Fixed magnetic moment

Message

jlrch · #1 Post by **jlrch** » Tue Jun 05, 2007 5:23 pm

Is it possible to perform a bulk calculation having fixed total magnetic moment?
I mean, for finite systems we can specify NUPDOWN and obtain two Fermi levels,
one for each spin direction. In principle this is also possible for bulk calculations.
However, NUPDOWN can only take integer values but I would like to make a plot
of energy vs magnetic moment.

#2 Post by **admin** » Wed Jun 06, 2007 7:44 am

the difference of spin up and spin down electrons in a cell has to be an integer number. please interpolate the curve if required

hra · #3 Post by **hra** » Fri Jun 08, 2007 11:45 pm

WHAT

First of all, as far as I know, nothing prevents you to give NUPDOWN non-integer values.

The difference between spin up and down in a cell can get any real number value (calculate some ferromagnetic metal and see for yourself).

#4 Post by **admin** » Fri Jun 15, 2007 11:29 am

the main purpose to fix NUPDOWN is to fix a certain spin multiplet state (singlet, duplet,.......) for a molecule or a cluster (though it is read as real number, you are right hra). Please mind that for systems without gap the difference between up and down charge may depend on the quality of the k-mesh (up/down bands crossing the fermi level), thus fixing the total magnetization of the cell is rather to be done with care in that case.