The Lilienfeld transistor, and evidence of a Bell Labs coverup

Talk about an idiot ... quoting an entire long article for a three line response.

Jim

(snip)

I should have top posted to save you any effort.

.....

If you look carefully at the Nobel citation, you find that the prize was awarded "for the discovery of the transistor effect." That meant, during the 1950's, quite specifically the long lifetime and relatively free motion of minority carriers in indirect-bandgap semiconductors such as Si and Ge. The prize was not awarded for the invention of the transistor. It seems that the Nobel committee was not inclined to award the prize for inventions until

The post is an interesting pointer to parts of the history. NIH was certainly an element of the Bell Labs culture. However, note that a functioning radio (I presume that means a receiver) does not imply an active, gain-producing device. A rectifying detector is all that is required*. The true test is a self-sustaining oscillator, as the Bell Labs group certainly recognized. John Bardeen kept an audio oscillator made from an early point contact transistor and was proud to demonstrate it as late as the early 1980s.

While the field-effect transistor is now obviously the right way to do electronics, it is also now clear that the technology had to pass through the bipolar phase, to simply generate enough income to pay for the technological development that made FET technology (MOS, J, or MeS) feasible.

• There was an educational "transistor radio" kit sold not too long after the Regency transistor radio came out. My dad bought one and we assembled it (I was about 7 at the time). I found it again a few years ago; it had a coil antenna, tuning capacitor, crystal earphone, and a couple of other miscellaneous caps, but no battery. It was obviously a classic crystal radio which used the E-B junction of an alloyed Ge transistor as the detector and probably left the collector open.

- Bill Frensley

and

technological

was

The transistors could have been rejects that had open collectors. They would have got them cheap, or free.

Certainly. But the FET also has a "transistor effect," and it dates from 1925.

I'm mostly speculating on what might have happened had the Nobel Committee been aware of the Lilienfeld story, or at least aware of the references that Shockley was apparently concealing at the time. Perhaps Lilienfeld (who was still alive in 1956) would have had a share. Or perhaps not, if the Prize involved exclusively semiconductor theory (which Lilienfeld did not have.) There is both a BJT "transistor effect" and an FET "transistor effect," and Lilienfeld had discovered the latter and developed working devices. (Note that Lilienfeld was a high-vacuum physicst, and also the devloper of the modern electrolytic cap.)

In any case, this all appears to be a story of what happens when the dishonesty of businesses and the conflicts of interest play a part in physics research. Otherwise honorable scientists turn all weasley, and simple history is manipulated and distorted by the marketing department.

A crystal radio? Uh... that's silly. Think about it: if Lilienfeld merely built a standard crystal radio in 1930 or thereabouts, it makes absolutely no sense that he'd "show it around," or that Stockman would include such a story in his review-letter pertaining to amplification by crystals durnig pre-Bell-labs history. You honestly think that Stockman was talking about a crystal detector hooked to headphones? Really? Also, I suspect that you didn't bother to check out those patents. One of Lilienfeld's figures depicts a modern transistor radio: a detector with multi-stage transistor amplifier connected to a loudspeaker.

I guess it boils down to two possibilities: whether Shockley and Johnson were liars, or whether Lilienfeld the liar. The legal deposition shows that Johnson committed a lie of omission in Physics Today, while Shockley et. al. committed a scientific sin: carefully avoiding any reference to the physicist whose earlier work they were replicating. Their actions are perfectly acceptable in the world of business and politics, but in physics such deceptive tactics are the very opposite of science and are quite disgusting.

Irrelevant, since self-sustaining crystal oscillators were known long before transistors. So were gain-producing crystal circuits. See "crystadyne" and the Lennyr website with zinc oxide amplifiers. These were 2-wire parametric amplifiers, not transistors. Lilienfeld's devices were 3-lead MOSFETS built by thin-film deposition.

technological

I think you have it backwards. The above is shown false if the following is true: if Lilienfeld had actually built stable gain-producing devices before 1930, and if Bell Labs *intentionally* steered away from FETs *because* they could not be protected by patents; instead investing in undeveloped BJT technology rather than picking up the trail of Lilienfeld's already-developed thin-film devices.

