Hi,
I'm having some trouble with IRC calculation with Gaussian 09 package.
First, I have successfully optimized the TS geometry with TS(berny) opt+freq under SMD/ACN solvent condition.
Then, I have submitted the IRC with the option of stepsize=5 , maxpoints=100, maxcyc=300, at the same solvent / level condition.
Reason i did this is because I have failed the last IRC with stepsize=10 (at 19th step, with delta-x convergence matter).
This time, the calculation went smoothly, but at the 79th step, gaussian seemed to have some hard time converging.
However, if I see the step of 65 to 79, I see almost no energy differences, as well as geometry.
So... in this case, can I change maxpoints=100 to maxpoints=75?
or, should I touch other options to make it completely converge?
thanks
Dear Eunsoo,
For IRC, 10 to 25 steps (depends on your system) in both directions (forward and reverse) that end up correctly to the desired geometries are enough, then optimize the last geometries in both directions.
agree with Ahmad, or you can use opt=calcall keyword for the 100 steps
dear Eunsoo,
please inform me finally how you solve the problem?
which stepsize? which maxpoint?
best regards,
- Badges
- Science topic