vibration analysis using IBRION=5

[kai]

I am moving on to analyze the vibration after a set of sucessful c-NEB calculation. To my perhaps limited understanding, such a vibration analysis could give me information as:
1) confirm the transition state by giving only one imaginary frequency
2) give pre-factor in arrhenius formula (from imaginary frequency in 1)
3)do zero point energy correction for both initial and transition states, thereby correcting the barrier height.

I am wondering if such a vibration calculation is as simple as changing IBRION=5, NSW=1, according to the vasp mannual? The vasp version I am working on is the 4.6 compiled with the vtst2.03d?

If so, where shall i anticipate to find the frequencies in the OUTCAR file and is there any useful Austin script available to help the analysis?

It appears that ibrion=5 could only support gamma point calculation; what shall i do if I want to go for k 2x2x1?

with regards,

Kai

[graeme]

We do not have an automatic way of getting the IBRION=5 frequencies into our prefactor calculation. In principle, this should be just a matter of extracting the frequencies from the OUTCAR and writing a freq.dat file. You could do this by hand. We'll add this to our list of things to do.

The statement that IBRION=5 evaluates frequencies at the gamma point does not mean the gamma point for the electronic structure; it means that no vibrational frequency can have a wavelength longer than the size of the box.

[kai]

thx, Graeme