I am currently working on obtaining the transition state for the deprotonation of a carbocation by histidine. Initially, I froze the bond lengths between the histidine nitrogen and the proton, as well as between the proton and the methyl group being deprotonated. This approach resulted in a single imaginary frequency of over -1000 cm⁻¹ (please see the attached files: 1_input and 1_output).
I then attempted to release these constraints and recalculate the transition state. I have tried various keyword combinations, including calcfc, calcall, scf=qc, and IOP(1/8=1) in conjunction with the level of theory # opt=(ts,noeigen) freq mpw1pw91/6-31+g(d,p). However, the structure did not converge. The energy oscillates uniformly without reaching a minimum (attached is a graph showing this behavior).
Could you provide any suggestions on how to address this issue ?
Thank you for your time and assistance.
The method you use to calculate a TS is correct, Opt=(TS,calcfc,noeigen). However, it is important to have a good approximation to the TS. You should first perform a scanning of how energy changes when methyl is deprotonated (see attached). To do that, do a scan job:
# opt=modredundant plus other keywords
.../...
.../...
B 1 2 S 10 0.10
This is only an example, replace 1 and 2 with the atom label of your molecule, S means scan type job, 10 -0.10 means ten steps with an increase of 0.10 angstroms. You need to adjust these parameters to your problem.
After the scan finishes, select the highest point and run TS optimization.
Hope this is helpful.
Best, Pablo
Greetings!
Pablo Mtz is absolutely correct. I would just add that it's also possible to use QST2 or QST3 (in Gaussian) or NEB-TS (in other software packages) for preliminary reaction pathway searches.
- Badges
- Science topic