[Pw_forum] antiferro
Gabriele Sclauzero
sclauzer at sissa.it
Fri Nov 21 15:41:20 CET 2008
ali kazempoor wrote:
> Daer all
> I start the ferromagnetic calculation for the slab of MnAs containing 7 layer,but after
>
> finishing the run , the spin polarization of Mn atoms in surface rotate in 180
> and make the layer antiferromagnetic ,are this feasible?
Actually I don't know anything about the physics of your system, but since the spin
polarization in collinear calculations is a scalar between -1 and 1, I would reply yes, it
is possible to find opposite spin polarization on non equivalent atoms.
> starting_magnetization(1) = 3.0,
> starting_magnetization(2) = 0.01,
Be careful of your value of starting_magnetization(1)! It must between -1 and 1! I thought
that the program would perform a check, but I see that this is not the case. You should
find this warning in your output file:
Initial potential from superposition of free atoms
Check: negative starting charge=(component1): -0.009305
Check: negative starting charge=(component2): -193.251520
which I don't know if it is later treated properly in the code. So I suggest you to start
with a correct value of magnetization.
> degauss = 0.02 ,
This may be not small enough for spin polarized calculations, please check convergence.
> diagonalization=''
> mixing_mode = 'plain'
> electron_maxstep = 300,
you do not need to specify these fields, and you should not unless you know exactly what
are the consequences. Defaults are usually good
> conv_thr = .0001 ,
This looks too small to me. The default is 10^-6, but some people uses even 10^-8
Regards,
GS
--
La legge n. 133 uccide l'Università e la Ricerca italiana!
Save Italian Brains! (http://saveitalianbrains.wordpress.com/)
o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student |
| c/o: SISSA & CNR-INFM Democritos, |
| via Beirut 2-4, 34014 Trieste (Italy) |
| email: sclauzer at sissa.it |
| phone: +39 040 3787 511 |
| skype: gurlonotturno |
o ------------------------------------------------ o
More information about the Pw_forum
mailing list