Shockley Semiconductor Laboratory

The Bell Labs team was indeed searching how to make a semiconductor triode (the word transistor has been coined after the discovery of the point-contact transistor).

However they had expected that they would find a method to make a device similar to the metal-semiconductor or metal-insulator-semiconductor field-effect transistors that had been invented by Julius Edgar Lilienfeld in 1925 and 1928 (Lilienfeld had never succeeded to make them reproducibly, because by his time the role of impurities in semiconductors was not understood, so he was not aware about the extreme requirements for fabricating reproducible semiconductor devices).

Moreover, Shockley cannot have kept "secret" the bipolar junction transistor invented by him, except perhaps for at most a few weeks, if he had felt any reason to verify his computations before making them public, to be certain that his new theory is correct.

Less than half of year after the discovery of the transistor effect by Bardeen and Brattain, both Bardeen with Brattain and Shockley have filed almost simultaneously 2 patents, for their transistors. As normal, the patents have been assigned to the Bell Labs and the filings must have been prepared some time before that, together with the legal department of the company.

In the real timeline there is no time left for "secrecy". Shockley certainly has not kept any useful information for himself, because the theory published by him almost immediately is what has taught the first few generations of engineers specializing in semiconductor devices, leading to an explosive growth of the industry.

The BJT invented by Shockley has been the basis of the semiconductor industry, while the point-contact transistor has remained a historical curiosity. Point-contact transistors have been used for a few years only because the first junction transistors were too big, so they could not be used at radio frequencies. However the size problem has been solved quickly, so the use of point-contact transistors has been abandoned.

The only reason why the discovery of the transistor effect in point-contact transistors is important is because this has made Shockley think about which is the cause of the transistor effect, making him develop his theory that he has published very soon, first in abbreviated form in 1949 in the "Bell System Technical Journal", then in 1950 in his book "Electrons and Holes in Semiconductors".

Now when Shockley is mentioned everybody first comments that even if he has been a genius physicist in other aspects he was a stupid human being. Both aspects are of course true and it is not at all unusual for humans to do both very good things and very bad things. However his stupidity did not really have any harmful effects, besides many people feeling insulted by him, while his theoretical work, which has taught so many engineers how to do their jobs, remains a really important heritage, which has greatly contributed to human progress.

The natural thing about controlling the depletion zone is a field-effect transistor. That is why FETs, both metal-semiconductor FETs (with Shottky barrrier at the gate) and metal-insulator-semiconductor transistors have already been invented long before WWII, in 1925 and 1928, by Julius Lilienfeld (who has never succeeded to fabricate them, because the need and the means for controlling impurities were not known at that time).

The fact that you can control the current through a depleted region by injection of minority carriers has never occurred to anyone before Shockley, who has thought about this only because he has searched for an explanation of the weird effect discovered by chance by Bardeen and Brattain when experimenting with multiple point contacts to a semiconductor crystal.

So the discovery of the bipolar junction transistor had nothing easy and natural about it.

The sequence of events at the Bell Laboratories has been:

1. The need of detector diodes for radar signals during WWII has lead to the development of growing crystals of germanium and silicon that were free from impurities, unlike any material that has ever existed before.

2. After WWII, using the materials produced by the radar research, i.e. pure germanium crystals, Bardeen, Brattain and other colleagues have made experiments with the purpose of discovering a method for making semiconductor triodes like those invented by Lilienfeld, i.e. FETs.

3. The experiments for making FETs have failed (a method of making FETs has been invented only a few years later, also by William Shockley). Nevertheless, during the experiments a weird unexplained effect has been discovered, which was named as the transistor effect (the name is not very appropriate, because a transistor is not a trans-resistor, but a trans-conductor, but that was not understood at the beginning).

4. Shockley has searched for an explanation of the origin of the transistor effect. This has lead to a theory of the semiconductor P-N junction.

5. After understanding how semiconductor P-N junctions really work and which is the role of the minority charge carriers in their properties, Shockley has invented first the bipolar junction transistor (BJT) in 1948 and then the junction field-effect transistor (JFET) in 1951, which have been the most important semiconductor devices for about a quarter of a century, until the improvements in the fabrication technology of the MOS transistors have made MOSFETs the most important semiconductor devices.

Moreover, by the time of the discovery of the transistor effect, there were no junction diodes. The fabrication of junction diodes has begun later, as a byproduct of the bipolar transistor technology.

Before transistors, there were point-contact diodes for radio frequencies, made with silicon, germanium, galena and many other semiconductor crystals, and also rectifier diodes made with selenium or copper oxide.

The theory of these diodes was not understood well at that time, but they are metal-semiconductor diodes, i.e. Schottky diodes, not junction diodes. Shottky had given a theory of the metal-semiconductor diodes, but the theory was not helpful for designing such diodes, because it depended on material parameters that were not known at that time, so all practical diodes had been discovered empirically.

Shockley has been the first to give a good theory of the junction diodes, after the discovery of the transistor effect by Bardeen and Brattain. With the Shockley theory, you could design P-N junction diodes, by controlling the impurity levels in semiconductor crystals (instead of searching experimentally for random metal-semiconductor pairs with rectifying properties, like before his theory).

Shockley was like this. Repeatedly in his life we see this pattern of people scrambling to get away from working with, or under, him. It's unusual for people to dislike someone that much: that's not common. If it was one anonymous guy over his whole career lodging a complaint, we might ignore it. But so many people voted to not be around Shockley, with their feet and with their employment. That's not common for a scientist, Democrat, Republican, entrepreneur, or anyone really: there must have been some unpleasant quality about Shockley that made people not want to be around him, as this pattern repeated over his whole life shows.

If anything, now that he's (seemingly) an incompetent president of the US, he's likely to create bad times.

That may be true (plenty of examples), but it may also be an outlier. It might be more common for talented individuals to work cooperatively, in a way that leads to breakthroughs, but without anyone trying to steal all the credit. I emphasize this is just a theory, in a field that cannot be reduced to science.

I've met many talented people over decades, in effective teams, but worked with only one classic martinet -- Steve Jobs. To hear Steve tell it, he was the reason for Apple's success. But he didn't design anything -- he was a salesman, not an engineer. In fairness, his incompetent replacements were in every way worse.

This thread is perfectly timed. We're about to see another martinet try to steal credit for the accomplishments of others, while denying responsibility for errors along the way.

I wonder if this applies to computing where in the early days RAM and clock cycles were limited so programmers and engineers had to be more creative and do more for less, where now it's just buy another GB RAM stick for $50USD and try again.

I remember a youtube video with bwk (Brian Kernighan) and he was saying that when they were doing UNIX and computing in the 70s it was easy for them as they had all the low hanging fruit to pick where now you have to climb higher to get the fruit?

It's why I don't do video calls with my coworkers.