Museum of American Exceptionalism
← Invention

Bell Labs, Murray Hill, New Jersey · 16 December 1947 · Bardeen, Brattain and Shockley

A Magnificent Christmas Present

Two physicists at Bell Labs pressed two gold points onto a sliver of germanium, and a signal came out a hundred times stronger. Every phone, computer and chip since starts there.

John Bardeen, William Shockley and Walter Brattain at Bell Labs, 1948, in a publicity photo made for the announcement of the transistor

The moment · 16 December 1947 · Murray Hill, New Jersey

Five weeks of trying. Walter Brattain wraps gold foil over a plastic wedge, slits it at the tip with a razor and presses it onto germanium with a spring. Two contacts, two thousandths of an inch apart. A small signal goes in. It comes out a hundred times stronger.

We discovered something today.John Bardeen
John Bardeen, William Shockley and Walter Brattain at Bell Labs, 1948, in a publicity photo made for the announcement of the transistor. AT&T / Bell Telephone Laboratories.
100×signal gain from the first transistor
0.002 inbetween its two gold contacts
3inventors, one Nobel Prize
92Btransistors on one Apple chip, 2023

After the war, the American telephone network ran on vacuum tubes that ran hot and burned out, and on metal relays that wore out with use. Mervin Kelly, who ran research at Bell Telephone Laboratories, wanted something small, solid and cool to replace them. He put a team of physicists, chemists and engineers on it. Two years later, two of them built it on a lab bench in New Jersey.

10 nm 100 nm 1 µm 10 µm 100 µm 1 mm 1 cm FIRST TRANSISTOR, 1947 about half an inch tall GAP BETWEEN ITS GOLD CONTACTS 0.002 inch, about 50 µm INTEL 4004 FEATURES, 1971 10 µm IBM 2 nm TEST CHIP GATE, 2021 12 nm, about two dozen atoms ABOUT A MILLION TIMES SMALLER
From the 1947 transistor down to a 2021 IBM transistor gate, on a log scale where each tick is ten times smaller, drawn to scale.

  1. A tube man wants something better

    Mervin Kelly, a Missouri physicist who made his name building vacuum tubes for the phone system, becomes director of research at Bell Telephone Laboratories. Tubes run hot, eat power and burn out. The switches in telephone exchanges are metal relays. Kelly starts hiring solid-state physicists. One of the first is William Shockley, fresh from MIT. Kelly tells him the day is coming when telephone switching will be done electronically.

  2. The solid-state group

    As the war ends, Kelly sets up a new group for basic research in solid-state physics, led by Shockley and chemist Stanley Morgan. Experimenter Walter Brattain, at Bell since 1929, joins. In October theorist John Bardeen arrives from the Naval Ordnance Laboratory. Chemist Robert Gibney and physicist Gerald Pearson round it out. Shockley’s first design for a semiconductor amplifier fails to work. In March 1946 Bardeen works out why: electrons trapped at the surface block the effect.

  3. The miracle month

    Gibney suggests putting a voltage across an electrolyte to break through the surface barrier. It works. For five weeks Brattain and Bardeen try one setup after another, on silicon and then on germanium, in water, in electrolyte, with metal points coated in wax. On 8 December a germanium version amplifies 330 times. On 12 December Brattain accidentally washes off an oxide layer and finds the effect works at every frequency he tries.

  4. Gold, germanium and a plastic wedge

    Bardeen works out that the two contacts must sit about two thousandths of an inch apart. Brattain wraps a strip of gold foil over the point of a plastic triangle, slices it at the tip with a razor blade and presses the two gold edges onto a slab of germanium with a spring. A small signal on one contact controls a larger current through the other. The output is about 100 times the input. It is the first transistor.

    A replica of the first point-contact transistor: gold foil on a plastic wedge, pressed onto germanium by a spring
    A replica of the first point-contact transistor: gold foil on a plastic wedge, pressed onto germanium by a spring. Florian Schäffer / Heinz Nixdorf MuseumsForum.
    Bardeen to his wife, Jane, that eveningWe discovered something today.

