Small +ve freqs in HF

[Jeremy R Greenwood <jeremy &$at$& med.su.oz.au>]

 Greetings all,
 Some simple questions for those more familiar with ab initio than I:
 As I understand it, the presence of a small real frequency in a
 Hartree-Fock calculation is regarded as evidence that on increasing
 basis size, the constrained structure may turn out to be a saddle point,
 with the mode in question becomming imaginary.
 My question is: How small is small?
 I am performing calculations on semi-aromatic heterocycles, and at
 HF|6-31g* in Cs symmetry, I find a frequency of 8 /cm. Others around
 30-50 /cm. The ring is somewhat prone to puckering, and indeed some
 similar calculations have produced C1 structures.
 My other question is: Does inclusion of electron correlation (e.g. MP2)
 with analytic second derivatives tend to produce similar results in terms
 of symmetry breaking as increasing basis set? (I understand that answers
 to these questions may depend very much on the nature of the systems
 being studied.)
 Jeremy
 --
 Jeremy Greenwood
 Department of Pharmacology
 University of Sydney