Transistor Technologies

Summary

Transistor technologies require the expertise of an interdisciplinary group that includes physicists, chemists, and engineers. The semiconductor transistor was invented by physicists and may be the single most important invention of the twentieth century. As an electronic amplifier and on/off electronic switch, the transistor has revolutionized electronics and many fields of industry, such as banking, manufacturing, automobiles, aircraft, military systems, space exploration, medical instrumentation, household appliances, and communication. Transistors have been so greatly miniaturized in size that semiconductor chips can contain up to one billion transistors on a silicon chip no larger than a thumbnail.

Definition and Basic Principles

NPN and PNP are symbols for the two most important junction transistors. There are three legs or wires leading from each of these transistors, one each from the three separate regions. NPN means that a p-type region with positive holes separates two n-type regions with electrons carrying the electric current. Thus, there are two junctions. They both have natural potential barriers that prevent electrons in the n region from entering the p region and prevent holes from entering the n region until external voltage is applied. When external voltage is applied, the transistor functions. The p region is called the base of the transistor. When negative voltage is applied between the first n region and the base, the first n region becomes an emitter of electrons into the base region. Positive voltage is applied to the other n region between it and the base. That n region and its junction become the collector of the transistor because it collects the electrons emitted from the emitter. Its potential barrier becomes even larger and an excellent collector of electrons. An electrical signal applied to the emitter modulates the current from emitter to the collector. The input signal is amplified in the collector circuit, which is a high-resistance circuit.

A common current is flowing from a low-voltage input to a high-voltage output. There is both voltage and power gain. There could be current gain if the input signal source were placed in the base leg and the emitter were grounded. The p region has mobile holes, and electrons can be captured by the positive holes in electron-hole recombination. This is detrimental. Recombination is minimized by creating very thin regions and using very high-quality semiconductor material. The PNP transistor is identical to the NPN, save for the changing places of the electrons and holes.

Background and History

Walter H. Brattain and John Bardeen, physicists at Bell Laboratories, discovered the transistor effect in December 1947. They had attempted to produce a semiconductor surface field effect amplifier but failed. In the process of studying surfaces to find the cause of failure, they probed the germanium surfaces with a sharp metal point. They discovered that an electrical signal into one point contact produced an amplified signal in the circuit of a second point contact close to the first. The point contact transistor was born. The name “transistor” was chosen by John R. Pierce, a fellow physicist at Bell Labs. Pierce reasoned that because the new device had current input and voltage output, it should be viewed as a transresistance. Since other devices (conductor, resistor, varistor, thermistor) ended in “-or” the name of the new device should end in “-or,” hence the name transistor. Brattain's and Bardeen's supervisor was physicist William B. Shockley, who was not included among the inventors. Shockley went on to invent what would become an even better transistor: a p-n junction transistor. The properties of this transistor could be much more readily designed and controlled than those of a surface-point contact device because Shockley's transistor depended on the “inside” bulk properties of the semiconductor, which were much more easily controlled than the surface device. The early industry manufactured some point-contact transistors, but Shockley's junction transistor dominated the growing industry into the 1960s. As a result, all three, Bardeen, Brattain, and Shockley, received the Nobel Prize in Physics in 1956 for research in semiconductors and the discovery of the transistor effect.

How It Works

Producing Junction Transistors. Junction transistors can be produced during the original crystal growing process for the silicon (or germanium) crystal by adding known n-type and p-type impurities to the molten semiconductor as the solid crystal is slowly pulled from the melt.

An improved method for producing junction transistors uses metal-alloying techniques. For example, aluminum as a dopant (desired impurity) makes silicon p-type. With a small n-type chip of silicon, a small ball of aluminum can be placed onto each of the two surfaces of the silicon chip and the temperature raised to a level high enough to melt the aluminum but not melt the silicon. However, the molten aluminum does alloy with a small surface region of the silicon, and a PNP transistor is produced.

One of the best ways for producing NPN and PNP junction transistors is by diffusing dopants into the silicon. Diffusion of impurity atoms into silicon is a slow process and must be done at high temperatures without melting the silicon. Double diffusions and triple diffusions have been found useful. One double diffusion method might start with n-type silicon into which a p-type impurity is diffused (such as boron). This produces one junction. An n-type impurity (such as phosphorus) can be diffused on top of the p-type diffusion.

The n diffusion is made to go less deep than the p diffusion, but it must be of higher concentration to overcome and change part of the p region back into n, thereby producing an NPN transistor.

