Jay Last, a co-founder of Fairchild Semiconductor, was one of the most influential but least celebrated engineers at the dawn of the semiconductor era.
One of the least-well-known heroes of the semiconductor revolution, Jay Last, died on November 11, 2021. Last was one of the famous team of eight people that left Shockley Semiconductor Laboratory to found Fairchild Semiconductor. While there, he was leader of the team that developed the essential technologies that led to the first practical IC.
In 1946, between his junior and senior years in high school, Last hitchhiked from Pennsylvania to the orchards of San Jose, Calif., where he picked fruit for the summer. It was probably a lark at the time, but a momentous decision nevertheless. After earning a PhD in Physics from MIT in 1956, he ended up moving to the San Francisco Bay area to work at Shockley Semiconductor Laboratories in Palo Alto, California. His fruit-picking experience influenced his decision to move to California.
Shockley Semiconductor was founded in 1956 as a Beckmann Instruments subsidiary. At the time Last was wrapping up his doctoral thesis, working with a balky Beckman spectrophotometer, which led to plenty of interactions with Beckman’s people. As a result of these interactions, Arnold Beckman wanted to hire Last. William Shockley flew to Massachusetts from California to recruit Last at Beckman’s suggestion. During the job interview, Shockley helped Last surmount some stumbling blocks in his thesis work. Last was quickly impressed with Shockley’s intellect and agreed to join Shockley Semiconductor’s R&D team.
William Shockley started Shockley Semiconductor to develop diffused silicon transistors. Germanium was the preferred semiconductor material at the time, but germanium transistors have many shortcomings, including lower gain, lower operating frequencies, power-handling limitations, and thermal runaway. Silicon was seen as potentially a superior semiconductor material — if it could be tamed.
Several companies besides Shockley Semiconductor were also racing to develop silicon transistors. However, Shockley soon became obsessed with an entirely different sort of semiconductor device — the 4-layer diode — which he’d invented. Shockley took several members of his lab, including Last, off of silicon transistor development to focus on the development of the 4-layer diode.
Robert Noyce was the leader of Shockley’s silicon transistor development effort. He had been working with Last, Gordon Moore, and Jean Hoerni to overcome the technical challenges associated with making a diffused silicon transistor, and they were making progress when Shockley split the team to staff his 4-layer diode effort. Competing semiconductor companies and Bell Labs were also making progress on silicon transistor development, so Shockley’s diversion of researchers from the silicon transistor project was perceived by Noyce’s research group as a mistake.
Shockley was a difficult manager at best, psychologically abusive actually, and there was a rising tide of resentment towards him amongst his staff. Shockley had been particularly cruel to Last. When he gave Last a raise in salary, he told Last that he should never have agreed to the low hiring salary that he himself had initially offered. Being split off from Noyce’s development team to work on the 4-layer diode only made matters worse for Last.
Given the situation, it’s unsurprising that seven of Shockley’s researchers including Last, Julius Blank, Victor Grinich, Jean Hoerni, Eugene Kleiner, Gordon Moore, and Sheldon Roberts decided to leave together as a group. After several attempts, they recruited Noyce to join them as an eighth member of what would become known as the Traitorous Eight. Of the break, Last said, “It was clear that we were going to make the transistor that Shockley didn’t want to make.”
Fairchild Semiconductor Corp. incorporated on September 19, 1957. Once freed from Shockley, the team’s combined knowledge and creative energy soon made Fairchild Semiconductor the most influential semiconductor company ever created. Nearly every semiconductor company founded owes its existence to Fairchild.
Fairchild at first
Initially, Fairchild fixated on winning the race to develop silicon transistors. Noyce and Last focused on applying lithography to semiconductor manufacturing, a keystone technology needed to make these devices. Last earned a bachelor’s degree in optics from the University of Rochester in 1951, prior to getting his PhD. Rochester was the home of two of the largest optical companies in the US, Eastman-Kodak and Bausch & Lomb, so optics were deeply ingrained in Last’s skill set. In just ten months, Fairchild went from an empty building to a working silicon transistor, which the company demonstrated at Wescon in 1958.
Successful development of the silicon transistor only set the stage for what was to come.
What do you do after you successfully make silicon transistors using photolithography? You put several of them on a chip, wire them up, and make an integrated circuit. That inspiration hit Noyce like a lightning bolt, but it was Last’s job to turn the insight into silicon.
Noyce filed his integrated circuit patent in July, 1959, and that same month, told Last that he expected Texas Instruments would make a big deal about Jack Kilby’s integrated circuit at the Wescon conference in August. Last could not fabricate an actual IC from scratch in one month, but he did manage to demonstrate the concept by building one flip-flop on a ceramic substrate using four silicon transistor die, resistors drawn on the ceramic substrate with a graphite pencil, and conventional bonding wires to connect all of the components together. This demonstration hybrid was ready by Wescon. The work of building a real IC started immediately after the show.
Last led Fairchild’s integrated circuit technology development group. His group overcame many challenges. For example, they had to learn how to isolate the transistors from each other on the same die, and they needed to find a way to accurately apply the optimal metal for interconnecting all components on the IC. Last’s group succeeded and Fairchild produced working monolithic ICs in May, 1960. After much further refinement on practical device isolation techniques, the company announced its first line of digital ICs, dubbed Micrologic, at the IRE Show in New York City in March, 1961.
However, most of Fairchild’s management was solidly behind transistor and diode development, production, and sales. They were not in favor of IC development, fearing that ICs would cannibalize Fairchild’s transistor market. (They were right of course.)
R&D budget cuts and reductions in force during early 1961 came largely out of Last’s group and he resigned from Fairchild on January 31, 1961 along with Jean Hoerni and Sheldon Roberts. The trio founded another semiconductor company, the Amelco Division of Teledyne.
Amelco was devoted to the development of ICs. Last was Amelco’s director of R&D from 1961 to 1966. The company developed many devices for classified military and space projects, so it wasn’t well known in commercial circles. Last left Silicon Valley for Los Angeles in 1966 to become vice president of research and development for Teledyne, and in 1974, he retired from Teledyne and the semiconductor industry.
Last became an art collector and co-authored five art books: California Orange Box Labels, Fruit Box Labels, The California Style, California Watercolor Artists 1925–1950, and California Watercolors 1850–1970. He also wrote a book about the history of lithography titled The Color Explosion: Nineteenth-Century American Lithography.
Note: Much of the information for this article comes from Craig Addison’s 2007 Oral History Interview of Jay Last and Dr. Leslie Berlin’s biography of Robert Noyce: The Man Behind the Microchip: Robert Noyce and the Invention of Silicon Valley.
This article was originally published on EE Times.
Steve Leibson is a Principal Analyst at Tirias Research. He has 45 years of industry-leading expertise in the development of advanced electronic systems using a wide range of technologies and has held managerial and technical positions at several leading electronics companies including HP, Cadnetix, Tensilica, Cadence Design Systems, Xilinx, and Intel. An industry expert and thought leader since 1985, Steve has been writing about electronic development in several leading industry publications including EDN Magazine and Microprocessor Report. He served as the founding editor for Wind River’s Embedded Developer’s Journal. He has also published several books and book chapters covering many electronics topics including the use of processor IP for ASIC development and he has presented numerous technical seminars and webinars to technical audiences, spoken at major industry events worldwide, and has provided strategic consulting to many leading technology companies.