undocumented: DFT virtual spectrum vspec


Gets Around
I recently made some updates and improvements to the nw_spectrum.py script included with NWChem, as documented in this thread. One of its capabilities is supposed to be plotting DFT vspec virtual spectrum output. I can't find any documentation about vspec, or any input files demonstrating its use. I do do see source code apparently implementing the functionality. I presume that somebody at some time -- perhaps Kenneth Lopata, author of nw_spectrum.py -- has used this feature, hence a parser for it in nw_spectrum.py.

Is this feature undocumented and untested because it's unfinished? I would try to come up with a vspec demonstration input file for use with nw_spectrum.py by working backward from the source code, but I don't want to put in the effort if the feature is not actually working.

Clicked A Few Times
I came across this thread while searching for information on vspec, which I chanced upon in the source when looking for a way to calculate dipole matrix elements at DFT level.

I simply added the vspec keyword to the dft input block, which gave me a .vspec file at the end of the calculation. This contains the energy difference between pairs of molecular orbitals, what appears to be expectation values of x, y, and z position operator and oscillator strengths. From looking at dft_input.F, I see there are options to specify alpha or beta spin and number of excitations.

I would also to be interested to know if this feature is ready for use. The output I have obtained certainly looks reasonable.

Forum Regular
Hi, Yes. The vspec feature for scalar calculations can be used. I'll document it.

For an open shell calculation:

vspec alpha nocc nvir beta nocc nvir

For a restricted calculation:

vspec nocc nvir

Let me know if there are any issues.

Best,
-Niri

niri.govind@pnnl.gov

Clicked A Few Times
Hi Niri,

Thank you very much. I will be making a lot of use of this!

Best,
Mark


Forum >> NWChem's corner >> NWChem functionality