The History of the Transistor Marches On
From its first appearance at Bell Labs in 1947 to today, the transistor has revolutionized the electronics industry perhaps more than any other component.
Since the mid-20th century, the transistor has played a vital role in the innovation of modern technologies. While the transistor was primarily employed for amplification in an analog circuit and switching in a digital circuit, intensive research and development has continued to open doors for new transistor-based applications.
The transistor has evolved in sizes and types since its inception in the mid-20th century. Image used courtesy of Bell System Memorial
Thanks to very large-scale integration (VLSI) technology, billions of transistors can be placed on a single chip for use in computing applications. The M1 Ultra SoC from Apple, for example, is made up of 114 billion transistors—the largest number of transistors ever on a chip.
Dating back to the early 20th century, engineers have used transistors to amplify electrical signals. The first instance of this use case came from British electrical engineer John Ambrose Fleming when he invented the vacuum tube. However, vacuum tubes faced many drawbacks that would only be solved by the invention of the modern transistor.
The Point-contact Transistor: A Star is Born
The first recognized transistor was developed by Bell Labs researchers Walter Brattain and John Bardeen in 1947. After several efforts to make an amplifier with silicon, Bardeen and Brattain resolved to use a slab of germanium and two gold foils to make a point-contact transistor. They observed more holes for electrons when a gold foil was placed in close proximity to the surface of germanium. The Bell researchers also noticed that the current passing through the contact was further boosted and amplified at the other contact of the gold foil.
The point-contact transistor. Image used courtesy of Bell System Memorial
This discovery marked the dawn of a new transistor-led era in the electronics industry. In 1952, the point-contact transistor became widely accessible in commercial use and was instrumental in manufacturing telephone systems.
From Germanium to Silicon
In an attempt to improve the transistor design of Bardeen and Brattain, William Shockley fabricated the junction transistor from germanium in 1951. Shockley's junction transistor was simply a sandwich of semiconductors with three layers. The outer layers contained many more electrons than the middle layer. Shockley explained that this design allowed current to flow through the sandwiched semiconductors to make an amplifier.
While the point-contact and junction transistor relied on germanium, researchers soon thereafter noted that the component broke down at 180°F. This is because germanium introduces too many free electrons in transistors when it is heated to a very high temperature, breaking the whole component down.
Bell Labs' commercial junction transistors in 1951. Image used courtesy of Jack Ward, Transistor Museum and Computer History
This drawback inspired Gordon Teal, a researcher at Texas Instruments, to invent the first-ever silicon transistor in 1954. Teal's silicon transistor had the same working principle as the germanium transistor, but it could withstand high temperatures. The silicon transistor was an n-p-n structure and was fabricated via a grown-junction process.
MOSFETs Make the Modern Era
The development of the silicon transistor led to the invention of more silicon-based transistors, such as metal-oxide-semiconductor transistors (MOSFETs). The first MOSFET was fabricated by Bell Labs researcher John Atalla in 1960. The design was based on Shockley’s field-effect theories.
Unlike the sandwich junction transistor, a MOSFET has a channel of either n- or p-type semiconductors. An electric field, which acts like a faucet to turn on and off the current in the transistor, is generated when voltage is applied to the channel. To achieve a high switching speed, manufacturers often adopt an epitaxial decomposition process during fabrication. This process also yields high breakdown voltages in transistors.
The Next Generation of Nano-scale Transistors
According to Moore’s law, the number of transistors per unit area in an integrated circuit (IC) doubles every couple of years. This push for miniaturization creates complexity for the next generation of transistors—from microelectronics to nanoelectronics. Today, researchers are aiming to downsize transistors to the nanometer scale.
With silicon-based transistors now operating in nanometer sizes, engineers face the design and fabrication challenges associated with dwindling physical space. For instance, a 100nm-sized MOSFET might experience short-channel effects that adversely affect the transistor’s performance. What’s more, nano-sized silicon transistors experience high channel leakage currents.
In 2016, a team at Lawrence Berkeley National Laboratory claimed to create the world's smallest transistor, measuring 1nm. Image used courtesy of Berkeley Lab
To address these limitations, researchers are now looking into nanotechnology materials to fabricate transistors. Recently, researchers have explored 2D ultrathin monolayer materials such as molybdenum disulfide to create more reliable transistors than miniaturized silicon transistors. Carbon nanotube and graphene are also promising materials to replace silicon in transistors.
Additionally, a team of researchers at TU Dresden recently reported the "world's first" efficient organic bipolar junction transistor. The team used highly-ordered, thin organic layers based on crystalline films of n- and p-type doped rubrene to develop organic bipolar transistors. These transistors may increase performance for data processing and transmission.
How has the evolution of the transistor affected you throughout your career? Share your experiences in the comments below.