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

Carbon nanotube transmission

  1. Sep 19, 2009 #1
    Hi friends,

    I've calculated the transmission spectrum of ideal carbon nanotubes for various applied voltages on them. I obtained the transmission spectra shown below for various bias voltages. However, I could not interpret why the transmission spectra changes so much with the bias voltage.

    I have to explain the variation in the transmission spectrum analytically. For the best, I need the transmission spectrum function which will have some physical parameters that can be related to the applied voltage. Could anybody please help? Any help will be appreciated.

    http://img268.imageshack.us/img268/4789/87209245.png [Broken]

    Last edited by a moderator: May 4, 2017
  2. jcsd
  3. Sep 21, 2009 #2
    How did you calculate the transmission spectrum in the first place?
    What chirality of carbon nanotube is the simulation for, as only 1/3 of nanotubes are metals and 2/3 are semi-conductors.
    Did you use DFT or Tight binding, or some other theory.
    I don't think you supplied nearly enough information.
  4. Sep 21, 2009 #3
    What theory are you using here?

    It's a metallic CNT because at 0 eV, the transmission is 2, which means there are at least 2 conducting modes even at no bias. (no bandgap)

    But I have no idea what you mean by "different biases" because to read this plot, you need to interpret the x-axis (the difference between the two contacts' electrochemical potential) as the BIAS.

    So what other "bias" do you mean here? Is there a third terminal? Do you mean to say there exists a "gate" voltage also?

    Another thing that's probably wrong: Why would I ever see a DECREASE in conductance
    for increasing voltage?

    Unless you are considering a complicated structure (something like a double-barrier etc..) it's nonsensical to see a decrease in conductance...

    Post some more information and let's see
  5. Sep 22, 2009 #4
    Thanks for your interest.

    * I used (6,6) CNT (metallic).
    * I used the software: ATK.2008 from Atomistix (Quantumwise).
    * There is no third electrode. The CNT is connected to bias voltage from its right and left ends.
    * x-axis is the energy of the electron whose transmission probability is given in the y-axis if it is injected by the supply voltage.
    * I applied 0V, 1V and 2V bias voltages between the ends of the CNT and then obtained the transmission spectra as shown in the figure.
    * I think, the important point is that the transmission spectrum shows a decreasing trend as the voltage applied between the two ends of the CNT is increased.
    * I'm trying to obtain an analytical formulation of the transmission spectrum function using the standard Landauer-B├╝ttkiker Approach.

    Thanks for your answers.
  6. Sep 23, 2009 #5
    What kind of scattering is assumed here?
    as I said there are several issues, fluctuating transmission doesn't make snse.

    and WHY would the transmission spectrum show a decreasing trend? Do you have a simple reason for that or is it just something that comes out of your code?

    Try NEGF-Landauer it'll be much cleaner.
  7. Sep 27, 2009 #6
  8. Sep 30, 2009 #7
    It looks kinda fishy that the conductances are not exact integers for vanishing bias. For a perfect nanotube, the conductance should simply equal the number of propagating modes. And as said before, it does not really make sense that the transmission/conductance fluctuates like that. And as for the finite bias results, does the software calculate the conductance using the linear response result
    [tex]G=G_Q \int_{E_F-eV/2}^{E_F+eV/2} dE T(E)[/tex]?
    This is not a good idea for as strong bias voltages as 1 eV, for which non-equilibrium methods should be used (and even they might not be that realistic).
    Last edited: Sep 30, 2009
  9. Oct 3, 2009 #8
    Thank you Saaskis.

    Yes, for a perfect 1-D system, the transmission has to be intiger multiples however, the simulated system consists over 700 carbon atoms hence, lots of non-idealities also affect the results I think. The software I used is already DFT-NEGF simulator which is said to perform accurate calculations even for high bias by calculating the transmission spectrum as shown above.

    I did not use the linear response approximation but the Landauer formula to calculate the conductance and the current, in fact the software I use uses Landauer approach to calculate the current and the conductance. When the transmission is like in the fig., the current shows a saturation regime in high biases.

Share this great discussion with others via Reddit, Google+, Twitter, or Facebook