Calculating response prorperty for more than 30 frequencies

[ankit7540]

Hello all,
It appears that the general RESPONSE section has a limit of 30 frequencies which are accepted at a time. If I try to enter more than 30, only the first 30 cases are included in the calculation. Is this behavior as expected.

The calculation finishes without any error though.
(Full output file is attached with this post.)
Thank you.

[kennethruud]

Yes, 30 is the default maximum number of frequencies that can be run in a conventional linear response calculation. You can reset this by changing MFREQ in the file include/infrsp.h. An alternative is of course to do 30 frequencies in each calculation, and do separate calculations and restart from the converged wave function (assuming you have kept this information).

Since you are scanning a broad frequency range, you may want to consider looking into the complex polarization propagator/damped response theory functionality of the code, where the default number of simultaneous frequencies is set to 300, even without recompilation.


Best regards,

Kenneth

[ankit7540]

Thank you for the clarification.
I do have another question regarding the RESPONSE calculation.

For frequencies crossing the electronic absorption we see negative values for polarizability. Are these the real components. Can we also get the Imaginary components, as shown in the figures in the work : https://aip.scitation.org/doi/pdf/10.1063/1.2107627

Thanks.

[kennethruud]

Getting the imaginary components of the polarizability is what the Complex Polarization Propagator/Damped Response Theory functionality will be able to give you.

Kenneth

[ankit7540]

Thank you.