  5. The Christmas present

    The day before Christmas Eve, Shockley brings Bell Labs’ research leadership to see it. Speech goes into a microphone, through the transistor and out to headphones. Switch the device in and out and the voice jumps in volume with no loss of quality. Measured at a fixed frequency, the power gain is 18 or better. Shockley calls it a magnificent Christmas present. Bell Labs keeps it secret while the patents are prepared.

    Brattain’s lab notebook, 24 December 1947This circuit was actually spoken over and by switching the device in and out a distinct gain in speech level could be heard and seen on the scope presentation with no noticeable change in quality.

  6. The junction transistor

    Shockley spends the weeks after the demonstration on a design of his own. On 23 January he writes down the idea of the junction transistor: a sandwich of differently treated layers of semiconductor, with no delicate points at all. It is sturdier and easier to make in quantity. Morgan Sparks and Gordon Teal build the first working ones at Bell Labs in 1951. It becomes the workhorse transistor of the 1950s.

  7. The transistor goes public

    The military gets a private look first. Then, at Bell Labs’ headquarters on West Street, research director Ralph Bown shows the press the new device. Its name, transistor, came from engineer John Pierce and won a staff ballot in May. Bown tells reporters it can do just about everything a vacuum tube can do, and some things a tube cannot. The New York Times gives it a short item the next morning.

  8. Silicon and the pocket radio

    Bell Labs licenses the transistor to other companies for a $25,000 advance. At Texas Instruments in Dallas, Gordon Teal builds the first commercial silicon transistors, which keep working in heat that stops germanium. On 10 May he tells an engineering conference in Dayton he happens to have a few in his pocket. On 18 October TI and Regency of Indianapolis announce the TR-1, the first transistor radio sold. It has four transistors and costs $49.95.

  9. The Nobel Prize

    Shockley, Bardeen and Brattain share the Nobel Prize in Physics “for their researches on semiconductors and their discovery of the transistor effect.” Bardeen’s Nobel lecture the next day is titled “Semiconductor Research Leading to the Point Contact Transistor.” By now transistors are in hearing aids, pocket radios and the Bell System equipment that routes long-distance calls.

The team

John Bardeen, 1948Age 39
Theoretical physicistJohn Bardeen

Born in Madison, Wisconsin, in 1908 and in college at 15. Led mine and torpedo countermeasures at the Naval Ordnance Laboratory in the war. Explained the surface barrier and guided the experiments. In 1972 he became the only person to win the physics Nobel twice.

Walter Brattain, 1948Age 45
Experimental physicistWalter Brattain

Born in China in 1902 and raised on a cattle ranch in eastern Washington. Joined Bell Labs in 1929 and built magnetometers to find submarines in the war. Made the gold-foil wedge with his own hands. Later taught at Whitman College, his alma mater.

William Shockley, 1956Age 37
Leader, solid-state groupWilliam Shockley

Born in London to American parents in 1910 and raised in California. Led antisubmarine operations research for the Navy in the war and won the Medal for Merit. Invented the junction transistor, the sturdy design that made transistors practical to mass-produce.

Mervin J. Kelly, Bell Telephone LaboratoriesAge 53
Director of research, later president, Bell LabsMervin Kelly

Born in Princeton, Missouri, in 1894. Earned his doctorate in Chicago as an assistant to Robert Millikan. Built vacuum tubes for the phone network, then set out to replace them. Formed the solid-state group in 1945 and was president of Bell Labs from 1951 to 1959.

