Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Transistors and Quantum Physics

  1. Nov 30, 2018 #1
    My question is: What is the contribution of Quantum Physics to the discovery, of the transistor?

    In Adam Becker's book What is real? I read that, "the discovery of quantum physics in the early twentieth century led directly to the [discovery] of silicon transistors..." He implies that, the observation that atoms can be in two places at once, as formulated in QM led to the discovery of the transistor.

    I read in Wikipedia that "Quantum mechanics differs from classical physics in that energy, momentum, angular momentum and other quantities of a bound system are restricted to discrete values (quantization); objects have characteristics of both particles and waves (wave-particle duality); and there are limits to the precision with which quantities can be measured (uncertainty principle)."

    So which of these fundamental principles of QM "led directly to discovery of silicon transistors?"

    There is a clue in Wikipedia History of Transistor page: " John Bardeen eventually developed a new branch of quantum mechanics known as surface physics to account for the "odd" behavior they saw..." But does not explain what these "odd behavior" are. And it is not clear if Quantum mechanical concepts were used to discover the transistor, or if, on the contrary, John Bardeen developed a branch of QM from what he observed. The two are different.
     
  2. jcsd
  3. Nov 30, 2018 #2

    jambaugh

    User Avatar
    Science Advisor
    Gold Member

    • To understand the band gap structure which distinguishes a conductor from a semi-conductor and insulator you must have to look at the deBroglie relations from quantum mechanics which relate particle wavelengths to their momentum. You must also look at the quantized energies of the electrons in various elemental atoms, how the spacing of those energies, match up with or fail to do so the atomic spacing in crystals of material.

    • The doping of a semiconductor, and -for that matter- the very structure of the periodic table itself, is understood by the quantum theory of the dynamics of electrons bound within a central potential.

    • Understanding the behavior of the holes and conducting electrons in the resepective p-type and n-type semiconductors relies heavily on the Fermi-Dirac statistics of an electron.
    I can't say how much was empirically guided and then explained after the fact vs formulated predictions within QM. That would take some deep research, deeper than you'll likely find on a wikipedia article.
     
  4. Dec 1, 2018 #3
    Thanks for the nice answer.
     
  5. Dec 2, 2018 #4
    Limits are imposed by shortcomings of measuring instruments, not by fundamental principles. Heisenberg Uncertainty only applies to properties that don't pre-exist and are undefined until measured.
     
  6. Dec 2, 2018 #5

    tech99

    User Avatar
    Gold Member

    It is interesting that in WW2 there were properly made silicon diodes, and so by the 50s, people were definitely trying to make a crystal triode. Some FET-like devices were also made. One suspects that the theory followed on behind the experiment, and I believe the operation of the first transistor was not immediately understood.
    I have also found that some signal detectors around 1900 seemed to involve electron tunneling, and that was not understood at the time either!
     
  7. Dec 2, 2018 #6
    IIRC, some FETs were made in the mid-30s, but the German inventor could not get backing. Combination of rising social unrest, post-Depression caution and entrenched valve-based manufacture ? I've read this fairly-obscure work was not re-discovered until after US transistors reached production...
     
  8. Dec 4, 2018 #7

    analogdesign

    User Avatar
    Science Advisor

    Transistor history is a hobby of mine so I thought I would chime in here.

    At the very beginning, there was no theory (the quantum theory of solids didn't really get going until the 1920s) and things like the "cat-whisker" galena detector were complete mysteries. However, by the 1930s there was a strong research program in semiconductors and the theory was strong. These folks knew what they were doing. A lot of the work in crystal detectors (Germanium and Silicon) was done at MIT and Purdue (Lark-Horovitz's lab) during WWII. The famous Bell Labs group that invented the point-contact transistor was mostly working on other things during the war, but before the war they were deliberately trying to make a transistor (called a crystal triode in those days, as you said). They spent several years trying to make a field-effect device because it is the simplest conceptually and actually learned a lot about surface physics through their failures. But, at least for the first transistor, they knew exactly what they were doing when Bardeen made the thin base out of metal. There is a very interesting book about all this called Crystal Fire: https://www.amazon.com/Crystal-Fire-Transistor-Information-Technology/dp/0393318516

    I think you are a bit misinformed and this feels a bit like a conspiracy theory. The first patent for an FET was filed in the 1920s by Lillenfeld (a German-American). I think you may be thinking of Oskar Heil who patented a design during the depression. Unfortunately, the patents were based on theoretical concepts and no prototypes are known to have existed. There was no "entrenched valve-based manufacture" suppressing the innovation. In fact, making a FET is simple, in principle, but the details are very, very difficult. In fact, we didn't have the technology to do it (e.g. silicon purity, doping control, gate alignment, ion implantation) until the late 1960s to start making useful devices.

    Lastly, whatever you read was mistaken, the folks at Bell Labs trying to make a field-effect device knew quite well what Hiel was up to, and made sure their work did not violate his patents.
     
    Last edited: Dec 5, 2018
  9. Dec 4, 2018 #8
    "whatever you read was mistaken"
    My bad.
     
  10. Dec 4, 2018 #9

    rude man

    User Avatar
    Homework Helper
    Gold Member

    Heisenberg's Uncertainty Principle applies to all matter. It applies to an electron and it applies to a cannonball.
     
  11. Dec 4, 2018 #10

    analogdesign

    User Avatar
    Science Advisor

    No worries. I hope I didn't come across as harsh at all, since that wasn't my intention.
     
  12. Dec 4, 2018 #11

    analogdesign

    User Avatar
    Science Advisor

    I would also like to point out that Heisenberg is not directly relevant here. The theory of semiconductors flows directly from Pauli's Exclusion Principle. Once you buy the idea that fermions can't sit on top of each other (i.e. two particles in the same location can't share the same quantum numbers), the energy levels in a crystal really have no choice but to spread into bands. Then, if you solve the "particle in a box" first-year physics type problem with the twist that instead of a potential well the particle is in a periodic lattice, (i.e. a crystal), then the key concept of semiconductors, the bandgap, jumps right out.

    It seemed so simple in my first semiconductor physics class, but the leaps of intellectual faith it must have taken for these folks to get there is awe inspiring.
     
  13. Dec 4, 2018 #12
    Broiler research?? Purdue (for my Indiana friends!)
     
  14. Dec 5, 2018 #13

    analogdesign

    User Avatar
    Science Advisor

    Oops! Fixed.
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted