I have found a TS point which I suspect is the behaviour I am looking for. To confirm I have been trying some IRC calculations, but none have been successful. Unfortunately I get the following output:
PES minimum detected on this side of the pathway.
Magnitude of the gradient = 0.0001140
Calculation of REVERSE path complete.
Reaction path calculation complete.
Energies reported relative to the TS energy of -1177.662975
--------------------------------------------------------------------------
Summary of reaction path following
--------------------------------------------------------------------------
Energy RxCoord
1 -0.00006 -0.19244
2 0.00000 0.00000
3 -0.00005 0.12758
--------------------------------------------------------------------------
Total number of points: 2
Total number of gradient calculations: 13
Total number of Hessian calculations: 0
I am using irc=(RCFC,maxpoints=20) and geom=allcheck.
I suspect that the PES around the TS is very flat, so the HEP algorithm isn't stepping over it.
I have queued several different stepsize values to see if that helps.
Any other suggestions?
Hi Travis,
That has been a missive help! I've started twelve independent jobs running IRC over the same TS starting point. I have varied the iop(1/7=n) to 5, 10 and 100, with and without IRC=(gs2), which in turn is either the low theory or a high theory.
So far only the iop(1/7=5) using G09 irc and a low level of theory has crashed with:
Maximum number of corrector steps exceded.
Error termination via Lnk1e in /usr/local/g09/l123.exe at Wed Jun 4 02:08:00 2014.
The irc with iop(1/7=n), where n=5,10,100, and a higher level of theory (6-31+g(d)) don't seem to escape the TS state as they report that the FORWARD direction was completed after one step.
Seems as though I am left to rely on the old implementation IRC=(gs2), in either low or high level or theory with different iop(1/7=n) values.
I am considering taking my original low level theory TS data and performing a restart with a high level of theory.
Thanks for the help.