Molcas Forum

JakobS

Member

Registered: 2022-02-10

Posts: 5

Availability of orbital-free embedding gradients

Hi! I'm trying to use the OFEM keyword in Alaska to compute analytic orbital-free embedding gradients. However, I'm not sure about how to put together the Molcas input.
Skimming through the source code, orbital-free embedding gradients seem to be implemented together with density fitting, but this feature does not seem very well tested/documented and I can't find any literature which uses it.

Starting from the Br2--H2O test (test/grayzone/898.input), I simply replaced the CASPT2 section with a call to Alaska:

&ALASKA
  root=1
  OFEM=ldtf/pbe

(... and removed the CIonly and Fake RICD keywords from the RASSCF/Seward module)

I find that MCLR runs fine but ALASKA terminates with:

 ###############################################################################
 ###############################################################################
 ###                                                                         ###  
 ###                                                                         ###  
 ###    Location: get_dArray                                                 ###  
 ###                                                                         ###  
 ###                                                                         ###  
 ###    Data of wrong length:Nuclear charge                                  ###  
 ###                                                                         ###  
 ###                                                                         ###  
 ###############################################################################
 ###############################################################################

Is that the intended use of the OFEM keyword in Alaska? It seems to me that there's something wrong with the way I set up the RunFile in connection with the BSSE keyword (which follows the standard instructions of a OFEM calculation) which leads to an unexpected number of nuclei on the RunFile.

Thanks for any directions!

Cheers,
Jakob

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: Availability of orbital-free embedding gradients

Well, since it's not tested, I guess the functionality is as good as broken (even if it didn't crash). You could file a bug report.

JakobS

Member

Registered: 2022-02-10

Posts: 5

Re: Availability of orbital-free embedding gradients

Thanks a lot for the information Ignacio! I'll consider that.