Molcas Forum

Support and discussions for Molcas and OpenMolcas users and developers

You are not logged in.

Announcement

Welcome to the Molcas forum. You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".

#1 Yesterday 11:58:40

ka3
Member
Registered: 2026-03-24
Posts: 10

Basis function symmetry from h5

Hi,

I'm trying to calculate the overlap of two basis sets read from two different rasscf.h5 files. I can successfully read the shells from each h5 file and then compute the overlap of the two basis sets when no symmetry is used. However, I would really like to read each basis from a calculation that uses symmetry and compute the overlap matrix sorted in symmetric blocks. The shells I read from h5 are constructed from attributes such as CENTER_COORDINATES, CENTER_LABELS, BASIS_FUNCTION_IDS, PRIMITIVES, PRIMITIVE_IDS (this is done in the same fashion as the mort tool). The only attribute that seems to contain the symmetry is NBAS, which is then a list of basis functions for each irrep, I'm unsure how to compute the overlap matrix from this while using symmetric calculations.

Is there a way to get the irrep of each basis function? Or any other way to easy compute the overlap matrix in my external code with symmetry?

Any help would be greatly appreciated, thanks!

Offline

#2 Yesterday 15:40:34

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,213

Re: Basis function symmetry from h5

The h5 file has the DESYM_MATRIX dataset. It should be possible to get what you want from that. Maybe you need to get the inverse.

Offline

#3 Today 11:00:24

ka3
Member
Registered: 2026-03-24
Posts: 10

Re: Basis function symmetry from h5

Thanks, I'm not sure what this matrix is though - I presumed it was for desymmetrization? I'd like to compute the overlap matrix of a molecule in it's core-ionized state and it's valence excited states prior to ionization. Ideally I would still like to have each MO assigned to it's corresponding irrep in the point group. Is there a way to get the AO basis functions labelled by symmetry out of molcas directly instead of working with just primitives, that would make my task easier.

Offline

#4 Today 14:20:03

Ignacio
Administrator
From: Uppsala
Registered: 2015-11-03
Posts: 1,213

Re: Basis function symmetry from h5

The distribution of MOs (and AOs) is given by NBAS, so if:
NBAS = [24 15 16 23]
then MOs 1 to 24 are irrep 1, MOs 25 to 39 are irrep 2, MOs 40 to 55 are irrep 3 and MOs 56 to 78 are irrep 4.

Offline

Board footer

Powered by FluxBB 1.5.11

Last refresh: Today 17:53:55