Spin Torques and Damping Paremeter Gradient

[had...@seznam.cz]

Dear all,

Right at the beginning, I would like to thank the authors for this very interesting and powerful software package.

 I am new to Vampire so I am sorry for potentially banal questions. My question is similar to the question asked in https://groups.google.com/d/msg/vampire-users/e_QhrmxhkYI/662nubjvAAAJ, which was however left without an answer. Since the new 5th version claims to include Spin Transfer Torques, this question is getting a bit more actual, I believe.

My system consist of several regions with different materials which I found very easy to implement.  I would like to then apply spin torques to one of the regions to simulate a spin accumulation. Is that somehow possible? I did go through the manual and also through the source files, however, I did not find any way how this can be done. Did someone try to simulate systems with spin torques applied to the system in general?

My other question is then probably easier. Is there a method how a Damping parameter gradient can be implemented except of specifying several small domains of the same material with just different damping parameter?

I would like to thank you in advance for any help.
Tomáš Hadámek 

[richard....@gmail.com]

Dear Tomas,

Yes - this can be done in a very simplistic way by adding the following commands in the input and material file.

sim:slonczewski-spin-polarization-unit-vector = 0,0,1 # direction of spin current vector

material[1]:slonczewski-adiabatic-spin-torque = 0.1 !T # field equivalent to Slonczewski torque

material[1]:slonczewski-non-adiabatic-spin-torque = 0.01 !T # ditto for non-adiabatic term

(Sorry - this was meant to go in the manual)

For the damping gradient, this unfortunately requires different materials for each layer - I would recommend writing a program to do this automatically by generating a suitable material file.

All the best,

Richard

PS I am working on a new solver which should be able to deal with inhomogeneous magnetisation more realistically using a circuit model.

On Wednesday, March 13, 2019 at 5:19:39 PM UTC

[gurung...@gmail.com]

Dear Richard,

Thank you for the vampire code. I have similar questions regarding the spin-torque implemented in the vampire.

For the application of the spin-torque, is it possible to define the magnitude of the applied current (and other parameters like in the mumax code) besides giving directly the magnitude of the adiabatic and non-adiabatic spin-torque? It would be great and easy to understand for most of us if you can provide the details of the spin-torque(including the equations included in the LLG Slonczewski equation) in the manual for the vampire.  

Thank you

Gurung