[molpro-user] Simple unreasonable norm problem...

[Kirk Peterson]

Charlie,

probably this is just due to RHF and CCSD not being very happy dissociating your molecule from an overall singlet to two
doublet atoms (and a closed-shell Ca2+).  But outside of that, please realize that the default frozen core for Ca
is [Ar], so that your calculation only involves 2 correlated electrons. So the (T) in your input is redundant since CCSD
already corresponds to a FCI.   I would recommend using the MRCI program, presumably with some sort of CASSCF orbitals,
which will not have any problems dissociating this molecule.

regards,

Kirk

PS - perhaps with the small VDZ basis set, one of the charges on Ca is transferring to an H? Certainly the IP of H is larger
than the 2nd IP of Ca, but maybe not with this basis set.

On Oct 14, 2010, at 10:38 AM, Charles Wood wrote:

> Hi Molpro mailling list,
> 
> I apologise in advance if this is a simple problem to fix. I looked on 
> the mailing list's previous thread to no avail.
> 
> My problem is this:
> I am trying to run an extremely simple H - Ca2+ - H calculation (one 
> calcium 2+ ion centred between 2 hydrogen atoms). I'm using UHF and 
> UCCSD(T) programs in molpro2009.1. If the geometry is specified so each 
> of the H atoms are less than 1.5 Angstroms then the calculation works fine.
> 
> If the H atoms are equal to or more than 1.5 Angstroms from the Ca2+ 
> ion, I get an unreasonable norm and an unconverged calculation. The 
> input file and output error are at the tail of this message.
> 
> I saw some people suggest 'shift' but this looks like something 
> connected to the geometry....any idea??
> 
> Thank you for any help!
> Charlie
> *************
> INPUT:
> *************
> ***,HHCa
>  memory,64,mw
>  print,basis,orbitals
> 
> geomtyp = xyz
> symmetry,nosym
> angstrom
> geometry={
> 3,
> 
> h  -1.5   0.0   0.0
> h   1.5   0.0   0.0
> ca  0.0   0.0   0.0
> end
> }
> 
> basis=vdz
> wf,20,1,0
> 
> rhf
> uccsd(t)
> ---
> 
> *****************
> OUTPUT ERROR:
> *****************
> Starting UCCSD calculation
> 
> ITER.      SQ.NORM     CORR.ENERGY   TOTAL ENERGY   ENERGY 
> CHANGE        DEN1      VAR(S)    VAR(P)  DIIS     TIME
>   1      1.09064786    -0.06995192  -676.96454186    -0.06995192    
> -0.01242820  0.11D-02  0.11D-01  1  1     0.02
>   2      1.19987435    -0.09555561  -676.99014555    -0.02560369    
> -0.00078764  0.17D-03  0.36D-03  2  2     0.02
>   3      1.20574077    -0.09615967  -676.99074961    -0.00060406    
> -0.00003603  0.21D-04  0.11D-04  3  3     0.03
> 
> UNREASONABLE NORM. CALCULATION STOPPED
> 
> Norm of t1 vector:      0.08966229      S-energy:     0.00000000      
> T1 diagnostic:  0.04483114
> 
> D1 diagnostic:  0.06340081
> Norm of t2 vector:      0.44463631      P-energy:    -0.09615967
>                                         Alpha-Beta:  -0.09615967
>                                         Alpha-Alpha:  0.00000000
>                                         Beta-Beta:    0.00000000
> 
> Doubles amplitudes (print threshold =  0.500E-01):
> 
>         I         J         SYM. A    SYM. B    A         B      T(IJ, 
> AB) [Alpha-Beta]
> 
>         1         1         1         1         1         1     -0.41754002
>         1         1         1         1         1         9     -0.06746129
>         1         1         1         1         1        20      0.06808800
>         1         1         1         1         9         1     -0.06746129
>         1         1         1         1        20         1      0.06808800
> 
> ?ERROR: NO CONVERGENCE. TRIPLES WILL NOT BE DONE. THIS CAN BE AVOIDED 
> USING THE NOCHECK OPTION.
> 
> Spin contamination <S**2-Sz**2-Sz>     0.00000000
> 
> 
> RESULTS
> =======
> 
>  Reference energy                   -676.894589937024
>  UCCSD singles energy                  0.000000003496
>  UCCSD pair energy                    -0.096159674326
>  UCCSD correlation energy             -0.096159670830
> 
> !RHF-UCCSD energy                   -676.990749607853
> 
> Program statistics:
> 
> Available memory in ccsd:                63999882
> Min. memory needed in ccsd:                 28044
> Max. memory used in ccsd:                   28044
> Max. memory used in cckext:                 52759 ( 3 integral passes)
> Max. memory used in cckint:                134726 ( 1 integral passes)
> 
> 
> ? Error
> ? No convergence. This error exit can be avoided using the NOCHECK option
> ? The problem occurs in ccsd
> 
> _______________________________________________
> Molpro-user mailing list
> Molpro-user at molpro.net
> http://www.molpro.net/mailman/listinfo/molpro-user