Vibrational frequency analysis for CI-NEB result

[sapper1]

Dear Admin, Proff. Graeme,
Since, CI-NEB always puts an image to the transition state,is it necessary to perform vibrational modes analysis for CI-NEB results ? if it is recommended, how can we do such calculations faster for slab+molecule(adsorbed) with many images.
thank you very much

[graeme]

Frequency calculations are not necessary, but they are useful to verify that you have a first order saddle point and to calculate the harmonic prefactor of a reaction. Frequency calculations are only needed at the minimum and saddle point (the climbing image). Generally, these calculations are less costly than convergence of the NEB. They can be done in parallel (over displacements) and evaluated only for the degrees of freedom which change significantly during the reaction.

[sapper1]

Dear Proff. Graeme,
thank you for your quick reply.
does that mean we just need to use the POSCAR/CONTCAR of the image at TS and the initial image separately and do vibrational modes calculations using IBRION >5 rather than using all the images.
thank you
best

[graeme]

yes, that's right

[lgxyz]

Always, I get more than one imaginary frequency when I do frequency calculations for the converged saddles.
Can you give suggestions for getting only one imaginary frequency? Thank you very much!

[graeme]

Please search this forum for other posts about this issue and suggestions regarding convergence tests.