Gordon Teal
Gordon Teal was an American physical chemist whose crystal-growing work at Bell Labs made the junction transistor practical and whose move to Texas Instruments in 1953 produced the first commercial silicon transistor, announced in May 1954. Born in Dallas in 1907 and trained at Brown University in physical chemistry, he joined Bell Labs in 1930 and spent more than two decades there before returning to Texas, which makes his career a bridge between the laboratory era of the transistor and the commercial, silicon-based industry that followed.
His central insight at Bell Labs was that semiconductors needed to be single crystals, and that they could be grown to order. Early point-contact transistors were made from polycrystalline germanium, a patchwork of tiny grains whose boundaries scattered charge carriers and made behaviour erratic. Teal argued that a uniform single crystal would allow the theory of Shockley and others to be tested and exploited, and, starting in 1948, he and the Bell Labs technician John Little built a crystal puller based on the Czochralski method, in which a seed crystal is dipped into molten germanium and slowly withdrawn. Their single-crystal germanium of 1948 to 1950 was reported in a 1950 paper, and it gave the Bell Labs physicists material in which the lifetime of minority carriers was long enough for a real junction transistor to work.
Teal's crystal puller became the grown-junction transistor when dopants were changed during the pull. In 1950 Morgan Sparks and Teal produced the first grown-junction transistors by adding impurities to the melt in sequence so that a single crystal contained an n-type layer, a thin p-type layer and another n-type layer. The result, described in a 1951 paper by Shockley, Sparks and Teal, was a working junction transistor made from a bar cut out of a grown crystal. This was the practical realisation of Shockley's 1948 junction theory, and it made possible the first production junction devices at Western Electric. Teal also worked on silicon crystals at Bell Labs, because silicon promised higher-temperature operation and much lower leakage than germanium, but silicon's high melting point and reactivity made it far harder to purify and grow, and the device community at large remained focused on germanium in the early 1950s.
Teal left Bell Labs for Texas Instruments in 1953, which was effectively the beginning of the silicon era. Texas Instruments, the Dallas company that had taken a transistor licence in 1952, hired him as a research director, and he put together a small group, including the chemist Willis Adcock, to make silicon crystals and grown-junction n-p-n silicon transistors. In May 1954, at a conference in Dayton, Ohio, he announced that the company had a commercial silicon transistor available. Silicon devices could operate at higher temperatures than germanium ones, which suited military electronics, and the announcement gave Texas Instruments a lasting competitive edge. It also shaped the commercial map: the company that first sold silicon in volume became a dominant semiconductor supplier, and Jack Kilby's 1958 integrated-circuit work took place at a firm that had already staked its future on silicon.
His later career combined research leadership with recognition from the engineering profession. Teal remained at Texas Instruments for more than a decade, in senior research and technical roles, and retired in the late 1960s before dying in 1997. He was honoured by engineering societies and counted among the figures whose work made modern solid-state electronics possible. His recognition has always been somewhat quieter than that of Shockley, Bardeen and Brattain, because crystal growing and materials chemistry are less obviously glamorous than device physics, yet nearly every later step depended on his insistence on high-quality single crystals. The same Czochralski pulling method, scaled up and automated, still produces most of the silicon wafers used by the chip industry today.
A frequent misconception is that Teal invented the silicon transistor by himself and that silicon was a sudden breakthrough. Silicon transistors depended on years of prior work in purification, doping and contact-making, much of it done by Bell Labs colleagues; the achievement was to bring those pieces together in a commercial, reproducible product. Another is that the 1954 devices were good by modern standards. They were relatively expensive, limited in frequency and fragile in manufacture, and they succeeded because there was a real military and industrial need for transistors that did not fail at higher temperature.
Read Gordon Teal as the person who treated the semiconductor as a materials problem first: by insisting on grown single crystals, then on silicon, he supplied the raw material and the commercial proof that the rest of the transistor industry was built upon.