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

Schottky barrier Rectifying characteristics

  1. Feb 13, 2012 #1

    A Schottky barrier is said to have rectifying characteristics, which from what I can understand is bad ...but would the rectifying characteristics mean that current passing through the junction would be flat ....level is probably a better way to describe it rather than flat.

    I only ask because from what I have been reading, a low Schottky barrier will give ohmic contacts, and therefore a linear current. So, I would have thought rectifying characteristics would have been good ....using the analogy of the rectifier converting AC to DC?


  2. jcsd
  3. Feb 13, 2012 #2
    I don't quite follow what you are talking, do you mean schottky diode have more resistance? I am not sure that is true.
  4. Feb 13, 2012 #3
    Hey, thanks for the reply.

    I am trying to research reducing the barrier height in transistors. From what I have been reading it says that when the metal and semiconductor for a pn junction, the contact can have a Schottky barrier which exhibits rectifying characteristics.

    I may have this completely wrong, but I think to get good current through the device, you want to reduce the barrier height by unpinning the Fermi level ....I haven't even got to that part yet ...Although I do have Physics of Semiconductor Devices by Sze, which is by all accounts the bible :)

  5. Feb 13, 2012 #4
    I am bailing out, I only talk about the application of Schottky, not the semiconductor physics.
  6. Feb 13, 2012 #5
    Ha ...no worries ...I am not even sure I have it right myself.

    Many thanks.

  7. Feb 13, 2012 #6
    I think rectifying characteristics just means it conducts in one direction (forward bias) and does not conduct in the other direction (reverse bias)

    That may just be stating the obvious, although what else could it mean?

    What I know about Schottky diodes:
    instead of a PN junction they have a metal in contact with the semiconductor material, that way only one type of charge carrier needs to move for current to pass through the device. Because of this they operate with faster switching characteristics since you don't have to wait for electron-hole pairs to recombine and such (i think).

    maybe a schottky diode is not exactly what we are talking about here. If i assume schottky barrier means the voltage bias necessary for the junction to conduct, then I believe if that barrier becomes very small (accomplished by doping?) then the junction will behave like a ohmic load.

    Its probably obvious that I don't know much about this, but it is interesting. How is your research progressing?
  8. Feb 15, 2012 #7

    Thanks for the reply.

    You're right, rectifying characteristics means it allows current to flow in one direction, and blocks in the other. Which clears that up at least, and explains why you want ohmic contacts.

    The research is progressing anyway haha ...I am writing an interim report at the minute, trying to explain what I know. Fingers crossed haha.

    Again, thanks for all the replies!

  9. Feb 22, 2012 #8
    Schottky barriers are not completely bad. There are publications where researchers have reported better transistors using schottky contacts for source and drain.
    That said, the schottky barrier height gives rise to an additional contact resistance at the interface of metal and semiconductor. This resistance limits the current that can flow across the channel region of the transistor and as Ohmic contacts are needed.
  10. Mar 1, 2012 #9
    A Schottky barrier is the potential barrier that prevents charge carriers from being injected from the semiconductor into the metal or vice versa.

    For an ohmic contact, you want a very low (or no) Schottky barrier because you desire a linear current with no potential barrier in either bias direction. A Schottky barrier contributes to the contact resistivity, which is undesirable in an ohmic contact. In the thermionic emission regime, the Schottky barrier height is the primary impediment for current flow in an ohmic contact. However, if your underlying semiconductor is doped to a high enough degree (thinning the depletion region) or you are at a low enough temperature (not enough thermal energy to excite carriers over the Schottky barrier), thermionic-field emission (thermally assisted tunneling) or field emission (tunneling) will become the dominant transport mechanism.

    For a rectifying (Schottky) contact, you want to have as high current drive in forward bias and as low current drive in reverse bias as possible. A high Schottky barrier in a rectifying contact prevents leakage current in reverse bias. In forward bias, the quasi-Fermi level (in an M/n-type contact) will rise, increasing the carrier concentration in the conduction band exponentially. Eventually, a large concentration of carriers will be able to surmount the Schottky barrier, allowing current to flow. In reverse bias, the quasi-Fermi level will decrease (same case as before), decreasing the carrier concentration in the SC conduction band. This effect can be seen in the diode equation, where at large reverse bias the total current density simply goes to a small negative constant.

    A few other notes about the Schottky barrier:
    - The barrier HEIGHT does NOT depend on doping, only the shape (width) of the barrier, which can lead to tunneling.
    - Fermi level pinning can either cause a high OR low Schottky barrier depending upon where the pinning location (near the so-called "charge neutrality level") is. For example, if the charge neutrality level is near the valence band for an n-type semiconductor, a positive Schottky barrier is almost inevitable unless the pinning is extremely weak (very low density of interface states). On a p-type semiconductor, this would almost ubiquitously yield an ohmic contact (negligible/negative Schottky barrier) A good example of this is metal/Ge contacts. However, be warned that the causes of Fermi level pinning are very complicated. You'll be going down quite a rabbit hole if you seek to learn more about that. The simplest explanation is that interface states, coming about from many factors, cause the charge neutrality condition of the junction to change, thus modifying the band bending of the semiconductor, "pinning" the Fermi Level.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook