Author Topic: calculating mobility of mos2  (Read 3082 times)

0 Members and 1 Guest are viewing this topic.

Offline sathyajit

  • Heavy QuantumATK user
  • ***
  • Posts: 31
  • Country: in
  • Reputation: 0
    • View Profile
calculating mobility of mos2
« on: February 3, 2015, 11:58 »
Hi,

I am trying to calculate the mobility of MoS2 with the help of the tutorial in

http://quantumwise.com/publications/tutorials/item/832-elastic-scattering-mean-free-path-mobility-impurity-scattering-in-a-silicon-nanowire

I want to know what changes I need to make to calculate the mobility of MoS2 and various other 2D materials ( ALL in intrinsic/undoped form) ?
This tutorial assumes doping and calculates mobility.

Thanks,
Sathyajit

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5565
  • Country: dk
  • Reputation: 93
    • View Profile
    • QuantumATK at Synopsys
Re: calculating mobility of mos2
« Reply #1 on: February 9, 2015, 00:47 »
Yes, the tutorial assumes doping and more specifically it computes the impurity-scattering limited mobility. If you are interested in the phonon-limited mobility, which is what matters in an intrinsic, undoped material, obviously you need to include phonons in your calculation, which is not so trivial (but will be possible in ATK 2015).

Although experimentally there is of course one specific mobility for a material, it's actually a rather tricky quantity to calculate. You can try to include all possible scattering mechanisms, and then the calculation is typically done using a Monte Carlo simulation, but it requires that you know all scattering rates. These can be computed from first principles (again, will be possible in ATK 2015) but often need to be calibrated against experiments. Or, you can try to pick a specific scattering mechanism "x" and compute the x-limited mobility . But the approach in these cases can be quite specific for the  mechanism - you couldn't use the approach in our tutorial for phonons, since you need to account for vibrations, but these can on the other hand be captured in an MD simulation, as presented in T. Markussen's presentation at our conference QuantumHagen 2014 (the slides are available on the abstract page).