By the numbers

1E455the room in Building 1 at Murray Hill where the transistor effect was found
18×power gain measured at the 23 December 1947 demonstration
$25,000advance Bell Labs asked of each company licensed to make transistors
$49.95price of the Regency TR-1, the first transistor radio, in 1954
100,000+TR-1 radios sold, putting the word transistor into everyday speech
2,300transistors on the Intel 4004 of 1971, the first microprocessor
50 billiontransistors IBM fit on a chip the size of a fingernail in 2021
2Nobel Prizes in Physics won by John Bardeen, the only person to win two
A silicon wafer of Intel Haswell processors, each die a grid of transistors far too small to see, 2012
A silicon wafer of Intel Haswell processors, each die a grid of transistors far too small to see, 2012
Brattain at the bench with Bardeen and Shockley behind him, from Bell Telephone Magazine, 1948
Brattain at the bench with Bardeen and Shockley behind him, from Bell Telephone Magazine, 1948
The Regency TR-1 of 1954, the first transistor radio sold, opened to show its circuit board
The Regency TR-1 of 1954, the first transistor radio sold, opened to show its circuit board
A replica of Jack Kilby's first integrated circuit, built at Texas Instruments in 1958
A replica of Jack Kilby's first integrated circuit, built at Texas Instruments in 1958
The chip layout of the Intel 4004 of 1971, the first microprocessor, with 2,300 transistors
The chip layout of the Intel 4004 of 1971, the first microprocessor, with 2,300 transistors
Bell Telephone Laboratories at Murray Hill, New Jersey, 1942, where the solid-state group did its work
Bell Telephone Laboratories at Murray Hill, New Jersey, 1942, where the solid-state group did its work

Watch, listen, see

Films, photographs and first-hand accounts from across the web. Each opens in a new tab.

The Genesis of the Transistor, a 1965 Bell Labs film
Video · 17 minThe Genesis of the Transistor

A 1965 Bell Labs film reunites Bardeen, Brattain and Shockley to show how they built the first transistor.

AT&T Archives
Title card of The Transistor, a 1953 Bell Labs film
Video · 10 minThe Transistor, 1953

Bell Labs explains its new invention to the public and predicts wrist radios and portable TVs.

AT&T Archives
From a 1980s Bell Labs film on the invention of the transistor
Interview · 11 minThe Inventors Tell the Story

A 1980s Bell Labs film in which the inventors of the transistor recall how they did it.

AT&T Archives
A replica of the first point-contact transistor: gold foil on a plastic wedge, pressed onto germanium by a spring
DocumentBrattain’s Lab Notebook

The handwritten notebook page recording the first demonstration of the transistor, December 1947.

PBS Transistorized!
Brattain at the bench with Bardeen and Shockley behind him, from Bell Telephone Magazine, 1948
InteractiveTransistorized!

The full story of the invention, with biographies, notebook pages, a timeline and interactive science lessons.

PBS / American Institute of Physics
U.S. Patent 2,524,035, Bardeen and Brattain's point-contact transistor, granted 3 October 1950
DocumentThe Transistor Patent

Bardeen and Brattain’s patent for the point-contact transistor, filed June 1948 and granted October 1950.

Google Patents
John Bardeen, 1948
DocumentBardeen’s Nobel Lecture

Bardeen’s own account, given in Stockholm in December 1956, of the research that led to the transistor.

NobelPrize.org
A replica of Jack Kilby's first integrated circuit, built at Texas Instruments in 1958
InteractiveThe Silicon Engine Timeline

The museum’s semiconductor timeline, from the point-contact transistor to the integrated circuit and beyond.

Computer History Museum

“What we have to show you today represents a fine example of teamwork, of brilliant individual contributions and of the value of basic research in an industrial framework.”

Ralph Bown, director of research, Bell Telephone Laboratories, announcing the transistor, New York, 30 June 1948

After

Bell Labs shared what it had made. It licensed the transistor to any company that would pay the advance and held symposiums to teach licensees how to build it. On 12 September 1958 Jack Kilby of Texas Instruments showed his managers the first integrated circuit, a whole circuit on one sliver of germanium. In July 1959 Robert Noyce of Fairchild Semiconductor filed a patent on a silicon version built for mass production. Kilby won the Nobel Prize in Physics in 2000.

From there the count climbed without stopping. Intel’s 4004 of 1971 held 2,300 transistors. Apple’s M3 Max of 2023 holds 92 billion. By 2018 the industry had built an estimated 13 sextillion of them, and the Computer History Museum calls the transistor the most frequently manufactured human artifact in history.

In 2009 the IEEE dedicated a Milestone plaque at Bell Labs in Murray Hill. It records that from 17 November to 23 December 1947, Brattain and Bardeen, under the direction of Shockley, discovered the transistor effect there and built and demonstrated the point-contact germanium transistor.

Primary sources