Jack Kilby, inventor of the microchip, dies

The Nobel laureate and gentle giant Jack Kilby, who has died of cancer aged 81, was acknowledged as the inventor of the microchip.

He had just joined Texas Instruments in 1958 when he came up with what is called the monolithic idea — to put all the elements of an electronic circuit on a single silicon chip. This packed many more transistors into smaller spaces and therefore made it possible to create the thousands or millions of wires needed to connect all the transistors and other devices at the same time, and for no extra cost. At the time, interconnections were being done by hand.

The silicon chip made electronics devices dramatically cheaper to manufacture, helping to create industries that are now worth $1-trillion annually. TI’s first big order was for chips for Minuteman intercontinental ballistic missiles in 1962, but chips now control most music and media players, digital cameras, mobile phones and other devices, as well as calculators, computers and games consoles.

Kilby was born in Jefferson City, Missouri, but grew up in Great Bend, Kansas, where his father worked his way up to be president of the Kansas Power Company. The boy wanted to become an electrical engineer, like his father, and planned to go to the Massachusetts Institute of Technology, but failed the entrance exam.

Soon after he enrolled at the University of Illinois, the Japanese bombed Pearl Harbor, and so Corporal Kilby was soon repairing, and improving, army radios in north-eastern India. Postwar, he graduated in 1947 in electrical engineering, and went to work for Centralab, an electronics company in Milwaukee, where he gained experience in transistors and took a masters from the University of Wisconsin.

The slow-spoken, very tall midwesterner spent a decade at Centralab and patented 12 inventions. When the giant TI offered him a job in its semiconductor research group, he moved to Dallas with his wife Barbara and two daughters. That’s where he spent the rest of his life.

TI was famous for its cheap transistors, which were creating a mass market in radios. Kilby’s monolithic idea took the company into pocket calculators and other devices, and changed the world. End of story? Not quite. Another electrical engineer, Robert Norton Noyce, came up with the same monolithic idea working independently.

Noyce had completed his PhD at MIT and gone to work for Philco, in Philadelphia, also pioneering the uses of transistors. Noyce was hired by William Shockley, the inventor of the transistor, to join his new semiconductor company in Palo Alto, California (which eventually became Silicon Valley). Noyce and his friend Gordon Moore later founded Intel, to manufacture these new integrated electronics (hence the name).

Noyce, who was an electronics superstar, should have had the idea first, and his realisation of it was the most complete and easy to manufacture. However, the lab books show Kilby beat him to it: July 1958 versus January 1959.

Both TI and Fairchild wanted to patent the microchip, and TI got there first.

Unfortunately, its application included ”flying wires”, based on Kilby’s hand-made demonstration chip. This idea was fundamentally wrong; the application added that connections could also be laid down in a conducting material such as gold (also wrong). Noyce’s application made it clear that the interconnections were created at the same time as the transistors on the chip, which became the basis for the chip manufacturing industry.

Both patent applications were granted, and Kilby versus Noyce was set for a decade in the courts, at a cost of many millions of dollars.

Kilby won the first case, on the grounds that his application was good enough, but this decision was reversed on appeal. TI took the case to the Supreme Court, which also ruled in Noyce’s favour — but did so almost 11 years after Kilby had his idea, and it no longer mattered. The industry was moving far too fast to bother about the glacial speed of the law.

TI and Fairchild had already agreed to grant each other licences to make chips, and both were licensing the rights to other companies and collecting royalties.

Everybody agreed that both Kilby and Noyce deserved credit for the monolithic idea, they were regarded as joint inventors, and both were awarded America’s National Medal of Science.

It emerged later that an English engineer, GWA Drummer, at the Royal Radar Establishment, had had the monolithic idea by 1952, and had tried to build a microchip before either.

Kilby continued to work for TI, where he made important contributions to the development of handheld calculators and thermal printers. He took on a consulting role after 1970, became a distinguished professor at Texas A&M University, and retired from TI in 1983.

He was awarded the Nobel Prize for physics in 2000, as well as many other awards, and held more than 60 patents.

He is survived by his daughters, Janet and Ann.