Another way to produce NPN and PNP junction transistors is by means of ion implantation. In this case, atoms of a dopant are ionized and then shot into the surface region of the silicon with very high energy, driving the impurity into the silicon to whatever depth is desired in the design. Because this process is very energetic, it damages the crystal structure of the silicon. The silicon must then be annealed at some higher temperature in order to remove the damage but, at the same time, leave the impurity in place. All p-n junction transistors are called bipolar transistors because both n-type and p-type regions are used in each transistor.

Surface Field Effect Transistors. Surface field effect transistors are called unipolar transistors because in any particular transistor only n-type regions and electrons are used, or only p-type regions and holes are used. An external electric field is applied to the surface of the silicon and modulates the electrical conductivity. In place of an emitter-base-collector of the junction transistors, the field effect transistor has a source, gate, and drain. The gate is a thin metal region separated from the silicon by a thin layer of insulating dielectric silicon dioxide. The gate applies the electric field. The source and drain are both n-type if the current flowing through the device is electrons. The source and drain are both p-type if the current flowing is holes. Surface field effect transistors have become the dominant type of transistor used in integrated circuits, which can contain up to one billion transistors plus resistors, capacitors, and the very thinnest of deposited connection wires made from aluminum, copper, or gold. The field effect transistors are simpler to produce than junction transistors and have many favorable electrical characteristics. The names of various field effect transistors go by the abbreviations MOS (metal-oxide semiconductor), PMOS (p-type metal-oxide semiconductor), NMOS (n-type metal-oxide semiconductor), CMOS (complementary metal-oxide semiconductor—uses both p-type unipolar and n-type unipolar).

Applications and Products

One of the great advantages of the transistor, either in the form of a single transistor or many transistors on a chip, is that it is generic. The transistor or chip can be used in any electronic product requiring amplifiers and on-off switches with small size, ruggedness, low power loss, desirable frequency characteristics, and high reliability. The transistor can be made larger for higher power usage, smaller for high-frequency usage, extremely high reliability for military and space usage, and inexpensive for household and consumer products. Transistors are used in all modern electronics. Many different forms of the junction transistor were created at many companies during the first fifteen years of the industry. Engineers at Fairchild Semiconductor learned how to produce reliable surface field effect transistors, and the entire industry moved in that direction in the late 1960's and continues to do so. The industry had also started working with the semiconductor silicon, a more desirable choice than germanium. Many companies used both germanium and silicon, producing both diodes and transistors. National Semiconductor was formed in 1959 as the first company to use only silicon and to produce only transistors and its own early integrated circuits. Integrated circuits were originally invented by Robert Noyce at Fairchild, and later Intel, and by Jack Kilby at Texas Instruments in the late 1950's. The integrated circuit (the so-called chip) sealed the future success of the electronics revolution. Intel has become a major force in the semiconductor industry.

It would not be an exaggeration to say that every manufacturing company, every service organization, most homes in the United States, all advanced defense and military equipment, all aircraft, all research, all modern automobiles, and all modern communications use electronic equipment of some kind. In each of thousands upon thousands of applications, the electronic circuitry is different, but the generic transistor in single units or in massive numbers of transistors on single chips are common to all. These transistors and transistor chips are “hidden” in small packages, which in turn are hidden in the larger containers that house the particular electronic equipment. Intel has tried to change the hidden perspective by having computer manufacturers place a label displaying “Intel inside” on the outside covers of personal computers (PCs). Perhaps that helped PC users to realize that what was going on inside the personal computer was not magic but the result of one of the greatest inventions of the twentieth century—the transistor.

Here are a few examples of where transistors and transistor-packed chips are used:

• Aircraft: The entire control instrumentation and “fly by wire” flight systems.
• Automobiles: Engine controls and accessory controls
• Business and personal communication: Communications equipment, cell phones, all of the electronics of the Internet, smart phones.
• Financial institutions: Large computers.
• Household appliances: Microwave oven controls, washing machines, timer controls, televisions.
• Manufacturing: Automatic controls of many products.
• Medical instrumentation: All types of body scanners.
• Military systems: Intercontinental ballistic missiles' computers, inertial guidance systems, and telemetering systems and unmanned drone aircraft.
• Personal computing: Personal computers: desktops, laptops, notepads.
• Research and meteorology: Supercomputers.
• Space exploration: Everything for both manned and unmanned flights.

