wien2triqs : errors

[zanat]

Hi I want to run a spin polarized calculation with wien2triqs.

I have this error in the spin polarized calculation :

 LDAFilename+'.qdmft' : file does not exist  

instead of .dmft there exist .qdmftup and *.qdmftdn.

So, I tried to make this modification by replace the .qdmft by .qdmftup and *.qdmftdn in the end of the file case.py (the end of my file case.py looks like this ):

# find exact chemical potential
SK.put_Sigma(Sigmaimp = [ S1.Sigma ])
SK.Chemical_potential = SK.find_mu( precision = 0.000001 )
dN,d = SK.calc_DensityCorrection(Filename = LDAFilename+'.qdmft')
SK.save()
correnerg1 = mult1 * 0.5 * (S1.G * S1.Sigma).total_density()
correnerg1 -= mult1 * SK.DCenerg[0]
#  case.qdmftup/dn
if (MPI.IS_MASTER_NODE()):
    f=open(LDAFilename+'.qdmftup','a')
    f.write("%.16f\n"%correnerg1)
    f.close()
    # .qdmftdn
    f1=open(LDAFilename+'.qdmftdn','a')
    f1.write("%.16f\n"%correnerg1)
    f1.close()

please Help me.

is those corrections is correct or I need some more modifications in case.py ? N.B.: In the line : dN,d = SK.calc_DensityCorrection(Filename = LDAFilename+'.qdmft') the up/dn is added in python modules (if SP = 1 )

[aeantipov]

Hi,@zanat. It's a bit hard to understand the error from the text. Can you please edit the text and use code syntax or at least <pre> and </pre> tags to split the code from the text?

[zanat]

Hi, @A32167. I'm very sorry. I tried to reformulate the text. I hope that problem is clear now.

[zanat]

Hi, @A32167. By the way, I can't find any issues about the spin-polarization calculation, nor in documentation examples.

[aichhorn]

Hi, There is one important issue. Magnetic LDA+DMFT calculations are done with paramagnetic LDA input, you never run Wien2k in spin polarized mode. And from that paramagnetic input, you then try to stabilize a magnetic solution. This can be done by adding a small magnetic field to the local hamiltonian in the first DMFT iteration. Then reduce this magnetic field, because the self consistent colution has to be stable without it. So try to start over this whole calculation with paramagnetic LDA input.