Info about analytic gradients for wb97m-v or b97m-v functionals for use with OPTIMIZE keyword

feature-request
1

Hi,

I am trying to use the recently proposed wb97m-v and b97m-v functionals to optimize molecular geometries.

I found out that, even with the latest stable version available (Psi4 1.2.1, Git: Rev {HEAD} 406f4de), the optimisation step run only with the keyword optimize(‘B97M-V’, dertype=‘energy’), that is with the additional parameter dertype=‘energy’.

Without this parameter for the optimize keyword, I obtain the following error:

Traceback (most recent call last):
File “/Users/andrea/psi4conda/bin/psi4”, line 269, in
exec(content)
File “”, line 38, in
File “/Users/andrea/psi4conda/lib//python3.6/site-packages/psi4/driver/driver.py”, line 1111, in optimize
G, wfn = gradient(lowername, return_wfn=True, molecule=moleculeclone, **kwargs)
File “/Users/andrea/psi4conda/lib//python3.6/site-packages/psi4/driver/driver.py”, line 638, in gradient
wfn = procedures[‘gradient’][lowername](lowername, molecule=molecule, **kwargs)
File “/Users/andrea/psi4conda/lib//python3.6/site-packages/psi4/driver/procrouting/proc.py”, line 2080, in run_scf_gradient
grad = core.scfgrad(ref_wfn)

RuntimeError:
Fatal Error: V: RKS cannot compute VV10 gradient contribution.
Error occurred in file: /Users/github/builds/conda-builds/psi4-multiout_1530689453513/work/psi4/src/psi4/libfock/v.cc on line: 904

So I am assuming that the analytic gradients for these functionals are not yet available in the stable version. Am I right? Or I am doing something wrong?

Any guess about when they will be available?

Thank you very much for your help.
Cheers,
Andrea

2

The original hold up was that we did not have range separated gradients available. Now that we do, adding in VV10 should be rather straightforward. Do you have a timetable in mind to get these gradients in?

3

Hi,

thank you for your kind reply.

About the timetable, since I am (very) interested in using/testing these functionals for obtaining optimised geometries of weakly-bound molecular clusters, I confess that, at least for me, the sooner they will be available the better…

Cheers,
Andrea

4

This is something that I would like in as well, let me see what I can do. Poke me in a few weeks if you not heard back.

5

It would be great!
Cheers,
Andrea

6

Hi,

do you have any updates?

Cheers,
Andrea

7

We got most of the way there, but there is an additional term that will require infrastructure updates before we can make it work.

8

Ok, thank you very much for the update.

If you can, please keep me posted.

Cheers,
Andrea