William Shockley and Solid State Electronics
Written by Historian   
Thursday, 04 February 2021
Article Index
William Shockley and Solid State Electronics
The transistor


The Junction Transistor

Shockley now knew that a solid state amplifying device was possible and he needed to find a way to make it practical.

For five weeks after the invention of the point contact transistor he thought hard and on New Year's Eve 1947 he filled 19 pages of his lab book with basic ideas. By the 23rd of January he had the construction of the junction transistor more or less completed.

Instead of a pair of wires to inject holes a third region of semi-conductor was used - creating an NPN sandwich. The first region acts as the source or emitter, the middle region controls the current through the entire device and acts as a control grid or base, the third is the collector. A small current flowing between base and collector causes a much larger current to flow between the emitter and collector - this makes the transistor a current amplifier.


The junction transistor is relatively easy to make once you know how to handle the semi-conductor materials. Shockley, Bardeen and Brattain applied for a joint patent in 1948 on the technology.  Bell licensed the design to any company who would pay $25,000 - a bargain given that they also offered help on getting the manufacturing right. The licence was waived to any company manufacturing hearing aids - a concession to Alexander Graham Bell's original work with the deaf.  The cheap licence was partly altruism and partly to stave off an anti-trust action like the one in progress against AT&T.

The rest of this part of the story you will already know. In 1953 the first transistor hearing aids were produced, in 1954 the transistor radio was invented and in 1956 IBM started to build the transistor replacement for the SAGE air defence computer. The military wasn't too keen to have its brand new 55,000 valve computer replaced and so the first transistorised computers were actually brought to the market by Univac and Philco in 1957.

From there the pace increased rapidly. Silicon took over from Germanium as the semiconductor of choice and the integrated circuit was invented. But the junction transistor was the first practical step. Without Shockley the discovery would still have happened. Other groups were only months behind the Bell Labs team - but for once there is little doubt about who was first.


At first it wasn't obvious how to package transistors to make best use of their size advantage. Early transistor modules featured lots of empty space! 

In 1955 Shockley did the almost unheard of thing for a Bell Labs employee - he left to form his own company. A small shed housed Shockley Semiconductors in the area south of Palo Alto that would become known as Silicon Valley. He hoped to make use of his knowledge of semi-conductors to build better transistors but he took a wrong turn in choosing to work with a four-layer PNPN structure. The four layer switch was generally thought to be a dead end at the time - although eventually it turned out to be quite useful - but Shockley would have been better off working on silicon rather than germanium transistors. The operation was far from a success but from it grew many of the companies that eventually filled Silicon Valley.

Shockley shared the Nobel prize with Bardeen and Brattain  in 1956. A Swedish journalist tried to phone Shockley early in the morning but the line was so bad he thought that someone was trying to play a joke on him. Soon afterwards it was announced on the radio and Shockley drove to the lab and took everyone out for a champagne breakfast.

Beyond Success

Schockley was a talented engineer but his management style was naive and very strange. He made salaries public in an attempt to stop secrets. When a delay made him suspect sabotage he insisted that an employee took a lie detector test. In 1957 a group of engineers including Robert Noyce, Gordon Moore and Jean Hoerni left Shockley Semiconductors to start Fairchild Semiconductors. Shockley regarded the group who left with contempt - he usually referred to them as "the Traitorous Eight".  In 1960 the company was purchased but Shockley stayed on as a consultant. He lectured at Stanford and eventually returned to Bell as a consultant in 1965.


This would be more or less the end of the story - a Nobel prize, the co-inventor of the transistor and the creator of the Silicon Valley phenomenon - but Shockley strayed outside physics and engineering.

He believed in and promoted a form of eugenics - the idea that humans should be bred to improve the species rather than allowed to propagate haphazardly.

I'm sure that Shockley thought that the idea was entirely logical and not in the least bit threatening but any theory even close to eugenics has the tendency to be branded as some sort of right wing search for the master race or at the very least an unpleasant form of racism.

Shockley had to put up with his lectures being disrupted by demonstrators and many threats to his life. Yet he always stood firm for the right of his opponents to say what they were saying. At one demonstration where the students were chanting "Off Shockley" their microphone went on the blink - he fixed it for them!

For all his benevolence and paternalism, his suggestions - paying people with a low IQ and genetic diseases to be sterilised and setting up sperm banks for high IQ donors - seem distasteful even if you treat them as the outcome of a rigorous logic. But Shockley wasn't a geneticist and was an amateur in the field - he was accused of using pseudo science to back up his political views.  In 1980 he brought a libel suit against a newspaper who had claimed that he was a Nazi. He won but was only awarded $1 in damages.

He tried to be adopted as a Republican candidate for the senate  in 1982 but only got 8000 votes. As time went on he was more and more willing to talk about his theories of eugenics and less and less about the transistor, electronics and solid state physics. In time he became  notorious rather than famous for what he had done.



Computer Languages by Committee - the 1960s

The 1960s saw the growth of interest in computer languages but, unlike today where successful languages are often designed by small teams and even single-minded enthusiasts, this was the decade of the [ ... ]

Alan Sugar - Amstrad and the CPC

In the UK Alan Sugar's Amstrad was the first company to look at computing with an eye to producing something cheap and cheerful and in doing so revolutionised the computer marketplace.

Other Articles






Last Updated ( Thursday, 04 February 2021 )