physics, materials, and history of transistors

[John Denker <jsd@MONMOUTH.COM>]

At 09:45 AM 5/8/99 -0500, brian whatcott wrote in part:

>  I derogate the efforts of the several following generations of
> physics/electronics researchers who COULD have given us the
> transistor but failed to do so for a conspicuously long period.

My, how easy it is for someone who
 * has never patented anything in his life,
 * doesn't understand the physics of the problem,
 * doesn't understand the materials science, and
 * doesn't know the history
to derogate the work of people who really have invented great things.

> In the opening years of the twentieth century,
> Fleming provided an effective application of the Edison diode
> effect to radio. DeForest was able to generalize this concept
> to the three electrode thermionic triode. He made this leap
> very fast and very effectively.
>
>    Despite the clearly analogous semiconductor diode device
> which was to be found in many households, NOBODY took
> the corresponding DeForest leap into the three terminal
> semiconductor for fifty years or so.

History lesson:

  US Patent     1,745,175
  Title Method and Apparatus for Controlling Electric Currents
  Filed         8 Oct 1926
  Issued  28 Jan 1930
  By Julius Edgar Lilienfeld
  http://www.ece.cmu.edu/afs/ece/usr/dwg/public_html/course/patents.html

It is basically a field-effect transistor and its application to
amplifiers.  (I've seen articles claiming it is a bipolar transistor, but
that's baloney.)

Physics lesson:

The device most analogous to a deForest triode is a field-effect
transistor.  The bipolar junction transistor that was invented "fifty years
or so" later is not "clearly analogous";  I would say it is not analogous
at all.

Materials science lesson:

Even though Lilienfeld and others saw the analogy, the materials of the
time did not permit construction of practical FETs.  The great advantage of
BJTs is that they can be constructed out of less-perfect materials.

> I don't disparage the magnitude of Bardeen, Brattain and
> Shockley's seminal discovery in 1948: I DO point out that
> the material for this experiment was sitting on the lab bench
> and in the domestic living room for the preceding forty years
> or more.

The invention of BJTs led to an industry which created the technology of
very pure materials.  This in turn permitted a leap "back to the future",
namely FETs.

BJTs had to wait for the principle to be invented.
FETs had to wait for the materials.

The contention that both the principle and the materials were "sitting"
around for fifty years is absurd.

--- jsd