Jack Kilby

Jack Kilby: The Summer Nobody Else Was Working

In the summer of 1958 Texas Instruments emptied out for the company holiday. Jack Kilby, hired weeks earlier, had not accrued any vacation, so he stayed in a deserted laboratory in Dallas and worked on a problem the industry called the tyranny of numbers. On 12 September he showed his managers a sliver of germanium wired to an oscilloscope. The trace was a continuous sine wave. A complete electronic circuit had been made from one piece of material, and everything since — the microprocessor, the mobile phone, the internet — descends from that afternoon.

An Ice Storm in Kansas

Jack St. Clair Kilby was born on 8 November 1923 in Jefferson City, Missouri, and grew up in Great Bend, Kansas, where his father ran a small electric company serving customers scattered across the rural west of the state. The formative episode came during an ice storm that took down the power lines. Kilby watched his father use amateur radio to keep in contact with the utility's customers when nothing else would reach them. Electronics, from that point, was not an abstraction but the thing that worked when everything else had failed.

He graduated from Great Bend High School and took a bachelor's degree in electrical engineering from the University of Illinois in 1947 — a year, as he later observed, before the transistor was invented, which put him at exactly the right age for the field that followed. He married Barbara Annegers in 1948; they had two daughters, Ann and Janet. He went to work in Milwaukee for a manufacturer of radio and television components, studying at night for a master's in electrical engineering from the University of Wisconsin, which he completed in 1950.

The Tyranny of Numbers

The problem that consumed the industry in the 1950s was not making circuits work but making them big. Every additional transistor, resistor and capacitor had to be individually manufactured and individually soldered to its neighbours. A complex machine needed tens of thousands of hand-made joints, each one a chance for failure. The reliability ceiling was hard and low.

Kilby joined Texas Instruments in 1958 for a specific reason: it was, he said, the only company that agreed to let him work on electronic component miniaturisation more or less full time. His insight, arrived at in that empty summer laboratory, was that if all the components could be made from the same semiconductor material, they could be made at once, in one piece, and the soldering would vanish. He fabricated a phase-shift oscillator on a bar of germanium and demonstrated it on 12 September 1958. On 6 February 1959 he filed United States Patent 3,138,743 for miniaturized electronic circuits.

His version required tiny wires to be attached by hand between components — the flying-wire problem that Robert Noyce at Fairchild Semiconductor solved a few months later by printing aluminium connections onto a flat oxide surface. The two companies litigated for years and eventually cross-licensed. Kilby never disputed the shared credit; he consistently credited Noyce as a partner in getting the integrated circuit commercially accepted, and the two men shared the Charles Stark Draper Prize in 1989.

Making It Ordinary

Kilby's second act was proving the thing was useful. At Texas Instruments he led the teams that built the first military systems and the first computer to use integrated circuits. Then, in a decision of enormous commercial consequence, TI set him the task of finding a consumer product that would demonstrate the chip to ordinary people. With Jerry Merryman and James Van Tassel he co-invented the handheld electronic calculator — a device that in a few years destroyed the slide rule and put an integrated circuit into millions of pockets. He also co-invented the thermal printer that such calculators used.

In 1970 he took leave from Texas Instruments to work as an independent inventor, spending much of the following decade on silicon technology for solar power generation. From 1978 to 1984 he was Distinguished Professor of Electrical Engineering at Texas A&M University. He retired formally from TI in 1983 but stayed connected to it for the rest of his life. He held patents on the integrated circuit and seven other inventions — a small number for a man of his stature, and a fair indication that he was interested in a few large problems rather than many small ones.

The recognition came steadily. Election to the National Academy of Engineering in 1967. The National Medal of Science from Richard Nixon in 1969. The IEEE Medal of Honor in 1986. The Draper Prize in 1989. The National Medal of Technology from George H. W. Bush in 1990. The Kyoto Prize in Advanced Technology in 1993. Honorary doctorates from Illinois, Wisconsin, Southern Methodist and Yale. In 1995 the IEEE created a signal processing medal in his name while he was still alive to see it. Then, in 2000, forty-two years after the demonstration, the Nobel Prize in Physics, shared with Zhores Alferov and Herbert Kroemer, for his part in the invention of the integrated circuit.

Why Jack Is Called a Genius

Kilby's genius was reductive rather than elaborative. Faced with a problem defined by the number of separate parts, he asked whether the parts needed to be separate at all — and answered by making resistors and capacitors out of semiconductor material, which is not what semiconductor material is good at, purely so that everything could be one piece. It was an ugly solution in engineering terms and a brilliant one in conceptual terms, and he found it alone, in an empty building, in a few weeks.

The Nobel committee's endorsement is unambiguous, and the Royal Swedish Academy rarely honours industrial engineering at all; the 2000 award to Kilby was in part a statement that the integrated circuit belonged in the same category as fundamental physics. The National Medal of Science and the IEEE Medal of Honor say the same in an American register.

The counter-case is real and Kilby himself would have made it. His original device was not manufacturable; Noyce's planar version, not Kilby's, is the ancestor of every chip made since. The idea of integration was in the air — the tyranny of numbers was a named problem that many laboratories were attacking, and had Kilby not solved it in 1958 someone would have shortly after. He was also, by the accounts of those who knew him, an exceptionally modest man who resisted the lone-inventor framing. What is not disputable is priority and nerve: he built the thing first, and he did it as a new employee with no standing, on a hunch, during a holiday.

Legacy

Kilby died of cancer in Dallas on 20 June 2005, aged eighty-one. His name is on the Kilby Labs at Texas Instruments, a computer centre at Edinburgh Napier University, an IEEE medal, an award foundation established in 1980 and a statue at the University of Texas at Dallas; his manuscripts went to Southern Methodist University's DeGolyer Library. The more accurate memorial is that the entire electronics industry is a series of footnotes to a demonstration that lasted a few seconds on an oscilloscope screen in an empty lab.

Achievements

• National Medal of Technology and Innovation — 1990
• Nobel Prize in Physics — 2000
• Affiliated with Texas Instruments
• Educated at University of Illinois Urbana-Champaign, University of Wisconsin–Milwaukee and University of Wisconsin–Madison
• Worked as inventor, university teacher and computer scientist