CCL:M:Charges and dipole moments (fwd)

  Re below discussion on charges:
  >    Quite the contrary -- a charge model can do better than the underlying
  > wave function if it is DESIGNED to do so (what we have coined a Class IV
  > charge model). The CM1 charge model (reference Storer, J. W.; Giesen, D.
  > J.; Cramer, C. J.; Truhlar, D. G. " Class IV Charge Models: A New
  > Semiempirical Approach in Quantum Chemistry" J. Comput.-Aid. Mol.
  > 1995, 9, 87. available on the web at
  > is a mapping
 procedure that
  > takes lower quality charges (e.g., semiempirical ZDO Mulliken) and
  > them to very high-quality charges that reproduce experiimental dipole
  > moments better than the expectation value of the dipole moment operator
  > acting on the semiempirical wave function.
 	It is very interesting that this can be done.
 	I am curious if anyone has tried to fix up the semiempirical
 dipole moment operator rather than the charges themselves.  After all,
 an exact semiempirical dipole moment operator should really be
 parameterized to include corretaion effects.
 	With Best Regards
 	Chuck Martin
 	Charles H. Martin
 	Quantum Theory Project
 	University of Florida
 	Gainesville, FL  32606