[CP2K-user] [CP2K:16563] How to use ECP potential in FORCE_EVAL

[hutter at chem.uzh.ch]

Hi

basis sets and pseudopotentials are not coupled in the CP2K input.
You can mix (even if it doesn't make sense) whatever you want.

Shell sets (per atom) is the first lone number in the basis set definition.
In your case this is 4, so you get 84*4 Shell sets in total. As, again in
your case, each set has exactly 1 shell, you get the same number of
total shells.
Other basis sets have less sets (e.g. MOLOPT have 1 set), but more shells per
set. Everything is as it should be, there is no bug here.

regards

Juerg Hutter
--------------------------------------------------------------
Juerg Hutter                         Phone : ++41 44 635 4491
Institut für Chemie                E-mail: hutter at chem.uzh.ch 
Universität Zürich                  
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp2k at googlegroups.com wrote: -----
To: "cp2k" <cp2k at googlegroups.com>
From: "Eugene" 
Sent by: cp2k at googlegroups.com
Date: 02/07/2022 08:02PM
Subject: Re: [CP2K:16561] How to use ECP potential in FORCE_EVAL

Hi Jeurg,
Cool it works. But number of MO is not the same for ECP and GTH. Now for ECP we have:
  Total number of            - Atomic kinds:                                   2
                             - Atoms:                                         84
                             - Shell sets:                                   336
                             - Shells:                                       336
Number of Shell cets is not equivalent to number of atoms, while in GTH this is the case. Maybe it is common for ECP. So, it means CP2K Documentation is not complete. Should I give a BUG report?
Warm wishes,
Eugene
понедельник, 7 февраля 2022 г. в 21:54:50 UTC+3, jgh: 
Hi 
 
check your basis set definition: the p shell says: 
 
 2  1  1  6  0  1 # Gauss type functions l=1 set 1 (3p orbital) 
 
That would be zero p-orbitals. Probably a typo. It should be 
 
 2  1  1  6  1  0 # Gauss type functions l=1 set 1 (3p orbital) 
 
Please also note that this is a not very well tested option. 
I'm not aware of any successful application using ECPs with CP2K. 
 
regards 
 
Juerg Hutter 
-------------------------------------------------------------- 
Juerg Hutter                         Phone : ++41 44 635 4491 
Institut für Chemie                E-mail: hut... at chem.uzh.ch  
Universität Zürich                   
Winterthurerstrasse 190 
CH-8057 Zürich, Switzerland 
--------------------------------------------------------------- 
 
-----cp... at googlegroups.com wrote: ----- 
To: "cp2k" <cp... at googlegroups.com> 
From: "Eugene"  
Sent by: cp... at googlegroups.com 
Date: 02/07/2022 05:30PM 
Subject: [CP2K:16557] How to use ECP potential in FORCE_EVAL 
 
According to cp2k documentation https://manual.cp2k.org/trunk/CP2K_INPUT/FORCE_EVAL/SUBSYS/KIND.html#POTENTIAL it is possible to set POTENTIAL variable value as ECP, in order to use Effective core potential. I've converted basis set from http://www.tc.uni-koeln.de/PP/clickpse.en.html according to CP2K documentation. The section KIND in my input file is: 
 &KIND V 
BASIS_SET STOLL-PBE 
POTENTIAL ECP 
&POTENTIAL 
V nelec 10 
V ul 
2      1.000000000            0.000000000       
V S 
2     14.490000000          178.447971000       
2      6.524000000           19.831375000       
V P 
2     14.300000000          109.529763000       
2      6.021000000           12.570310000       
V D 
2     17.480000000          -19.219657000       
2      5.709000000           -0.642775000       
&END 
&END KIND 
The basis set in new file is: 
V STOLL-PBE 
 4 
 2  0  0  7  1  0 # Gauss type functions l=0 set 1 (3s orbital) 
  12.8432080  0.460448 
  11.3757530 -0.495309 
   5.4069740 -0.350821 
   1.4659270  0.654979 
   0.5980800  0.582548 
   0.0887900  0.017081 
   0.0353180 -0.007455 
 2  0  0  7  1  0 # Gauss type functions l=0 set 2 (4s orbital) 
  12.8432080  0.126932 
  11.3757530 -0.139770 
  5.4069740  -0.089862 
  1.4659270   0.204669 
  0.5980800   0.321169 
  0.0887900  -0.657043 
  0.0353180  -0.474893 
 2  1  1  6  0  1 # Gauss type functions l=1 set 1 (3p orbital) 
  31.8898680  0.004658 
   8.2371780 -0.119249 
   4.3283730  0.125195 
   1.5405260  0.620704 
   0.5280810  0.416780 
   0.0899620  0.014015 
 2  2  2  5  1 # Gauss type functions l=2 set 1 (3d orbital) 
  22.6804330  0.031628 
   6.8613120  0.153807 
   2.2754450  0.378661 
   0.7319220  0.507276 
   0.2007460  0.275921 
 
CP2K starts normally but exit with error message: 
 CPASSERT failed   qs_mo_types.F:295 
By checking source code I have error here: 
 CPASSERT(nmo >= nomo) 
 
Could someone kindly give a working example for force calculations with ECP potential. 
 
Many thx in advance, 
Eugene   
  --  
 You received this message because you are subscribed to the Google Groups "cp2k" group. 
 To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns... at googlegroups.com. 
 To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/a65b2506-cc62-475b-8514-3ad646aea6a3n%40googlegroups.com. 
  
  
  -- 
 You received this message because you are subscribed to the Google Groups "cp2k" group.
 To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
 To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/32e8d80d-d886-4a51-b040-de7a787533a3n%40googlegroups.com.
 

-- 
You received this message because you are subscribed to the Google Groups "cp2k" group.
To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/OFCA1A3C5C.BE25258C-ONC12587E3.00346A27-C12587E3.00346A29%40lotus.uzh.ch.