Molcas Forum

Sihan Zhang

Member

Registered: 2025-05-15

Posts: 10

Issues on HDF5 printing of RASSI

Dear developers

I am currently developing RIXS polarization calculation approaches based on the output of OpenMolcas. In the past I worked with the .out file, but I noticed that HDF5 can be a better choice for me to use in my package. Therefore, I tried with this, and got many issues:

1. How can I ask RASSI module to print the transition vectors of SOC states in .h5. I find only SF states data is stored, and SOS_EDPIMOM, SOS_ANGMOM, SOS_SPIN (both real and imag) are all zero, even thought I can find non-zero values in .out.

2. How can I store higher order transition tensors or velocity gauge transition tensors information in .h5. Before I printed them in .out by MEES/MESO and Properties to print data like 'MLTPL  2' 1. I expected it is possible to do so by using some keywords.

3. Are all of the data in HDF5 is in atomic units?

For reference, my package is available on GitHub:
https://github.com/ZhangHMDS/Polarixs

It is designed to compute RIXS planes with linear polarization dependence from OpenMolcas's output. In the future I will include M1 and SP transition in this package.

Best Regards!
Sihan

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,204

Re: Issues on HDF5 printing of RASSI

1. Try with the MESO keyword. I think it will also activate storing the data in the .h5

2. Only some data is stored in the the .h5. If you don't find the placeholder in the file, there's probably no way to save that data in the .h5 currently.

3. They should be, unless the description says otherwise.