Why is interaction energy of a halogen-bonded complex unexpectedly and surprisingly positive?

Dear respected forum members,

The halogen-bonded F-F…NCNa complex was fully optimized at MP2/aug-cc-pVTZ level. Then the value of counterpoise-corrected interaction energy (CP-IE) was estimated through CCSD(T)/CBS approach of PSI4 employing “aug-cc-pV[TQ]Z” basis sets. The input file of CBS calculation is as follows (I use PSI4 1.5a1.dev117):

memory 55 gb

molecule dimer {
0 1
F 0.00000000 0.00000000 -3.65247200
F 0.00000000 0.00000000 -2.23792300

0 1
N 0.00000000 0.00000000 0.32488100
C 0.00000000 0.00000000 1.50710700
Na 0.00000000 0.00000000 3.79061400

units angstrom
symmetry c1

set {
freeze_core true
cc_type df

energy(cbs,scf_wfn=‘scf’,scf_basis=‘aug-cc-pV[TQ]Z’,scf_scheme=scf_xtpl_karton_2,corl_wfn=‘ccsd(t)’, corl_basis=‘aug-cc-pv[TQ]z’,corl_scheme=corl_xtpl_helgaker_2,bsse_type=[‘cp’,‘nocp’],scf_alpha=5.79,corl_alpha=3.05)

Very surprisingly and highly unexpectedly, the CP-IE for this complex is a large positive value of 58.04 kcal/mol!! indicating a quite destabilization. While F-F with a value of maximum electrostatic potential of V_s,max= 13.93 kcal/mol associated with “sigma-hole” of fluorine atom is not a very strong Lewis acid, the NCNa with a minimum electrostatic potential value of V_s,min= -64.77 kcal/mol corresponded to lone electron-pair of nitrogen atom is a strong Lewis base; thus, an almost strong F…N halogen bond interaction is expected. Or, at least, the value of CP-IE should never be positive.

I would be highly grateful if you let me know what is the origin of such a highly positive interaction energy and what solution is suggested. A part of the output is also respectfully presented at the end of this message.

Yours sincerely,

PS. Based on my experiences, the output of CBS extrapolation for a binary complex always includes 12 sections printed out as “==> Stages <==” but the above-mentioned calculation comprises 15 of such section, while the studied complex is a binary one.


  N-Body: Complex Energy (fragments = (1, 2), basis = (1, 2):  -453.92286085738567)

==> N-Body: Counterpoise Corrected (CP) energies <==

n-Body Total Energy [Eh] I.E. [kcal/mol] Delta [kcal/mol]
1 -454.015163867553 0.000000000000 0.000000000000
2 -453.922663778010 58.044682513375 58.044682513375

==> N-Body: Non-Counterpoise Corrected (NoCP) energies <==

n-Body Total Energy [Eh] I.E. [kcal/mol] Delta [kcal/mol]
1 -454.015163867553 0.000000000000 0.000000000000
2 -453.922860857386 57.921013338352 57.921013338352

=> Loading Basis Set <=

Name: DEF2-SVP
Keyword: BASIS
atoms 1-2 entry F          line   150 file /home/saeed/miniconda/share/psi4/basis/def2-svp.gbs 
atoms 3   entry N          line   110 file /home/saeed/miniconda/share/psi4/basis/def2-svp.gbs 
atoms 4   entry C          line    90 file /home/saeed/miniconda/share/psi4/basis/def2-svp.gbs 
atoms 5   entry NA         line   190 file /home/saeed/miniconda/share/psi4/basis/def2-svp.gbs 

Psi4 stopped on: Saturday, 30 October 2021 06:39PM
Psi4 wall time for execution: 1:23:46.11

*** Psi4 exiting successfully. Buy a developer a beer!

likely this issue again: frozen core problem for system with Na and I · Issue #2012 · psi4/psi4 · GitHub

Also, please, don’t write me an email and make a forum post.

Dear Holger,
Many thanks for your kind attention.