UT theoretical chemistry code forum

Dear Sir,
After the CI-NEB convergence, the vibrational frequency for the TS is caclulated with only the adsorbates allowed vibrate. INCAR file :
SYSTEM= S hygrogenation on Pd
ENCUT=300
#ISMEAR=0
#SIGMA=0.1

NSW=60
IALGO=38
GGA=91
IBRION=5
POTIM = 0.02
NFREE = 2
#NPAR=1
ISIF=2
#RWIGS=0.5 0.628
#LORBIT=2

LWAVE=F
#LREAL=.FALSE.
I didn't got any imaginary frequency. I thought maybe the images used in CI-NEB is not enogh, so I increased the images and repeat CI-NEB calculation. However, after calculating the vibrational frequency of the subsequent sanddle point, I just got the same results. It does not work. Then, what are the possible reasons? and How to solve the problem? Please give me some suggestion, thank you very much!

Another question:
If there are more than one imaginary frequency in the sanddle point, what should I do next? Please give me some suggestions. Many many thanks!

If the climbing image converges (ediffg < -0.01) there is no need to run more images. The requirement for a valid saddle is only that this point is well converged.

My first guess at a problem would be your frequency calculation. You must use a small ediff value to get accurate forces (~1e-7) and use small displacements (~0.002 Ang). Calculate the modes for all non-frozen atoms in your calculation, not just the adsorbate atoms.