Brief History of Transistors

written by Prof. Sungsik Lee on 7 May 2017 (see more in Director’s Blog)

line

juliusBy definition, transistor is a combined terminology of transfer and resistor. This concept was come up with replacing the vacuum tubes in principle. And the modern type of transistors, which is currently known as MOSFET, was firstly patented in 1926 by Julius Lilienfeld, Polish-American physicist and inventor. He is also known as a former student of Max Planck who is one of fathers of quantum mechanics. At that time, due to lack of understanding on semiconductor and its manufacturing technologies, it was just an idea.

mosfets

Figures: typical layout and cross-section of MOSFETs. Here, it is found that the Logo of the LEE lab is composed of the MOSFET layout on the left-hand side. The animated images (An atomistic simulation for electron density as gate voltage (Vg) varies in a nanowire MOSFET) are from the wikipedia on Moore’s law.

In 1931, semiconductors theory was introduced by Alan Wilson, and the concepts of rectifications (known as a diode operation) were explained by Mott, Schottky, and Davydov. This was accelerating the understanding and manufacturing semiconductor devices in the level of laboratory and university.

3 guysOne of them was Bell laboratory which we all knew as the origin of telephones, transistors, CCD camera, and many other revolutionary electronic devices. In this lab, Shockley was leading a team in the 1930s, who was at that time struggling with a MOSFET development based on the original patent from Julius in 1926. While Brattain was still positive, Bardeen made his decision on why the team failed to make a successful MOSFET yet since he calculated trap density at the interface was too high to get any amplification factor in their fabricated MOSFETs. So, finally, Shockley made his decision to shut down the project. Alternatively, Shockley, Bardeen, and Brattain invented in 1947 a successfully working device which is currently known as BJT, leading to winning the Nobel Physics Prize in 1956. At the time of the invention, it was not a BJT, but a point-contact transistor with Germanium, looking very bulky and dirty. Although it was, it was for the first time, changing our life from very bulky vacuum tubes to tiny transistors (consistent with our 2nd research philosophy – fundamental breakthrough). Only after 1951, these guys got the BJT devices with a pnp or npn type. Note that a pn junction was discovered by Russell Ohl in 1940 at Bell lab, leading to the micro-electronic devices, including the pn-diode, np-solarcell, npn / pnp-BJT, and other applied FETs, such as JFET.

daewonSince Shockley and his team dropped the MOSFET project, another team in Bell lab was not so negative to it. So, they tried many many times to improve the interface quality in a MOSFET structure and eventually succeeded to get it as the first ever successfully working MOSFET in 1959. One of the key developers was Dawon Kahng, who is a Korean and died in 1992. And it was patented in 1963, leading to another revolution since BJT. Until the 1980s, BJTs were still dominant. Since early 1990, MOSFETs have been dominating our entire semiconductor and electronic industries, including Samsung, LG, Toshiba, and many others, because of its low power and complementary operations, so it has also been called the CMOS technology (CMOS = n-MOSFET + p-MOSFET).

Now, we have to think of what could be another revolutionary device in semiconductors besides those BJT and MOSFET. If you have an answer for this question, the future semiconductor industry may be yours. For this, the LEE lab is also trying to find the next type of transistors, so join us if you like.

*Note that source of photos is Wikipedia, the cross-section of MOSFET is from google image search, and layout of MOSFET is from the LEE lab. And this article is referenced with the website of Bell lab, Wikipedia, etc.

line