Re: [ESPResSo-users] magnetic field due to magnetic dipoles

[Andrew Jewett]

I am extremely far from being an experienced espresso user.
Perhaps I should not post a reply.
(I am posting because it helps me understand espresso better.)

> Previously I studied simpler systems and I know that magnetic field due to
> other dipoles may increase magnetic moments by 20%.

  It sounds like you want to measure the magnetic field at each
location order to manually set the strength of each dipole moment (at
the beginning of your simulation).  Note: I am under the impression
ESPresSo is not (yet) able to modify the strength of a dipole in
response to the magnetic field around it.  (I thought that the dipole
magnitudes of individual particles are permanent.)  I could be wrong.
So I assume you plan to keep these dipole strengths constant as you
run the simulation.

Regarding how measure the magnetic field, you could try to calculate
it directly using your own script, or you could use espresso.  Here is
one way to use espresso:

At the beginning of your simulation, you could make N copies of your
simulation (one for every magnetic particle).

In each simulation, you could
0) delete one of the magnetic particles, and calculate the potential
energy of the system without that particle

Then add a magnetic particle of unit magnitude
1) point the new particle in the x direction and calculate the energy
2) point the new particle in the y direction and calculate the energy
3) point the new particle in the z direction and calculate the energy

Subtract the energy from 0) from 1), 2), 3).  The energy difference
should be proportional to find the strength of the magnetic field in
each direction due to the surrounding particles.

(Unfortunately you would have to do this N times.)

Hopefully somebody else who knows ESPresSo better knows of an easier
way to calculate the magnetic field directly.

  If my comment is not relevant to what you are trying to do, please ignore it.

Cheers!
Andrew

On Fri, Nov 9, 2012 at 10:51 AM, tomaž mohorič <address@hidden> wrote:
> Dear all,
>
> I am a new user of Espresso and I would have a question for more experienced
> users. I read the documentation (user's guide) but couldn't find the answer.
>
> I would like to simulate a system of magnetic dipoles. As far as I
> understood, Espresso is able to calculate the energy of such a system and
> forces acting on each particle.
>
> In my case I would like to know what is the magnetic field (due to all other
> dipoles) at the position of each particle. Problem is not so trivial, since
> the magnetic field due to magnetic dipole decreases as 1/r**3 and one should
> take into account contribution from all particles (for beginning I assume
> repeating cells in x and y axis).
>
> Previously I studied simpler systems and I know that magnetic field due to
> other dipoles may increase magnetic moments by 20%.
>
> My question is: is there any way to obtain the magnetic field due to
> magnetic dipoles from Espresso?
>
>
> Thanks for your help and kind regards.
>
> Tomaz Mohoric