[Wien] A bug in qtl f projection

[Kim Kyoo]

Dear All,
I found a bug in the "qtl" package.
SRC_qtl/transf.f should be debugged as following



==================================================
In the very beginning, the definitions of s3, s5 are missing

      s2=sqrt(2.)
*
!!KK--------------------------------------------------------------------------
*
*      s3=sqrt(3.)*
*      s5=sqrt(5.)*
*
!!-------------------------------------------------------------------------------
*
      do m1=-l,l
       do m2=-l,l
        T(m1,m2)=0.
       enddo
      enddo

      if(nm.eq.1)then    ! identity matrix, basis Ylm
       write(6,100)L
100    format(' L=',i3,'. Unitary transformation to Ylm basis')
       do m1=-L,L
         T(m1,m1)=1.
       enddo
      else if(nm.ge.2)then        ! Real basis       (*)
! ordering: real y_L,M^+, y_L,M^-, yL,M-1^+, y_L,M-1^-,.... y_L0
       write(6,101)L
101    format(' L=',i3,'. Unitary transformation to real basis')
       k=-L
       do M=-L,-1
        T(k, M)=1./s2
        T(k,-M)=1./s2
        k=k+1
        T(k, M)= img/s2
        T(k,-M)=-img/s2
        k=k+1
       enddo
        T(L, 0)=1.
      endif


      if(L.eq.3)then
       if(nm.eq.5)then     ! f states in cubic symmetry treated here
*
!!KK--------------------------------------------------------------------------
*
*        T=0*
*
!!-------------------------------------------------------------------------------
*

L=3,nm=5 should be treated separately but this condition holds for (*) too
spoiling T matrix. so we need to initialize again
or one can modify the condition (*) to exclude L=3,nm=5 condition
==================================================
tested result was reasonable.

Best,
Kyoo

Kyoo Kim
dept of Physics
POSTECH, Pohang, Korea
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20130720/2c6828b0/attachment.htm>