Software is needed to tie all of the electronics together and to tell the electronics what to do, and the software differs from one application to the next. The software and the electronics hardware are married. The transistors are operating at lightning speed as on/off switches and as amplifiers of smaller electric signals, but switches must be told what to do. All forms of computers have central processing units (semiconductor chips), semiconductor memories, and analogue-to-digital and digital-to-analogue semiconductor chips for interfacing with the real world.

Careers and Course Work

The semiconductor transistor chip industry has one of the the largest percentage of employees with doctoral degrees in all of industry. Highly-educated and trained people are also needed in the related industries that supply equipment and materials to the semiconductor industry. A semiconductor fabrication facility (fab plant) is so advanced with environmental requirements and manufacturing equipment that final capital costs can total several billion dollars for just one facility. If nanoscience and nanotechnology are successful in developing mass production processes for nanotransistors on chips, the requirements for extremely clean facilities and for very advanced manufacturing equipment will be huge.

Undergraduates will find many colleges and universities with courses in solid-state physics, microelectronics, and materials science and engineering. These courses are necessary for those expecting to enter the modern field of semiconductors. Following that preparation, it is important to gain experience in one of the nation's many nanoscience and nanotechnology centers, each of which is usually part of a university. University of California, Berkeley, Los Angeles, and Santa Barbara; The Johns Hopkins University; State University of New York at Albany; Purdue University; Carnegie Mellon University; Rice University; Illinois Institute of Technology; Cornell University; Harvard University; University of Massachusetts at Amherst; Columbia University; and the University of Pennsylvania are a few of the many schools that have nanoscience centers.

Social Context and Future Prospects

It would not be an overstatement to claim that virtually every manufacturing company, every service organization, and most homes in the United States use electronic equipment containing transistor-loaded semiconductor chips. Research in nanoscience and nanotechnology will continue to create a culture and a society even more heavily dependent on the availability of advanced electronics. Breakthroughs in transistor nanotechnology have been made in the twenty-first century. For example, a transistor gate one nanometer long was created in 2016, challenging the previous assumption that five nanometers was the smallest a transistor gate could be. Then, in 2022, scientists in Tsinghua University, China, built a transistor gate that was 0.34 nm, the size of a single carbon atom. Without question, research and development will continue to push transistor technologies even further.

Bibliography

Bondyopadhyay, Probir K., Pallab K. Chatterjee, and Utpal K. Chakrabarti, eds. Proceedings of the IEEE: Special Issue on the Fiftieth Anniversary of the Transistor 86, no. 1 (January, 1998).

Callister, William D., Jr., and David G. Rethwisch. Materials Science and Engineering: An Introduction. 8th ed. Hoboken, N.J.: John Wiley & Sons, 2010.

Desai, Sujay B. "MoS2 Transistors with 1-Nanometer Gate Lengths." Science, vol. 354, no. 6308, 2016, pp. 99–102.

"Electrical and Electronic Engineering Technologists and Technicians." U.S. Bureau of Labor Statistics, 18 Apr. 2022, www.bls.gov/ooh/architecture-and-engineering/electrical-and-electronics-engineering-technicians.htm. Accessed 8 June 2022.

"Materials Engineers." U.S. Bureau of Labor Statistics, 18 Apr. 2022, www.bls.gov/ooh/architecture-and-engineering/materials-engineers.htm. Accessed 8 June 2022.

Muller, Richard S., and Theodore I. Kamins, with Mansun Chan. Device Electronics for Integrated Circuits. 3d ed. New York: John Wiley & Sons, 2003.

Potoroaca, Adrian. “Chinese Researchers Create the World's Smallest Transistor Gate.” TechSpot, 11 March 2022, www.techspot.com/news/93745-chinese-researchers-create-world-smallest-transistor-gate.html. Accessed 7 June 2024.

Ramsey, Jeremy J. Nanotechnology: An Introduction. 2nd ed., Elsevier, 2016.

Reid, T. R. The Chip: How Two Americans Invented the Microchip and Launched a Revolution. Rev. ed. New York: Simon & Schuster, 2001.

Suplee, Curt. Physics in the Twentieth Century. Edited by Judy R. Franz and John S. Rigden. New York: Harry N. Abrams, in association with the American Physical Society and the American Institute of Physics, 1999.

Sze, S. M., and Kwok K. Ng. Physics of Semiconductor Devices. 3d ed. Hoboken, N.J.: John Wiley & Sons, 2007.