How the Transistor Was Invented

The Mechanism

For most of the 1940s, every radio, telephone, and early computer ran on vacuum tubes - glowing glass bulbs the size of a thumb that amplified electrical signals by boiling electrons off a hot filament. Vacuum tubes were big, hot, fragile, and power-hungry. The ENIAC, the first room-sized electronic computer, ran on 18,000 of them. Bell Telephone Laboratories - the research arm of the American telephone company - had been hunting since the late 1930s for a smaller, cooler, more reliable replacement. The group leader, William Shockley, had assembled a solid-state physics team to look for a way to amplify a signal using a semiconductor crystal - a material whose ability to conduct electricity sits between a metal (which conducts freely) and an insulator (which does not conduct at all). The theorist on the team was John Bardeen, a quiet 39-year-old from Wisconsin who had come to Bell Labs in 1945. The experimentalist was Walter Brattain, a 45-year-old who had grown up on a cattle ranch in Washington State. In November 1947 Bardeen had a key insight: if you put a small bias voltage on a semiconductor surface, you could control how much current flowed through it. Brattain spent the next four weeks at the bench trying to build a working amplifier around that idea. The critical innovation came on a small plastic wedge. Brattain wrapped a single thin strip of gold foil around the apex of the wedge and then sliced the gold along the apex with a razor blade - producing two electrically separate gold contacts a hair-width apart. He pressed the wedge down onto a small slab of high-purity germanium. A third wire connected to the bottom of the germanium completed the circuit. The two gold strips were called the emitter and the collector; the germanium underneath was the base. Apply a small voltage to the emitter, and a much larger current would flow through the germanium and out through the collector. On the afternoon of 16 December 1947 Bardeen and Brattain tried it. The output current was about a hundred times bigger than the input. The device was amplifying. They wired a small loudspeaker across the output and could hear the signal from an audio oscillator coming out the other side. Brattain went home that night in the carpool and told his fellow riders he had just performed the most important experiment of his life. He was right. Their group leader William Shockley had been out of town in Chicago for the Christmas holiday and missed the entire demonstration. He spent the next four weeks alone in his hotel room reverse-engineering the design and producing a better one: the bipolar junction transistor, made of three layers of differently-doped semiconductor crystal with no metal point contacts at all. The junction transistor became the dominant architecture of the 1950s and 1960s, and it is the structural ancestor of every transistor in every commercial device on the planet today. The first commercial transistor product appeared in 1952 - a Sonotone hearing aid. By 1954, 97 percent of American hearing aids used transistors. The Regency TR-1 transistor radio appeared the same year. The integrated circuit came in 1959. The first microprocessor in 1971. In 1956 Shockley, Bardeen, and Brattain shared the Nobel Prize in Physics - but the three men, by then no longer on speaking terms, refused to be photographed together at the ceremony. Bardeen went on to win a second Nobel Prize in Physics in 1972, for a completely different discovery about superconductivity. He is the only person in history ever to win the physics Nobel twice. By 2026, a single piece of silicon the size of a postage stamp can contain more than 200 billion transistors - all of them descendants of the small plastic wedge of gold foil pressed against a slab of germanium in a ground-floor laboratory in New Jersey on a winter afternoon in 1947.