DMFT paramagnetic self consistency

[davoudn]

Dear All,

I tried the DMFT self consistency as is stated in http://ipht.cea.fr/triqs/doc/user_manual/solvers/dmft/dmft.html, my problem is that for the spin up and down blocks the real part of the resulting (matsubara frequency) green's functions are quiet different for paramagnetic simulation(means there is a magnetic moment while there is a paramagnetic condition). I would appreciate any comments to figure out the issue.

Thanks in advance
davoud

[mferrero]

Hi! Sorry for the late answer. We are in the process of releasing a new version of TRIQS and have been a bit busy. Usually this kind of problem appears if the simulation is not long enough. Namely, you will have to increase the number of cycles to get better statistics. Then most likely the up and down components will start to look the same.