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

Highly degenerate semiconductor uses

  1. May 2, 2012 #1
    "Hypothetically" lets say that I have a material that intrinsically behaves like a highly degenerate p-type direct gap semiconductor (significant p-orbital contribution to the DOS at and above the fermi level). Can anyone think of what in the world such a material may be useful for? I know that such doping is useful for ensuring a good ohmic contact with a metal, but what devices might that be useful for? Keep in mind, this is intrinsically heavily p-type, I didn't dope it to get it that way and so there is no gradient of p-type carrier (which is typically desirable for pn or pin diodes).
    Thanks anyone who reads this.
  2. jcsd
  3. May 3, 2012 #2
    Ah! The wafer spammers have followed me here!
  4. May 3, 2012 #3
    If intrinsically it had its Fermi energy lying deep inside the conduction band, it will not be called a semiconductor in the first place.
  5. May 4, 2012 #4
    Thats a tough call because, unlike most metallic band diagrams, there is a largish (around 1.6 eV) direct band gap above the the fermi level, valence states, and some of the p-band conduction states. As an intrinsic crystal, it really does more closely resemble the band diagram for a highly degenerate semiconductor than it does a metal, but I'm at a loss as to what such a thing could be useful for...
  6. May 4, 2012 #5
    If it has a direct band gap, then it should be good for LEDs. If you can inject electrons above the band gap, you should get plenty of 1.6eV (780 nm) photons coming out. Unfortunately, that is just below red in the near infrared and technologically not all that interesting.
  7. May 4, 2012 #6
    Now that's an interesting idea. I had been thinking in terms of absorption, I hadn't been thinking in terms of emission. It might actually be interesting as a solid state laser active medium...
  8. May 4, 2012 #7
    GaAs is in the same ball park, and that also has a direct band gap


    However, I am afraid that to make any useful semiconductors out of your mystery compound, you might have to find a way of filling up the valence band (e.g. by doping), or by using it in heterostructures.
  9. May 4, 2012 #8
    In terms of a p-n semiconductor laser, I would agree with you. However, what about a solid state laser (probably optically pumped...)? The low fermi level likely wont help the efficiency, but once the population of electrons is inverted, that direct gap should still lead to lazing as the electrons relax back down, shouldn't it? Thanks M Quack, I appreciate being able to bounce ideas off someone.
  10. May 4, 2012 #9
    Ideally for lasing you want a metastable excited state so that stimulated emission dominates spontaneous emission. Depending on the matrix elements that might not be the case for a direct band gap - to be checked. For example one could try to optically excite the material (as you propose) and then measure the decay time of the luminescence. Then compare to GaAs or similar materials used in semiconductor lasers. I am not sure if you could directly measure the carrier lifetime as your material should be conducting anyways.

    I would imagine that direct injection of electrons is much more efficient than optical pumping. You need a higher photon energy/band gap to excite electrons across the gap. So if you want to pump with a semiconductor source (LED) you need CdTe, AlGaAs or similar. Then you have to ask yourself why do the intermediate step, and why not use AlGaAs as laser material in the first place.

    I am not an expert, all these arguments are seat-of-the-pants.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook