[lammps-users] Magnon heat flux & GK thermal conductivity

LAMMPS LAMMPS Mailing List Mirror
1

Hello Dear Lammps users

I want to calculate magnon thermal conductivity(Of course, not all atoms are magnetic), but I have a problem .
i use : fix nve all nve/spin lattice frozen .
the heat flux becomes 0 in all directions :
Step Temp v_Jx v_Jy v_Jz v_k11 v_k22 v_k33
0 0 0 0 0 0 0 0
10000 0 0 0 0 0 0 0
20000 0 0 0 0 0 0 0
30000 0 0 0 0 0 0 0
40000 0 0 0 0 0 0 0
50000 0 0 0 0 0 0 0
60000 0 0 0 0 0 0 0
70000 0 0 0 0 0 0 0
80000 0 0 0 0 0 0 0

thanks all

2

With “lattice frozen” your atoms won’t move and compute heat/flux knows nothing about spins, so your output makes perfect sense to me.

3

Thank you very much.
Do you think if I change from “frozen” to “moving” the problem will be solved?
And then I will only see the contribution of magnons in thermal conductivity?

4

Thank you very much.
Do you think if I change from “frozen” to “moving” the problem will be solved?

no.

And then I will only see the contribution of magnons in thermal conductivity?

of course not.

5

Before using the nve/spin thermostat ,I used a langevin thermostat for magnetic atoms and reached them to desired temperature. Is it sufficient?
or should I do something different?

6

Before using the nve/spin thermostat

fix nve/spin is an integrator not a thermostat, fix langevin is a thermostat not an integrator.
thus you cannot replace one with the other. they do different things.

,I used a langevin thermostat for magnetic atoms and reached them to desired temperature. Is it sufficient?

sufficient for what?

again, you need to be careful what you measure. most features in LAMMPS know nothing about spins, so they will just ignore them. if there would be a contribution then only indirectly through the components that know about them. having an additional “invisible” property and interaction may change other measurable properties.

or should I do something different?

I think you have not properly thought through what you want to do and what you need (as a method) to get the desired results.
and now are just fishing to get me to tell you whether any choice of random LAMMPS features you know may give you what you want.
That is not a scientific approach. and I won’t play this kind of game.