[molpro-user] r12-ccsd(t): convergence problem

[Kirk Peterson]

Dear Oleg,

actually the problem is in your wf directive in your hf calculation and not the F12. The ground
state of O2 is a triplet Sigma-, which has a spatial symmetry of A2 in C2v (irrep #4) yet you've specified symmetry 1 (A1).  So your
hf directive should look like:

{hf;wf,16,4,2}

regards,

Kirk


On Oct 19, 2011, at 9:36 PM, Ol Ga wrote:

> Dear Molpro users and Developers,
> 
> I tried to optimized O2 molecule in its triplet spin state. However,
> the job file execulation was terminated because of convergence
> problem:
> 
> CPU time for basis constructions                 0.06 sec
> Fock operators(MO) rebuilt from dump record.
> CPU time for fock operators                      0.06 sec
> 
> FockProj BlockDiagonalize() eigenvalues: (norb=                      106 )
>   E(Occ):    -20.69187681330067        -20.69180955842683
>  -0.8942909102302152       -0.7098648711596274       -0.7098648711578895
>  -0.4352382802670008       -0.4352382802650147
>   E(Virt):  -0.9626366760849758       -0.1241281701396240
>   9.3930670448409470E-002   0.1339011030254502        0.1339011030263404
>   0.1382729585582448        0.1766153660807595        0.1766153661291486
>   0.1977640322360459        0.3318168340490611        0.5185510274457877
>   0.6739940734255396        0.6869627796392771        0.6869627796477453
>   0.7173176797233280        0.7173176797674587        0.8159713781304202
>   0.9074790801596273        0.9445745034304920        0.9445745034542171
>    1.147960402616340         1.147960402624714         1.174498303122963
>    1.297731477118842         1.297731477265529         1.696095951441474
> (a lot of values)
>    11.86879275284763         13.47227273821943         37.03234052522317
>    37.03234052539107         37.49565306264802         37.62684820202917
>    37.62684820292329         38.27374021249445         40.92039981942011
>    42.96722325266681
>   nOcc:                        7   nTotal:                      106   nCore:
>                        2
> 
> ? Error
> ? Encountered below-occupied eigenvalues in virtual block. This may
> indicate serious deficiencies in the orbital or JKfit-basis sets
> (basis_exch). To ignore this error, use explicit,ignore_bad_singles=1.
> ? The problem occurs in f12_shared/Mp2Singles()
> 
> GLOBAL ERROR fehler on processor   0
> 
> The input was :
> 
> ***
> memory,107,m
> geomtyp=xyz
>          ! use cartesian coordinates xmol style
> geometry={
> 2
>          ! Number of atoms
> O2
>   O         0.609118940000      0.000000000000      0.000000000000
>   O        -0.609118940000      0.000000000000      0.000000000000
> }
> basis=vtz-f12
> {hf;wf,16,1,2,0}
> uccsd(t)-f12,df_basis=avtz/mp2fit,df_basis_exch=vtz/jkfit,\\
> ri_basis=vtz-f12/optri
> {optg,gradient=1.d-5,MAXIT=100,method=DIIS}
> frequencies,FREQREC=5400.2,TASKREC=5500.2
> ---
> 
> I would ask you please to suggest me how can I treat this error.
> 
> Sincerely,
> Oleg B. Gadzhiev
> PhD student
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user