Why the keyword noscf didn't work?

Just Got Here

9:30:56 PM PDT - Mon, Oct 29th 2012
Dear all, I am a new user of NWChem. I tried to obtain the combined wavefunction file from the two fragments, and calculated the hartree-fock energy without updating the wavefunction. However, the program seems to ignore the keyword 'noscf' and still perform many scf steps to turn the charge localized states to delocalized states. Could you tell me how to fix this problem? Thank you very much for your help!! scf uhf singlet vectors input fragment He1.movecs He2.movecs output dimerA.movecs noscf end task scf Sincerely yours, Helina

Forum Vet

2:46:44 PM PDT - Tue, Oct 30th 2012
Helina, For the following input deck the "noscf" keyword seems to be working properly. This case was run with the current NWChem 6.1.1 release version. Bert start he2 geometry he 0 0 0 end basis he library cc-pvdz end scf uhf vectors output he.movecs end task scf geometry he 2 0 0 he -2 0 0 end scf uhf singlet vectors input fragment he.movecs he.movecs output dimer.movecs noscf end task scf Quote:Whitetrees33 Oct 30th 4:30 am Dear all, I am a new user of NWChem. I tried to obtain the combined wavefunction file from the two fragments, and calculated the hartree-fock energy without updating the wavefunction. However, the program seems to ignore the keyword 'noscf' and still perform many scf steps to turn the charge localized states to delocalized states. Could you tell me how to fix this problem? Thank you very much for your help!! scf uhf singlet vectors input fragment He1.movecs He2.movecs output dimerA.movecs noscf end task scf Sincerely yours, Helina

Just Got Here

8:11:45 AM PDT - Wed, Oct 31st 2012

Hi Bert,

Thank you very much for your help! The noscf flag works with the input file you provide, but I am still confused about the final orbital is the combination of the two wavefunction of the He atoms rather than the original localized orbitals even if I turn off the symmetry option. Could you tell me how to fix this problem?

Vector    1  Occ=1.000000D+00  E=-9.142339D-01

             MO Center=  1.1D-18,  0.0D+00,  2.4D-12, r^2= 4.3D+00

  Bfn.  Coefficient  Atom+Function         Bfn.  Coefficient  Atom+Function  

 ----- ------------  ---------------      ----- ------------  ---------------

    1      0.419032  1 He s                  6      0.419032  2 He s          

    2      0.362755  1 He s                  7      0.362755  2 He s

start hehe

basis

he library cc-pvdz  file /lustre/home/apps/NWChem/data/libraries/

end

geometry nocenter noautoz noautosym
he -2 0 0
end

scf

uhf

vectors input atomic output he1.movecs

end
task scf

geometry nocenter noautoz noautosym
he 2 0 0
end

scf

uhf

vectors input atomic output he2.movecs

end
task scf

geometry

he -2 0 0 

he 2 0 0

end

scf

sym off

adapt off

uhf

singlet

vectors input fragment he1.movecs he2.movecs output dimer.movecs

noscf

end
task scf

Forum Vet

9:09:35 AM PDT - Wed, Oct 31st 2012
There is still overlap between the wave functions on either atom, and bonding and anti-bonding molecular orbitals will be formed. If you go to +5 and -5 angstrom there is no overlap and localization does occur. Bert Quote:Whitetrees33 Oct 31st 3:11 pm Hi Bert, Thank you very much for your help! The noscf flag works with the input file you provide, but I am still confused about the final orbital is the combination of the two wavefunction of the He atoms rather than the original localized orbitals even if I turn off the symmetry option. Could you tell me how to fix this problem? Vector 1 Occ=1.000000D+00 E=-9.142339D-01 MO Center= 1.1D-18, 0.0D+00, 2.4D-12, r^2= 4.3D+00 Bfn. Coefficient Atom+Function Bfn. Coefficient Atom+Function ----- ------------ --------------- ----- ------------ --------------- 1 0.419032 1 He s 6 0.419032 2 He s 2 0.362755 1 He s 7 0.362755 2 He s start hehe basis he library cc-pvdz file /lustre/home/apps/NWChem/data/libraries/ end geometry nocenter noautoz noautosym he -2 0 0 end scf uhf vectors input atomic output he1.movecs end task scf geometry nocenter noautoz noautosym he 2 0 0 end scf uhf vectors input atomic output he2.movecs end task scf geometry he -2 0 0 he 2 0 0 end scf sym off adapt off uhf singlet vectors input fragment he1.movecs he2.movecs output dimer.movecs noscf end task scf

Forum >> NWChem's corner >> Running NWChem