QuantumATK Forum

QuantumATK => General Questions and Answers => Topic started by: Cyrille on November 12, 2015, 13:54

Title: fixed spin in transport calculations?
Post by: Cyrille on November 12, 2015, 13:54

Dear all

I have recently managed to perform fixed-spin calculations (angle of the spin set to a given value) for an Iron wire with SOC (OMX pseudo/basis)
For the moment it is only a "standard" DFT calculation with "BulkConfiguration"
I could set the spin orientation by using the following script (taken from FeMgO.py)
==================================================
B0 = 5*eV
theta=30*Degrees   
#----------------------------------------
# Intial spin
#----------------------------------------   
spin_list = [(Iron, 1, theta, 0*Degrees)]
initial_spin = InitialSpin(scaled_spins=spin_list)
spin_list = [(Iron, theta, 0*Degrees)]
fixed_spin = FixedSpin(spin_list, B0)

bulk_configuration.setMagneticField(fixed_spin)
=========================================================
The calculation works perfectly fine. In particular I am able to follow the evolution of the band-structure with the angle between spin and wire axis...

I now try to do the same type of calculation but in transport (DeviceConfiguration) in order to study the evolution of the transmission with the angle of the spin (AMR),
But my calculation crashed.

Is the option .setMagneticField(fixed_spin) available in DeviceConfiguration?

Thanks in advance

Cyrille

Title: Re: fixed spin in transport calculations?
Post by: Jess Wellendorff on November 13, 2015, 11:16

Hi Cyrille. I believe the FixedSpin class so far only works for BulkConfigurations, but I will need to check up on it to be sure...

Title: Re: fixed spin in transport calculations?
Post by: Quhe on March 21, 2016, 15:35

Has this problem been resolved?Is it possible to do a reliable transport calculation with SOC (NEGF+SOC) using ATK now?

Quote from: Cyrille on November 12, 2015, 13:54

Dear all

I have recently managed to perform fixed-spin calculations (angle of the spin set to a given value) for an Iron wire with SOC (OMX pseudo/basis)
For the moment it is only a "standard" DFT calculation with "BulkConfiguration"
I could set the spin orientation by using the following script (taken from FeMgO.py)
==================================================
B0 = 5*eV
theta=30*Degrees   
#----------------------------------------
# Intial spin
#----------------------------------------   
spin_list = [(Iron, 1, theta, 0*Degrees)]
initial_spin = InitialSpin(scaled_spins=spin_list)
spin_list = [(Iron, theta, 0*Degrees)]
fixed_spin = FixedSpin(spin_list, B0)

bulk_configuration.setMagneticField(fixed_spin)
=========================================================
The calculation works perfectly fine. In particular I am able to follow the evolution of the band-structure with the angle between spin and wire axis...

I now try to do the same type of calculation but in transport (DeviceConfiguration) in order to study the evolution of the transmission with the angle of the spin (AMR),
But my calculation crashed.

Is the option .setMagneticField(fixed_spin) available in DeviceConfiguration?

Thanks in advance

Cyrille

Title: Re: fixed spin in transport calculations?
Post by: Jess Wellendorff on March 23, 2016, 13:04

Yes, I believe FixedSpin now works for device configurations also.