History of the Transistor
The history of the transistor is the story of how a laboratory attempt to build a better solid-state amplifier in 1947 became the most manufactured object in human history, with the number of devices on a single modern chip now in the tens of billions. It is not a single invention but a chain: earlier ideas about rectifiers and amplifiers, a breakthrough at Bell Labs, a series of improved device types, and then integration that packed enormous numbers of transistors onto one piece of silicon.
The prehistory consists of two lines of work, rectifiers and amplifiers, that the transistor eventually merged. In 1874 Karl Ferdinand Braun observed that contacts between metal and certain crystals conduct better in one direction than the other, an early rectifier that later led to crystal radio detectors. In 1906 Lee de Forest introduced the Audion, the triode vacuum tube, which could amplify signals and became the basis of radio, telephone repeaters, and the first electronic computers. In 1925 Julius Edgar Lilienfeld filed a patent describing a device that controlled current in a semiconductor with an electric field, a concept resembling a field-effect transistor, but the materials of the day were too poor for it to work.
The decisive event was at Bell Labs in December 1947, when John Bardeen and Walter Brattain built the first working transistor under William Shockley's group. The point-contact device, made from germanium with two closely spaced metal contacts, showed amplification when demonstrated on 23 December 1947, and Bell Labs announced it publicly on 30 June 1948. Shockley, who had pushed for a field-effect design and was left out of the first patent, worked out the theory of the junction transistor in 1948, and the first junction transistors were made in 1950. Junction devices were more rugged, quieter, and easier to manufacture than point-contact ones, and they became the dominant form for the next decade.
The 1950s were the decade of materials and applications. Gordon Teal's work on single-crystal growth made reliable germanium and then silicon transistors possible, and Texas Instruments announced the first commercial silicon transistor in 1954. The same year the Regency TR-1 pocket radio put transistors in consumers' hands, and transistorized computers began to appear by the middle of the decade. In 1956 the Nobel Prize in Physics went to Shockley, Bardeen, and Brattain, which formally marked the transistor as a scientific milestone. Shockley then founded his laboratory in California, which, through the Fairchild spin-off, seeded Silicon Valley.
The decade's end brought the two ideas that turned the transistor from a part into a platform. Jack Kilby demonstrated a germanium integrated circuit on 12 September 1958, and Robert Noyce's planar, silicon version, built on Jean Hoerni's process, made integration manufacturable in 1959. That same year Mohamed Atalla and Dawon Kahng at Bell Labs built the first working MOSFET, the metal-oxide-semiconductor field-effect transistor, which realized the field-effect idea that Lilienfeld had patented decades earlier. The MOSFET's simplicity and small size made it ideal for dense chips, and CMOS logic built from it, introduced in 1963, became the basis for most modern processors and memory.
Scale followed: from a few transistors per chip in 1960 to about 2,300 on Intel's 4004 in 1971, and now tens of billions on the largest processors and AI accelerators. The progression tracked Moore's law, driven by repeated shrinking of the MOSFET. Today's devices are made with extreme ultraviolet lithography and three-dimensional structures, and they are used in billions of phones, cars, appliances, servers, and satellites. Most of the transistors ever made were made in the last few years.
Common misconceptions help show what the history actually is. The transistor was not invented by one person; Bardeen, Brattain, and Shockley share the credit, with Lilienfeld's earlier patent and decades of crystal physics behind them. It also did not replace the vacuum tube overnight, since tubes remained in use for years and still serve in niche roles. And the first transistors were not silicon, nor were they the field-effect devices that now dominate: the line runs from germanium point-contact to germanium junction, to silicon, to the silicon MOSFET.
Read the history of the transistor as a relay race in which each runner fixed the previous one's weakness: the idea of a semiconductor amplifier came first, Bell Labs made it work, better crystals and silicon made it reliable, integration made it cheap, and the MOSFET made it small enough to put tens of billions on a chip.