[Wien] severe bug in LDA+U or HYBR calculations in complex case (no inversion symmetry

[Peter Blaha]

Unfortunately in the last version (10.1) a severe bug in   lapw1c
has been introduced.
It concerns all LDA+U or Hybrid-DFT (or EECE) calculations, where the
density matrix (case.dmatup/dn and case.vorbup/dn) files have non-zero imaginary
components and there is no inversion symmetry present (files case.in1c instead of case.in1).

The fix is relatively easy:

edit SRC_lapw1/inilpw.f    and change the reading of vorb:

...
       DOUBLE PRECISION   PI, E1, E2, RNN
...
               do i=-l,l
         	do j=-l,l
         	  read(7,*)vorb(jat,l,i,j)   ! this line is bad !
                 END DO
                enddo

to
...
       DOUBLE PRECISION   PI, E1, E2, RNN, rval,cval
...
               do i=-l,l
         	do j=-l,l
         	  read(7,*) rval,cval
                   vorb(jat,l,i,j)=cmplx(rval,cval)
                 END DO
                enddo

The version on the web has also been updated.
-- 

                                       P.Blaha
--------------------------------------------------------------------------
Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna, A-1060 Vienna
Phone: +43-1-58801-15671             FAX: +43-1-58801-15698
Email: blaha at theochem.tuwien.ac.at    WWW: http://info.tuwien.ac.at/theochem/
--------------------------------------------------------------------------