CCL: IRC for flat PES

 Sent to CCL by: "Pierre  Archirel" [pierre.archirel..u-psud.fr]
 This is an answer to E. V. Horbatenko
 If the PES is very flat there is a difficult issue in the separation of the
 center of mass of the molecule. It is very useful to compare the
 "negative" frequency characterising the TS with the six
 "zero" frequencies coming from the global translation and rotation. I
 have two examples of this, for the same molecule, with B3LYP and int=ultrafine:
 Example 1:
  Low frequencies ---  -36.0058  -2.8583  -0.0001   0.0005   0.0009   5.9011
  Low frequencies ---   10.3067  18.8761  33.4494
 it can be seen that the "TS" frequency (-36.0058) is much larger (by a
 factor 10) than the first "zero" frequency (-2.8583). Actually the IRC
 calculation worked nicely.
 Example 2:
  Low frequencies ---  -13.7156  -6.3252  -3.9374  -0.0003   0.0003   0.0005
  Low frequencies ---    5.4561  16.2158  28.6961
 In this case the "TS" frequency (-13.7156) is only two times larger
 than the first "zero" frequency (-6.3252). I could not make the IRC
 calculation even start. Obviously in this case the "TS" direction is
 polluted by global rotation.
 Have you compared these values?
 Pierre Archirel
 LCP, Universite Paris-Sud, France
 pierre.archirel.u-psud.fr