error when I call WB97M_V

[Ebtisam zaglool]

hi

I tried to call WB97M_V using 

omega=0.3
 srx=0.15
 {int;ERFLERFC,mu=$omega,srfac=$srx}
 {uks,HYB_MGGA_XC_WB97M_V,vv10nl;
     wf,8,,2;
 shift,0.0,0.0
 orbprint,5
 }

but it gives me this error, 

 ? Error
 ? ndiff inconsistent
 ? The problem occurs in dfti_block

 GLOBAL ERROR fehler on processor   0                                       
any idea what is the correct way to call it! 

I attached the out file

[Yevhen Horbatenko]

Dear Ebtisam:

I can suggest a workaround, although not sure if it is exactly what you want.

Rather than providing all those parameters wB97M-V depends on by yourself,

you can install an external library of dft functionals libxc:   https://libxc.gitlab.io

and then call wB97M-V as , e.g., 
{df-ks,HYB_MGGA_XC_WB97M_V;wf,charge=-1,spin=0}

Change the command above to what you need exactly.

Once you have the lib installed, Molpro will be able to find it by itself.
If it does not happen, you can put something like 

PATH=/your/path/to/libxc6.2.2/bin:${PATH}
LD_LIBRARY_PATH=/your/path/to/libxc6.2.2/lib:${LD_LIBRARY_PATH}

inside your job submission script.

A year ago i also had a problem with wB97M-V, unfortunately,  do not remember exactly the problem.

I remember only that this functional could not run too.
I was using molpro 2024.2. 

Ended up installing this external library, and it solved my problem.

Kind regards,

Yevhen

[andreas...@gmail.com]

Dear Ebtisam,

it is unfortunately currently not possible to use VV10 self-consistently in conjunction with meta-GGAs in a KS calculation. A patch for this will be provided in the upcoming new version of Molpro (I think we are about to prepare one quite soon).

Since the impact of the VV10 on the KS potential is quite insignificant I would recommend that you only compute the VV10 energy after convergence of the KS calculation using only the HYB_MGGA_XC_WB97M_V functional. A common input for this would look like this (see also the description in the manual):

{ks,<DFT functional names>; save,2100.2}
eloc=energy
{vdwDF; vv10,den=2100.2,toten=1}
evv10=energy

The variable evv10 will save the total energy (HYB_MGGA_XC_WB97M_V+VV10) in this example.

Best wishes,

Andreas

[tibo...@gmail.com]

Dear all,

Speaking of wB97M-V, I think it would be nice if wB97M-V was added to the list of functional keywords.

It tends to be among the best performing non-double-hybrid functionals in multiple benchmark studies, so being able to simply run {DF-KS, WB97M-V} to get the final wB97M-V energy could improve the user experience, especially for new users.

Combined with the simplified input introduced recently, this would be an easy way to point users towards a modern functional.

Best wishes,

Tibor