Re: CCL:Summary: Calculations without Born-Oppenheimer approximation "2nd Call"

"John McKelvey" <jmmckel*at*>:
 > Jens,
 > A question, please.  Neglecting the fact that FSGO's may be poor basis
 > functions, where do they stand with respect to BO if the positions of the
 > gaussians are optimized along with the SCF process?
 > Regards,
 > John McKelvey
 Dear John:
 I don't have the papers at hand. But as far as I see, the
 basic features of the approach by Boehm and associates are the
 following: They perform the basic calculations within the
 usual Born-Oppenheimer approximation, obtaining data for a
 very large number of nuclear configurations. This part of the
 investigation can be performed with a variety of electronic
 hamiltonians and basis sets, as in other standard ab initio
 procedures. However, and that is the important point, they
 don't determine the molecular "equilibrium configuration" in
 the usual Born-Oppenheimer manner, i.e., by selecting the
 single nuclear configuration that corresponds to a minimum on
 the Born-Oppenheimer potential hypersurface. In the Boehm
 procedure, the quantum and thermal nuclear degrees of freedom
 are accounted for by forming quantum ensemble averages over
 the set of nuclear configurations. The result is a wave-packet
 picture of the nuclear positions that goes beyond the Born-
 Oppenheimer picture.
 Yours, Jens >--<
 JENS SPANGET-LARSEN         Office:         +45 4674 2710
 Department of Chemistry     Fax:            +45 4674 3011
 Roskilde University (RUC)   Mobile:         +45 2320 6246
 P.O.Box 260                 E-Mail:        spanget*at*
 DK-4000 Roskilde, Denmark