Author Topic: Half integer transmission in hdf5 file  (Read 3040 times)

0 Members and 1 Guest are viewing this topic.

Offline kaihuang

  • New QuantumATK user
  • *
  • Posts: 2
  • Country: us
  • Reputation: 0
    • View Profile
Half integer transmission in hdf5 file
« on: April 21, 2024, 06:55 »
Hello,

I conducted the transmission calculation for a bulk system and attempted to plot the transmission concerning k by parsing the hdf5 file. In this instance, I initiated the DFT calculation, incorporating Spin-Orbit Coupling and noncollinear magnetic moments. GGA method and SG15-SO pseudopotential. Subsequently, the transmission was calculated. Within the output hdf5 file, I observed that the transmission is delineated in the data tree as follows:

- TransmissionSpectrum_0
    - transmission
        - 0
            - data
        - 1
            - data
        - 2
            - data
        - 3
            - data

For each 'data,' there is a 2D array. Two indices represent k points and energy, respectively.

However, two questions arise:

1. Is it correct that 0, 1, 2, 3 above represent the transmission of spin all, x, y, and z, respectively? Or do they represent spin all, y, z, and x? Or something else?
2. In the transmission->0->data, I noticed that the transmission plot matches the electron band structure well, which implies the validity of the result. However, I saw many half-integer transmissions, like 0.5, 1.5, 2.5, etc. It seems that each band only contributes 0.5 transmission channel rather than 1. So, I would like to inquire about the meaning of this dataset. Can I multiply it by 2 to get the right result?

Additionally, I attempted the transmission calculation without SOC, i.e., choosing just spin polarized. The transmission output in the hdf5 file is as follows:

- TransmissionSpectrum_0
    - transmission
        -0
            - data

This time, there's no spin projection. But again, each band contributes only 1 transmission channel rather than 2 (for the two spin eigenstates). Again, what's the meaning of this data in hdf5, and can I just multiply by 2?

Thanks,
Kai

 

Offline Anders Blom

  • QuantumATK Staff
  • Supreme QuantumATK Wizard
  • *****
  • Posts: 5571
  • Country: dk
  • Reputation: 94
    • View Profile
    • QuantumATK at Synopsys
Re: Half integer transmission in hdf5 file
« Reply #1 on: April 25, 2024, 08:08 »
Although the data in the HDF5 file is not obscured, the exact format is not something we document in detail. We wouldn't you just use atkpython to read the object from the file, then use the query functions that are well documented in the manual to provide the correct normalization depending on spin projection etc?