MCSCF convergence issue with natorb saving

[Gaurab Ganguly]

Dear developers,

I am trying to save state-specific natural orbitals in state-averaged MCSCF calculation for further NEVPT2 calculations. I am experiencing ab normal termination of the job when I am saving the state-specific natorbs (total 10 natorbs). When I am not saving the state-specific natorbs I have a quick convergence. Can you please suggest me what is going wrong.

part of the input:

=================================================

 {multi;start,2141.2;
        frozen,   0, 0, 0, 0,0,0,0,0;
        closed,  23,19,19,14,4,3,3,1;
        occ,     25,19,19,15,7,5,5,2;
        wf,      nelec = 187, sym = 6, spin = 3, charge = 0; state, 1;
        wf,      nelec = 187, sym = 7, spin = 3, charge = 0; state, 1;
        wf,      nelec = 187, sym = 8, spin = 3, charge = 0; state,60;
        natorb,2142.2,ci,print;thresh,0.03;orbprint,20;
        natorb,2143.2,state= 1.6;  ! state-specific natorbs
        natorb,2144.2,state= 5.8;  ! state-specific natorbs
        natorb,2145.2,state= 6.8;  ! state-specific natorbs
        natorb,2146.2,state=19.8;  ! state-specific natorbs
        natorb,2147.2,state=21.8;  ! state-specific natorbs
        natorb,2148.2,state=22.8;  ! state-specific natorbs
        natorb,2149.2,state=36.8;  ! state-specific natorbs
        natorb,2150.2,state=51.8;  ! state-specific natorbs
        natorb,2151.2,state=54.8;  ! state-specific natorbs
        }

=============================================================

Output:

=============================================================

 Natural orbital dump for state 36.8 (ms2=3) at molpro section  2149.2    (Orbital set 1)

 ms2, ndet, nfns, ntake                     3                   21                   14                   13
 ? Error
 ? Failure to find complete spin space
 ? The problem occurs in muwvfc:spieig

===============================================================

thanks and regards,

Gaurab

[Gaurab Ganguly]

Dear Prof Warner,

Many thanks for your reply. I am using Molpro 2015. In the manual, I found that they used natorb for NEVPT2 (The NEVPT2 program [Molpro manual]). That is the reason I used natorb instead of canorb. Here I attach the two outputs. One of them worked and the other did not. The difference between the two calculations is that in one of them I saved more natorbs and that one did not work.