Walter H. Brattain

The American physicist Walter H. Brattain (1902-1987), a co-inventor of the transistor, devoted much of his life to research on surface states.

Although he was born in Amoy, China (February 10, 1902), Walter Houser Brattain spent the early part of his life in the northwest of the United States. He was raised in the state of Washington on a cattle ranch owned by his parents, Ross R. Brattain and Ottilie Houser, and earned his B.S. degree in physics and mathematics at Whitman College in Walla Walla, Washington. Brattain earned that degree in 1924 and an M.A. degree from the University of Oregon in 1926. He then moved eastward, taking his Ph.D. degree in physics at the University of Minnesota in 1929. Brattain's advisor was John T. Tate, and his thesis was on electron impact in mercury vapor. In 1928 and 1929 he worked at the National Bureau of Standards in Washington, D.C., and in 1929 was hired by Bell Telephone Laboratories.

Brattain's concerns at Bell Laboratories in the years before World War II were first in the surface physics of tungsten and later in the surfaces of the semiconductors cuprous oxide and silicon. During World War II Brattain devoted his time to developing methods of submarine detection under a contract with the National Defense Research Council at Columbia University.

Following the war, Brattain returned to Bell Laboratories and soon joined the semiconductor division of the newly-organized Solid State Department of the laboratories. William Shockley was the director of the semiconductor division, and early in 1946 he initiated a general investigation of semiconductors that was intended to produce a practical solid state amplifier.

Crystals of pure semiconductors (such as silicon or germanium) are very poor conductors at ambient temperatures because the energy that an electron must have in order to occupy a conduction energy level is considerably greater than the thermal energy available to an electron in such a crystal. Heating a semiconductor can excite electrons into conduction states, but it is more practical to increase conductivity by adding impurities to the crystal. A crystal may be doped with a small amount of an element having more electrons than the semiconductor, and those excess electrons will be free to move through the crystal; such a crystal is an n-type semiconductor. One may also add to the crystal a small amount of an element having fewer electrons than the semiconductor, and the electron vacancies, or holes, so introduced will be free to move through the crystal like positively-charged electrons; such a doped crystal is a p-type semiconductor.

At the surface of a semiconductor the level of the conduction band can be altered, which will increase or decrease the conductivity of the crystal. Junctions between metals and n-type or p-type semiconductors, or between the two types of semiconductors, have asymmetric conduction properties, and semiconductor junctions can therefore be used to rectify electrical currents. In a rectifier, a voltage bias that produces a current flow in the low-resistance direction is a forward bias, while a bias in the opposite direction is a reverse bias.

Semiconductor rectifiers were familiar devices by the end of World War II, and Shockley hoped to produce a new device that would have a variable resistance and hence could be used as an amplifier. He proposed a design in which an electric field was applied across the thickness of a thin slab of a semiconductor. The conductivity of the semiconductor changed only by a small fraction of the expected amount when the field was applied, which John Bardeen (another member of Shockley's division) suggested was due to the existence of energy states for electrons on the surface of the semiconductor. Charges occupying such states would form a layer that screened the interior of the semiconductor from external fields, and so drastically reduce any effect on the conductivity by such fields. Brattain undertook the investigation of the properties of the surface states, and in the course of his experiments he and Bardeen discovered a means of constructing a solid state amplifier that was distinct from Shockley's field-effect device.

Brattain began his experiments by measuring the change in potential of the surface of a crystal of silicon (with reference to an electrode near that surface) when it was exposed to light. Brattain subsequently found that by introducing an electrolyte between his reference electrode and the semiconductor surface and applying a bias to the electrode, he could greatly influence the potential produced by illumination of the semiconductor. He and Bardeen concluded that ions in the electrolyte migrated to the surface of the semiconductor and nullified the effect of the surface charge already there. It then became possible to observe Shockley's field effect.

Brattain and Bardeen next introduced a second electrode into their apparatus, which was a point contact on the semiconductor that was insulated from the electrolyte. Their semiconductor was a thin layer of n-type silicon on top of a block of p-type silicon, and they found that an increase of the bias on the first electrode in the forward direction would produce a decrease of the current flowing into the point contact under a reverse bias. There was some amplification observed in this circuit (and in other similar ones using silicon and germanium), but the factors were small, and the electrolytes used did not allow good response at high frequencies.

Bardeen and Brattain hoped to improve their devices by using a gold film in place of the electrolyte, and they intended to isolate it from the semiconductor (in this case a block of n-type germanium) by means of an oxide layer on top of the germanium. The oxide coating was inadvertently washed off, however, and the gold film made contact with the germanium. It was with this arrangement that a new effect was observed. A forward bias on the gold film increased the current that flowed to the point contact, which was the opposite effect from what had been observed previously. Brattain and Bardeen supposed that there was a current of holes flowing from the gold film into the semiconductor, and then into the point contact. The new amplifying effect was named the transistor effect.

Bardeen suggested that the gold film and point contact could be replaced by two closely spaced contacts. Brattain constructed the two contacts by wrapping gold tape around the point of a wedge of polystyrene and scraping the gold away from the point of the wedge. The wedge was then pressed against a block of n-type germanium. On December 16, 1947, the device was incorporated into a small amplifier that had a gain of more than 18 and good frequency response. A week later the amplifier was demonstrated for the staff of Bell Laboratories, although a public announcement was not made until June of 1948. For their invention Brattain, Bardeen, and Shockley were awarded the 1956 Nobel Prize in Physics. Brattain also received the Stuart Ballantine Medal, the John Scott Medal, and in 1974 was inducted into the National Inventors Hall of Fame.

Brattain continued to carry out semiconductor research at Bell Laboratories until he retired in 1967. Between 1962 and 1972 he frequently taught courses at Whitman College, and from 1965 until 1975 he took part in a research program to model cell membranes as phospholipid layers.

Brattain was elected to the National Academy of Sciences in 1959 and received many other honors. In 1935 he married Keren Gilmore, and the couple had one son. Following the death of his first wife, Brattain married Emma Jane Kirsch Miller in 1958. Brattain lived in retirement in the state of Washington.

On October 13, 1987, Brattain died in Seattle, Washington, but not without leaving a permanent legacy. In its January 1997 profile of 25 visionaries, Workforce cited Brattain, along with his co-inventors Bardeen and Shockley. "AT&T's Bell Laboratories has," the article stated, "spawned numerous inventions, but none more significant than the transistor...today, the transistor serves as the basic building block for all solid-state electronics." In the late 1990s, Brattain's invention could be found in cellular telephones, fax machines, computers, automatic cameras, satellites, and many other electronic devices.

Further Reading

• The invention of the transistor has been described by Lillian Hoddeson in her article "The discovery of the point-contact transistor," Historical Studies in the Physical Sciences, 12 (1981-1982). Bardeen, Shockley, and Brattain also recount their experiences in their Nobel addresses: John Bardeen, "Semiconductor research leading to the point contact transistor;" William Shockley, "Transistor technology evokes new physics;" and Walter H. Brattain, "Surface properties of semiconductor," all in Nobel Lectures: Physics, 1942-1962 (Amsterdam, 1964). Appended to each of these addresses is a short biography of the author.
• Bardeen's discussion of semiconductor surface states appeared in the article "Surface states and rectification at a metal semiconductor contact," Physical Review, 71 (1947). The first published description of the transistor is Bardeen and Brattain's article "The transistor, a semi-conductor triode," Physical Review, 74 (1948).