There is no reason why a company can't make scads money selling non-tube miniature amplifiers built from a public domain invention. Patents only let them sue competitors, as well as giving them major propaganda cred for "inventing THE transistor" rather than merely "popularizing Lilienfeld's device." If really interested, read those references. The Bell Labs deposition apparently states that Lilienfeld FETs were sucessfully built by Shockley and produced significant gain ...and that nearly all of Bell Labs' FET patent claims were then shot down because of Lilienfeld's prior art already in the public domain. Fewer of their BJT claims were shot down by Lilienfeld's device.

On the other hand, if Bell Labs had simply started selling Lilienfeld public-domain MOSFETs, Shockley and crew might have been out of a job, and today we might be missing a very major chunk of solid state theory. Or perhaps it the theory would have been developed by honest physicists outside Bell Labs, not by those whose conflict of interest (read profit motive) causes them to intentionally deceive fellow scientists.

((((((((((((((((((((((( ( ( (o) ) ) ))))))))))))))))))))))) William J. Beaty Research Engineer snipped-for-privacy@chem.washington.edu UW Chem Dept, Bagley Hall RM74 snipped-for-privacy@eskimo.com Box 351700, Seattle, WA 98195-1700 ph425-222-5066 http//staff.washington.edu/wbeaty/

I don't think it's generally considered a "transistor effect"; that's why the letters "FE" (Field Effect) in the name FET are there. The Bell labs folks came up with the name "transistor" (transfer resistor), and they meant specifically the sort of action that takes place in a BJT. They were the first to realize the significance of minority carriers in semiconductors. FET's don't use a minority carrier effect.

technological

Good point.

Separate topic: I'd always wondered if nobody had ever bothered to try building a Lilienfeld MOSFET. Some variants should be easily achievable by hobbyists. But this 1995 article notes two recent successful attempts (I mean besides Shockley's apparent success in 1948.)

So now I can't be first. Rats. (Not that I actually would have, seeing my own level of ambition!)

((((((((((((((((((((((( ( ( (o) ) ) ))))))))))))))))))))))) William J. Beaty Research Engineer snipped-for-privacy@chem.washington.edu UW Chem Dept, Bagley Hall RM74 snipped-for-privacy@eskimo.com Box 351700, Seattle, WA 98195-1700 ph425-222-5066 http//staff.washington.edu/wbeaty/

Very interesting thread. Thanks for posting.

BTW, you might want to add a colon to your url. It works better as

Regards,

Mike Monett

Antiviral, Antibacterial Silver Solution:

technological

Has anyone mentioned that around that time, both before, during and after Lilienfeld, that some radio amateurs, not being rich enough to afford a tube, fiddled with point contactS (yes, plural) on crystals? Various configurations were used, including pressure on one or more of the contacts, as well as various *biasing* schemes. And there wer claims of gain . . but . . reproducibility was sadly lacking. Adding true and verified stories with dates would enrich the history!

As I pointed out in the 2004 post referenced below, AT&T had a need for millions of audio amplifiers, and as it would have cost them billions of dollars if they had engineered transistors into their equipment earlier, they conveniently waited until the Lilienfeld patents expired before they had Bell Labs "invent" the transistor.

And a can be seen from the patents, contrary to the claims of the Q.M. and S.R. Zombies, Q.M. and S.R. played no role in the invention of transistors.

And as can be seen by the notebooks of the so-called Bell Labs "inventors" they fumbled their way into getting their point contact transistor to work, and no magical Q.M. computations were involved.

But did you have to requote it all?

Late at night, by candle light, "Bill Beaty" penned this immortal opus:

I vaguely recall an Amateur Scientist article from the mid-70s that involved baking some metal oxide on glass slides using a strip of aluminum as the gate. Didn't seem very useful but nice for a science fair or something.

- YD.

Robert Baer wrote: ...

Hi Robert

Maybe they were negative differential resistance devices (tunnel diode effect before they were "invented")?:

Today radio amateurs also make home made tunnel diode amplifiers and oscillators:

By Nyle Steiner K7NS 2001: Zinc Negative Resistance RF Amplifier for Crystal Sets and Regenerative Receivers Uses No Tubes or Transistors:

More nostalgic stuff:

The Wireless World and Radio Review. October 1, 1924 and October 8,

Radio News, September, 1924, pages 294-295, 431: The Crystodyne Principle:

Radio News, September, 1924, page 291: A Sensational Radio Invention By HUGO GERNSBACK:

/Glenn

Yep! Below is a physics paper about making Galena point-contact transistors. The researchers found that they could get some gain only if they used a freshly-cleaved Galena surface, some very sharp metal point-contacts, and most importantly, 0.05mm electrode spacing (that's

Crystal Triode Action in Lead Sulphide, P. C. Banbury, H.A. Gebbie, C. A. Hogarth Proc. Conf. Semi-conducting Materials pp78-86. H.K. Henisch (ed), Butterworth's sci pub LTD 1951.

