use of GMH in TDFT

4 views
Skip to first unread message

Oleksii

unread,
Nov 5, 2025, 7:13:54 PM (13 hours ago) Nov 5
to NWChem Forum
Dear NWChem developers,

While exploring the source code of TDFT, I have come across the lines corresponding to the calculation of the electron coupling coefficient (Hab) using Generalised Mulliken-Hush theory:
#if 0
c
c        Calculate GMHCoupling Hab = mij*Eij/dsqrt((deltamu)**2 + 4*mij*mij)
         mij = dsqrt(tmom(2)**2+tmom(3)**2+tmom(4)**2)
         deltaEij = apbval(n)
         if (dabs(mij).le.mijtol) mij = 0.d0
         mijdeltaEij = mij*deltaEij
         deltamuij2 = (gsmom(2)-exmom(2))**2
     &              + (gsmom(3)-exmom(3))**2
     &              + (gsmom(4)-exmom(4))**2
         hab = mijdeltaEij/dsqrt(deltamuij2 + 4.d0*mij*mij)
         if (ga_nodeid().eq.0) then
                write (LuOut,*)
                write (LuOut,9252) dabs(hab)
                write (LuOut,*)
         endif
#endif


After uncommenting this block, recompiling the TDDFT module,  and insertion the "gmh" keyword in the configuration file, I was able to make the respective calculation run (have not yet done a proper benchmark though).

Can you please let me know if there are any caveats to using this module? It is also completely missing in the documentation.

Best regards,
Oleksii
Reply all
Reply to author
Forward
0 new messages