The Off-Key Bard
History

This Day in History — The First Working Integrated Circuit

On this day in 1958

On September 12, 1958, Kilby pressed a switch and a sine wave appeared — proving the first integrated circuit worked, on a sliver of germanium 7/16 by 1/16 of an inch.

The Off-Key Bard peers down at a tiny strip of semiconductor material connected by delicate wires. It doesn't look revolutionary. There are no screens, keyboards, blinking lights, or recognizable computer parts. But sitting on that workbench is an idea that will eventually help shrink computers from room-sized machines into devices we carry in our pockets.

"The breakthrough wasn't simply making a better electronic component. It was realizing that many components could be made together as one."

On this day in 1958, Jack St. Clair Kilby — a newly hired electrical engineer at Texas Instruments in Dallas — demonstrated a working integrated circuit to a group of company executives that included TI president Mark Shepherd. He held up a tiny bar of germanium measuring 7/16 of an inch by 1/16 of an inch, attached to an oscilloscope, and pressed a switch. A continuous sine wave appeared on the oscilloscope screen.

The first integrated circuit had worked.

Kilby later reflected on what neither he nor anyone else understood in that moment: "What we didn't realise then was that the integrated circuit would reduce the cost of electronic functions by a factor of a million to one. Nothing had ever done that for anything before."

The Empty Laboratory and the Tyranny of Numbers

The story of how the demonstration happened is inseparable from Kilby's status as a new employee.

He had joined Texas Instruments only in May 1958. Because he hadn't yet accumulated vacation time, he remained in the laboratory when TI instituted its traditional company-wide summer break — two weeks during which most of the research staff went home. Left largely alone in an empty laboratory, Kilby worked on a problem that had been nagging at electronics engineers for years: the "tyranny of numbers."

The tyranny of numbers was not about any single component. Transistors had already replaced bulky vacuum tubes and were working reliably. The problem was that building increasingly complex electronic circuits required assembling enormous numbers of individual components — transistors, resistors, capacitors — each manufactured separately and then physically wired together. A single complex circuit could require thousands of parts, each a separate object, each a potential point of failure, each adding size and weight and cost. There were real mathematical limits approaching on how many individually wired components could be assembled into reliable equipment.

In July 1958, Kilby made a sketch in his lab notebook: a circuit in which all the components were built from the same piece of semiconductor material. If resistors and capacitors could be formed directly from semiconductor properties — if the whole circuit could be integrated into a single substrate — the wiring problem largely disappeared. He got permission to pursue the idea.

By September, he had built a phase shift oscillator on a sliver of germanium 7/16" × 1/16". The device had fine gold wires bonded to it externally — the prototype was crude, by any standard — but the components were all part of the same piece of material.

On September 12, he pressed the switch. The sine wave appeared. He had solved the tyranny of numbers.

The Patent and the Competition

Kilby filed a patent for "A solid circuit made of germanium" on February 6, 1959. But he was not the only person who had been thinking about the problem.

A few months after Kilby's demonstration, Robert Noyce at Fairchild Semiconductor independently arrived at a related idea — and crucially, he designed it using silicon and the planar manufacturing process pioneered by Jean Hoerni. Planar processing allowed components and connections to be formed on a flat silicon surface, which made it dramatically easier to manufacture integrated circuits reliably in large quantities. Noyce's design also used vapor-deposited aluminum interconnections instead of Kilby's hand-bonded gold wires, which made mass production genuinely practical.

The invention of the integrated circuit is credited jointly to Kilby and Noyce. Kilby demonstrated the first working device. Noyce developed the silicon-based architecture that made the technology manufacturable at scale. Both filed patents; the patent dispute between their employers lasted for years and was eventually settled.

From Oscilloscope to Pocket Calculator to Billions of Transistors

Texas Instruments management, recognizing that the integrated circuit needed a compelling commercial demonstration, challenged Kilby to design a calculator as powerful as the large electromechanical desktop models of the day — but small enough to fit in a coat pocket. The result, the handheld electronic calculator, helped establish the integrated circuit as a commercially viable product in the late 1960s.

What followed is the story of the modern world. Integrated circuits allowed engineers to place increasing numbers of transistors onto ever-smaller chips. A circuit requiring a room full of wired components could be reduced to a single chip. Then to a chip with hundreds of components. Then thousands. Then millions.

Today, modern processors contain tens of billions of transistors on silicon dies smaller than a postage stamp — a direct line of descent from the sliver of germanium Kilby demonstrated on September 12, 1958.

Kilby was awarded the 2000 Nobel Prize in Physics for his part in the invention of the integrated circuit. Robert Noyce, who died in 1990, could not share the prize because Nobel Prizes are not awarded posthumously, though the committee specifically acknowledged his contribution. Kilby himself died on June 20, 2005.

"A circuit small, a simple start,
With many pieces joined as part.
From germanium the future grew —
And billions fit where once were few."

History reminds us: the greatest technological revolutions don't always begin with spectacular machines. On September 12, 1958, the future of computing was a sliver of germanium smaller than a fingernail, connected to an oscilloscope, in an otherwise quiet Texas laboratory — waiting for someone to press a switch.

Sources

← Back to the Desk — History