The required small spacing is indirect evidence that early hobbyists weren't seeing a bipolar transistor effect. Instead they were probably creating a resistive divider inside the material which biased their diodes and reduced or removed the turn-on voltage. That, or they were creating parametric amplifiers (based on the negative resistance operating regime.)

On the other hand, here is the website of H. P. Friedrichs, AC7ZL, with photos of home-built transistors, from his book "Instruments of Amplification."

Home built transistors (and Triode vac tubes)

Buy that book from Lindsay:

Instruments of Amplification

((((((((((((((((((((((( ( ( (o) ) ) ))))))))))))))))))))))) William J. Beaty Research Engineer snipped-for-privacy@chem.washington.edu UW Chem Dept, Bagley Hall RM74 snipped-for-privacy@eskimo.com Box 351700, Seattle, WA 98195-1700 ph425-222-5066

SET YOUR SIG BACK TO NORMAL.

Hmmmm. Maybe someone should see whether quantum-dot devices can be explained in semiclassical terms, or by using SED.

People who expose to doubt only their opponents' position would make good politicians. But those who dare to doubt both the contents of their textbooks as well as their own long-held knowledge, ah, the should consider a career in the sciences.

Now if you want to expose QM to skepticism, not only can you read Einstein, but also go and check out Dr. Willis Lamb of "Lamb Shift" fame. At one point he had a running bet with other physicists about the non-existence of EM quanta. He was betting that photons would turn out to be a delusion. See:

Anti-Photon

And also Lamb's book:

Interpretation of Quantum Mechanics

((((((((((((((((((((((( ( ( (o) ) ) ))))))))))))))))))))))) William J. Beaty Research Engineer snipped-for-privacy@chem.washington.edu UW Chem Dept, Bagley Hall RM74 snipped-for-privacy@eskimo.com Box 351700, Seattle, WA 98195-1700 ph425-222-5066

Tom Potter wrote: ........

No quantum mechanical "computations" were required, but the insight provided by band theory was essential to understanding and interpreting the experimental data that led to working devices. What passed for theory in that era was (of necessity) remarkably qualitative, with lots of attention to highly oversimplified (but solvable) models, then argument by analogy.

What I find most interesting about the Lilienfeld patents is the complete confusion about conduction mechanisms in solids. His concepts are apparently drawn mostly from ionic conduction in aqueous solutions. Of course, this should not be surprising, since that was the only halfway accurate model of conduction available until 1928 (the date of the last application) with the publication of Felix Bloch's paper.

In contrast the semiconductor researchers in the late 1930s and 1940s (not just at Bell Labs) had a conceptual framework which allowed them to sort out the difference between conduction due to a large density of not-especially-mobile charges (metals) from that of small densities of highly mobile charges (semiconductors). The later are what one needs to make an active device (current valve). The fact that the charge carriers that move most easily between the material then in use (germanium) and metallic contacts are positively charged (the holes) led them to percieve that the quantum insights were particularly crucial.

- Bill Frensley

Can't.

--

• Required crap appended to avoid restrictions imposed by brain +

• damaged idiots.

• Server Response: '441 Posting Failed (Rejected by POST filter)', +

• Port: 119, Secure(SSL): No, Server Error: 441,

• Error Number: 0x800CCCA9

It is interesting to see that in response to my statement: "Q.M. and S.R. played no role in the invention of transistors." William J. Beaty tries to create a strawman with his statement: "Hmmmm. Maybe someone should see whether quantum-dot devices can be explained in semiclassical terms, or by using SED."

or else he wrongly thinks that quantum-dot technology preceded Lilienfeld's "invention" of the transistor.

I must also point out that "explaining" an effect, or fitting it into some model, folows the observation of the effect.

It is also interesting to see that William J. Beaty asserts that EM is quanta, rather than the fact that high Q resonant systems, like atoms, absorb action only at resonance.

In other words, massive systems are quantum by virtue of their resonance and high Q, and they will absorb action at that frequency.

For example, if EM were quanta, there would be no Doppler effect.

Also note that a high Q, resonant (Quantum) system can be brought into resonance with an off frequency, non-quantum, EM wave with the appropriate relative velocity.

-- Tom Potter