This is another procedure used by our colleague George Yumnam to calculate the band structure using the mBJ+U+SO method:
init_lapw (with spin polarization)
initso (with spin polarization)
runsp_lapw -so -NI -p
init_mbj_lapw
runsp_lapw -so -i 1 -NI -p
save_lapw -d pbe
init_mbj_lapw
runsp_lapw -so -i 100 -NI -p
Then I create the files:
case.indm, case.indmc, case.indm_so, case.indmc_so, case.inorb, case.inorb_so
runsp_lapw -so -orb -NI -p
then, i wanted to perform the band structure calculation.
So, I did the following:
x lapw1 -band -up -p
x lapw1 -band -dn -p
x lapwso -up -orb -p
Then prepare the case.insp
x spaghetti -up -so -p
My name is Dr Abderrahmane Reggad. I'm a 54 year old and I am a researcher in materials' sciences. 
2 Comments
But according to recent post of Dr. Tran, mBJ +U is not effective.
ReplyDeletehttp://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg03146.html
Actually, I think you have misinterpreted the discussion in the provided link.
DeleteIt was intended to mean that MBJ already gives the correction to the exchange correlation functional, hence doing the +U calculation would not be effective.
If I take E_del(functional) as the correction after +U for a particular functional, then:
E_del(GGA/LDA) >> E_del(MBJ)
i.e. to say that there is very less energy corrections in case of MBJ after application of +U.
However, MBJ+U is valid and consistent with available data.