1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Still a semi conductor?

  1. Mar 29, 2005 #1
    Two students performing a Hall voltage measurement using germanium as a semiconductor reversed the magnetic field while holding the bias
    current constant but did not get opposite signs for the Hall voltage. The magnitude of Hall voltage detected using a voltmeter was the same as it must be. Based on the sign observed, they could determine the doping of the semiconductor from the a) Lorentz force law and b) direction of the current which they controlled through the bias current. Appropriate procedures were followed and several attempts were made after reconstructing the circuit to rectify the signs. The experiment was attempting to illustrate the basic properties of semiconductivity. This part of the experiment attempted to calculate the Hall coefficient, type of doping (n or p) and the carrier mobility.

    Part of experimental science is dealing with equipment breakdown and other
    non-idealities. When this happens they had to think carefully about the extent to which their results are affected, and what they could conclude about the physics they were investigating.

    Their numerical values for the Hall coefficient and carrier mobility were very close to the literature ones.

    What are they able to conclude about the bandgap theory of conductors from their experiment? More specifically, were they able to say whether the samples tested are n-type or p-type semiconductors, or if they're semiconductors at all? Are their Hall voltage measurements discredited (and
    why)?

    Here' s what I think: The Hall voltage calculations should be valid. The agreement with literature values shows this BUT they cannot say anything about the bandgap theory..i.e. they are unable to determine whether the two samples of germanium are of different doping. They are unable, as a result, to determine whether the samples are infact semi conductors. The sign part could be due to equipmental problems but they reconstructed the circuit so I would say that this is not the problem.

    Am I correct? Where and how am I messing up the interpretation of this experimental problem? With the exception of instrumental problems, what could be responsible for this sign changing flaw? :confused:

    James
     
  2. jcsd
  3. Mar 29, 2005 #2

    Gokul43201

    User Avatar
    Staff Emeritus
    Science Advisor
    Gold Member

    The wording of the experimental procedure is a little careless making this problem interpretation dependent.

    I'm assuming the following interpretation : A does a Hall measurement on sample 1 with an upward B-field and bias voltage Vb. B performs the same measurement on sample 2 (contacted identically to sample 1) with a downward B-field but the same bias voltage Vb (with the same polarity). A and B measure equal Hall voltages in magnitude and sign.

    If this is the question, then it's not hard to see where the change in polarity relative to a given Bz orientation comes from - the type of majority charge carrier, of course, which tells you ...

    If, on the other hand - and I think this is not what the question intends - a Hall measurement is performed on a single sample while switching the direction of Bz and no polarity change is obseved, then there has to be something fishy about the experimental set-up (like, for some crazy reason there's a rectifier in the output ... or I'm missing something obvious).
     
  4. Mar 29, 2005 #3
    Thanks for the reply. I just asked around and the second erxplanation you gave is the one applicable here. The polarity did not change. Can you elaborate a bit on what you said about the rectifier? How would this affect the experimental setup? Are they still able to conclude anything about whether it is n or p type based on this?

    James
     
  5. Mar 30, 2005 #4
    Anything guys? I am really lost.

    James
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook

Have something to add?