Author Topic: How to find the spin-up and spin-down electrons?  (Read 6735 times)

0 Members and 1 Guest are viewing this topic.

Offline bubble

  • Heavy QuantumATK user
  • ***
  • Posts: 59
  • Country: cn
  • Reputation: 0
    • View Profile
How to find the spin-up and spin-down electrons?
« on: November 3, 2015, 14:45 »
Hi,
   It is well known that, for magnetic systems (magnetic molecule or solid), the numbers of spin-up and spin-down eletrons are different. However, I found it is difficult to directly obtain the numbers of spin-up and spin-down electrons after a spin-polarized calculation in ATK. If I do not intend to perform a mulliken analysis, can I directly obtain the number of unpaired electrons after a spin-polarized calculation (i.e. the respective number of spin-up and spind-down electrons, and this should be the basic information for the output file)? Because in some cases, for example, I need to know the quantum number of the highest occupied states of the magnetic systems, where the spin-up and spin-down states are not degenerate.

In addition, even though a mulliken analysis is performed, can I directly obtain the number of unpaired electrons ? Because for a complicated system, to count the unpaired electrons on every atoms and then add all of them one by one is very unconvenient.
 
BTW, there may be a bug in 2015. I found that, in "Transmission Analyzer", only the plot of spin-up eigenstates can be given, where the spin-down eigenstate can NOT be directly shown.(At the same time, in the viewer, open the  corresponding  "properties"window, there is No choice about isosurfaces  for spin-down states. But the  isosurfaces  choice for spin-up state exists. )

Thank you!
« Last Edit: November 3, 2015, 15:20 by bubble »

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #1 on: November 4, 2015, 11:45 »
You can of course use some python scripting, this topic has been discussed on the forum already.
Also, the reference manual is a good source of information :)

The two script attached  can calculate the magnetic moment of your structure (which you can relate to the number of unpaired electrons) from an electron density and from the Mulliken population.

We'll check the TS issue.

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #2 on: November 4, 2015, 11:54 »
Quote
BTW, there may be a bug in 2015. I found that, in "Transmission Analyzer", only the plot of spin-up eigenstates can be given, where the spin-down eigenstate can NOT be directly shown.(At the same time, in the viewer, open the  corresponding  "properties"window, there is No choice about isosurfaces  for spin-down states. But the  isosurfaces  choice for spin-up state exists. )

everything works fine in 2015.0. You can plot the spin up and spin down transmission spectra.
Please attach your input/output files to check what is going on.

Offline bubble

  • Heavy QuantumATK user
  • ***
  • Posts: 59
  • Country: cn
  • Reputation: 0
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #3 on: November 5, 2015, 08:34 »
Quote
BTW, there may be a bug in 2015. I found that, in "Transmission Analyzer", only the plot of spin-up eigenstates can be given, where the spin-down eigenstate can NOT be directly shown.(At the same time, in the viewer, open the  corresponding  "properties"window, there is No choice about isosurfaces  for spin-down states. But the  isosurfaces  choice for spin-up state exists. )

everything works fine in 2015.0. You can plot the spin up and spin down transmission spectra.
Please attach your input/output files to check what is going on.
Thank you for your reply.
Althogh the transmission spetra are OK, please note that it seems that the 'spin-down eigenstate' in the Transmission Analyzer may have some problems,i.e., for a spin-polarized system, the spin-down eigenstate can not be well displayed through the 'eigenstate' option in Transmission Analyzer.
« Last Edit: November 5, 2015, 08:40 by bubble »

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #4 on: November 9, 2015, 14:17 »
I see now that something may be wrong.
thanks for reporting the issue. We will fix it and come back to you soon.

Offline bubble

  • Heavy QuantumATK user
  • ***
  • Posts: 59
  • Country: cn
  • Reputation: 0
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #5 on: November 14, 2015, 01:33 »
I see now that something may be wrong.
thanks for reporting the issue. We will fix it and come back to you soon.

Thank you for your reply, Martinez.
I have another question. Sometimes  the unpaired electrons in a magnetic devices are NOT an integer, or the influence  of some calculated parameter (such as the so-called 'temperature' of electrode) is significant . In these cases, how to plot the eigenstate of the device at fermi level or a peak in the bias window? Because it is difficult to fix the quantum number of the eigenstate. Is there another way to do this if the Transimission Analyzer has some problem.

Offline Subhban15

  • Heavy QuantumATK user
  • ***
  • Posts: 54
  • Country: in
  • Reputation: 0
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #6 on: November 30, 2015, 13:14 »
Hello!

The problem, probably, has not been resolved yet, because I am still not getting the transmission spectrum of down spin.

Is there a way out? I mean, can we somehow view the same by making some specific changes or using any script?

Thank you! :)

Offline Umberto Martinez

  • Supreme QuantumATK Wizard
  • *****
  • Posts: 479
  • Country: dk
  • Reputation: 26
    • View Profile
Re: How to find the spin-up and spin-down electrons?
« Reply #7 on: December 1, 2015, 08:51 »
No, it has not been solved. It will in a feature release, most probably in 2016 one.
In the meanwhile you can use all the corresponding analysis in a script as shown in the examples in the reference manual:
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.transmissioneigenvalues.html
http://www.quantumwise.com/documents/manuals/latest/ReferenceManual/index.html/ref.transmissioneigenstate.html