Imaginary freqs in Spartan (thanks, but...)

 Hello everybody,
 About a couple of weeks ago I asked a question about getting rid of
 imaginary frequencies in Spartan.  First I want to thank:
 Stefan Fau
 Ned Charles Haubein
 Artem Masunov
 Edmund Ndip
 for their interest and suggestions.  I'm thinking that I was not
 explicit enough regarding my problem, so here goes the situation:
 I've been doing calculations of solvation energies using semiempirical
 methods for about two years now.  At first, I used COSMO in MOPAC93,
 and it was common to get "minima" with imaginary frequencies.  Within
 MOPAC93, it was possible to tighten up some criteria, change the
 optimization method, or (last resort) run an IRC to get rid of those
 imaginary freqs.  It was painful but it usually worked (although in
 some recalcitrant cases I was unable to eliminate all the imaginary
 freqs).  Now I'm using the solvent models in Spartan and the problem
 with "minima with imaginary freqs" has reappeared.  So my questions
 are:
 a) Is there a way to run an IRC within Spartan?
 b) Are solvent models bound to produce spurious imaginary frequencies?
 if so, why?
 Thanx
 Victor
 ===
 Victor M. Rosas García
 Computational Chemistry Apprentice
 e-mail: quimico69.,at,.yahoo.com
 _________________________________________________________
 Do You Yahoo!?
 Get your free .,at,.yahoo.com address at http://mail.yahoo.com