Molcas Forum

Maxxie

Member

From: Zürich

Registered: 2018-11-20

Posts: 2

Symmetry Constrained Optimisation

I'm not so much new to Molcas, but I think this is a basic question.

I'm running calculations on a new system, and I've run into an error that I had not seen before because this is the first time that I'm doing symmetry restricted RASSCF calculations in Molcas.

I have a set of molecules that I want to constrain to a certain point group, but the error I get is that "Symmetry operators do not match the geometry". The molecule should be C2, so I set the GROUP card to "XY".

I am assuming that either I have defined this group card incorrectly, or Molcas is having issues because my starting geometry is not constrained to C2. If this is the case, is there anyway that I can tell Molcas to ignore the symmetry of my starting coordinates, but still have it optimise to a symmetric minimum? Would leaving the point group as C1, optimising to a minimum, and then reoptimising that structure for C2 work? Or would Molcas necessarily force that calculation to be C1?

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,012

Re: Symmetry Constrained Optimisation

Molcas requires your input geometry to match the specified symmetry, but there is a SYMThreshold keyword that lets you be slightly (or greatly) off.

You could also "force" the symmetry through constraints. Constraints need not be satisfied in the initial geometry. However, constraints affect only the geometry, so this won't allow you to reduce the computational cost through symmetry or set the wavefunction symmetry. But the optimized geometry should be symmetric enough that you can use it normally with symmetry.

Maxxie

Member

From: Zürich

Registered: 2018-11-20

Posts: 2

Re: Symmetry Constrained Optimisation

Thanks very much Ignacio, I will try your suggestions then update with the results.