How it began
Jack Kilby joined Texas Instruments in Dallas in 1958 to work on miniaturization. Most of the staff left on a company vacation for which he was not yet eligible, and working almost alone he conceived of a circuit in which all the components are made from the same piece of material. On September 12, 1958, he built a working oscillator from germanium. In 1959 Robert Noyce of Fairchild Semiconductor conceived a monolithic circuit based on Jean Hoerni's planar process and filed a patent in July. Fairchild engineers produced the first working monolithic circuits in May 1960.
Before it existed
As computer systems grew more complex, engineers looked for simpler ways to interconnect the thousands of transistors they used. Government agencies funded micro-module and hybrid circuit projects. During the 1950s groups at RCA, Bell Labs, IBM and elsewhere integrated several components on a chip, but the Computer History Museum notes that none of these produced a general solution.
How it works
Kilby etched slices of germanium to form transistor, capacitor and resistor elements and joined them with fine gold "flying wires." Noyce's approach diffused diodes, transistors, resistors and capacitors into silicon and connected them with aluminum lines deposited on top of a protective oxide coating. This removed the hand-attached wires and gave a practical way to manufacture complete circuits on a single chip in volume.
Who pushed back
Kilby's flying-wire connections were not a practical production technique, and only a few dozen units of the first commercial device, the Type 502 flip-flop of March 1960 priced at 450 dollars, were shipped for evaluation. Fairchild and Texas Instruments fought over the patents in court for years. The courts eventually ruled in Noyce's favor, but by then the companies had settled on a cross-license.
What changed first
Some of the earliest uses were in computer equipment for the Apollo space missions and the Minuteman missile. Kilby worked to bring the technology to ordinary products, and the National High Magnetic Field Laboratory credits him with helping the handheld calculator debut in 1965.
What lasted
Kilby and Noyce both received the National Medal of Science and are regarded as co-inventors. Kilby shared the 2000 Nobel Prize in Physics, and suggested that Noyce, who died in 1990, deserved to share the honor. The planar method remains the fundamental approach used to produce integrated circuits.
