Aug. 6-8 at the University of Delaware and then rotate annually to other institutions. Eastman himself, now the John L. Given Foundation Chair and Professor of Engineering at Cornell, will open the conference as keynote speaker, reviewing his 37 years of research in the field.

Thirty-seven years ago, Eastman was working with vacuum tubes called "Klystrons" that came in machines about the size of a refrigerator. But at a scientific meeting he saw a demonstration in which a voltage applied to a little piece of gallium arsenide caused oscillation. Oscillation was what circuits using vacuum tubes did; it was how they "made waves." Eastman decided that solid-state was the future of microwaves.

He works with "compound semiconductors," notably gallium arsenide, and more recently gallium nitride, with a few new candidates under examination. The best-known semiconductor, silicon, is made up of a lattice of identical silicon atoms. To make transistors you add a few impurities. The oddball atoms break up the orderly crystal structure and leave a few electrons free to move around, and a few empty spaces through which they can move. The movement of the free electrons can be controlled by applying outside electric fields, so the devices can be used to switch current flow off and on, as in computers, or amplify it or make it oscillate, as in radios and wireless telephones.

In a compound semiconductor, the crystal is made up of alternating atoms -- say, gallium and arsenic. When a crystal like this is bonded to a slightly different material, free electrons appear at the junction to make conduction possible. In some arrangements the atoms can be made to absorb energy and kick out photons, making light-emitting diodes (LEDs) or lasers.

Compound semiconductors can be made to oscillate at very high frequencies because, for complex reasons involving quantum mechanics, the free electrons behave as if they were lighter than those in other materials, Eastman said. "When I started, the transistor had just been invented, and we were hoping we could get the frequency up to 1 megahertz so we could have a.m. radio," he recalled. Today's devices typically operate in the 10s of gigahertz range.

The other big difference is power. Increasing the electric field applied across a silicon transistor quickly causes it to break down. Gallium arsenide handles much higher fields at higher frequencies, and the newer gallium nitride devices can put out as much as 50 times the power of gallium arsenide. This means either higher power in existing devices -- such as the transmitters in satellites -- or current power levels in smaller spaces -- maybe wristwatch telephones.

Eastman is also notable for nurturing new scientists and engineers. Over the course of his career he has so far trained 111 Ph.D. students, many of whom have gone on to become leading faculty members at other institutions or CEOs of high-tech corporations. One, Sandip Tiwari, is the current director of the Cornell Nanofabrication Facility.

Assuming that support for one student over an average five years comes to around $100,000 a year, Eastman estimates that Cornell's budget for his past graduate students about equals the projected cost of the new Duffield Hall. There has been plenty of payback, he says, as graduates bring research funding from their employers back to Cornell. Corporations often offer research funding, he adds, to have first pick of Cornell grads. "In some cases it's listed in their budget as 'recruiting,'" he said.

Eastman is a recipient of the Heinrich Welker Medal, the Senior Alexander von Humboldt Fellowship Award, the Aldert van der Ziel Award and the Award of the International Symposium on Gallium Arsenide; is a member of the National Academy of Engineers; a fellow of IEEE and American Physical Society; and a member of the Electromagnetic Academy.

He expects to retire in two years, leaving a legacy that sums up several decades of technological progress. "When I started we had big things that put out milliwatts," he recalled, spreading his hands to encompass something a little bigger than a breadbox. "Now we have things the size of a human hair that put out several watts."