MCSCF orbital energies
Jacky LIEVIN
jlievin at ulb.ac.be
Thu Aug 16 16:22:06 BST 2001
Dear molpro users
MCSCF orbitals are usually listed (command gprint,orbital) by order of
increasing orbital energies, except in some cases (see example below) in
which we found that the energy ordering does not occur for some low-lying
orbitals. As a consequence the definition of the closed and active space is
no more correct and the next mrci calculation is build on a wrong active
space. We don't know the origin of this problem and how to solve it. We
only found in the example below that an increase of the active space may
help but we don't know why
Any comment and suggestion is welcome. Many thanks in advance
Jacky LIEVIN
Sample input: calculation of a 5 delta state of VCl
___________________________________________________
***,VCl
memory,20,m
gprint,orbital,civector
r=2.3 ang
int
geometry={Cl;V,Cl,r}
ECP,2,ECP10MDF;
ECP,1,ECP10MWB;
SPD,2,ECP10MDF;C;
F,2,even,1,,0.8;
SP,1,ECP10MWB;C;
D,1,VTZ;
rhf;
occ,6,3,3,0
closed,4,2,2,0;
wf,20,4,4
multi;
closed,4,2,2,0;occ,6,3,3,1; ! MULTI 1
wf,20,1,4;lquant,2;
wf,20,4,4;lquant,2;
multi;
closed,3,1,1,0;occ,7,3,3,1; ! MULTI 2
wf,20,1,4;lquant,2;
wf,20,4,4;lquant,2;
multi;
closed,2,1,1,0;occ,7,3,3,1; ! MULTI 3
wf,20,1,4;lquant,2;
wf,20,4,4;lquant,2;
---
MULTI 2: the 4 lowest MO's are in the right order
_______
1.1 2.00000 -3.291584 0.002178 0.002094 -0.004011 -0.001982
0.000788 0.000232 0.000000 0.000831 0.000000 0.398
544
0.655887 0.582799 0.018516 -0.008198
0.001799 -0.000035 -0.000423 0.001543 0.000867 -0.000
517
-0.002639 0.000000 -0.001035 0.000000
0.000342 0.000000 0.000325 0.000000
2.1 2.00000 -2.057822 -0.013137 -0.013931 0.001620 0.003473
0.000099 -0.002832 0.000000 -0.000722 0.000000 -0.001
599
-0.003980 -0.001368 0.006101 -0.001021
0.000290 0.122660 0.716797 0.422498 0.007286 -0.000
748
0.006929 0.000000 0.001412 0.000000
0.000661 0.000000 -0.008655 0.000000
3.1 2.00000 -1.003809 0.444777 0.532868 0.185142 0.038915
-0.007808 0.002689 0.000000 0.023296 0.000000 -0.023
858
-0.049160 -0.042378 0.057118 -0.012434
0.006962 0.011875 0.072761 0.028513 -0.026847 0.006
506
0.043111 0.000000 0.026100 0.000000
-0.003941 0.000000 -0.005567 0.000000
4.1 1.98863 -0.475588 0.133341 0.219596 -0.786185 -0.129850
0.007866 -0.009015 0.000000 -0.034420 0.000000 0.036
796
0.074880 0.090417 -0.195576 0.043989
-0.006667 -0.010675 -0.067981 -0.016489 0.112361 -0.034
623
-0.095729 0.000000 -0.076661 0.000000
-0.004637 0.000000 0.007448 0.000000
MULTI 3: MO's 1.1 and 3.1 are permuted --> 1.1 becomes active and 3.1 closed!
_______
1.1 2.00000 -2.057140 -0.012709 -0.013422 0.001744 0.003488
0.000104 -0.002830 0.000000 -0.000708 0.000000 0.003
166
0.003845 0.005604 0.006389 -0.001135
0.000320 0.122646 0.716686 0.422621 0.007302 -0.000
757
0.006951 0.000000 0.001432 0.000000
0.000658 0.000000 -0.008646 0.000000
2.1 2.00000 -1.013399 -0.444235 -0.532288 -0.183413 -0.038533
0.007731 -0.002682 0.000000 -0.023234 0.000000 -0.001
695
0.007053 0.004946 -0.057795 0.012843
-0.007049 -0.011626 -0.071272 -0.027847 0.026524 -0.006
400
-0.042704 0.000000 -0.025863 0.000000
0.003925 0.000000 0.005509 0.000000
3.1 1.99923 -3.281112 -0.026991 -0.033306 -0.010383 -0.003599
0.001227 0.000174 0.000000 -0.000381 0.000000 0.398
928
0.656958 0.583752 0.016224 -0.007703
0.001400 -0.002184 -0.013154 -0.005201 0.001695 -0.000
668
-0.004973 0.000000 -0.002160 0.000000
0.000609 0.000000 0.000729 0.000000
4.1 1.98853 -0.474996 0.132538 0.218658 -0.786397 -0.129938
0.007888 -0.008990 0.000000 -0.034412 0.000000 0.039
566
0.079444 0.094509 -0.195528 0.043879
-0.006658 -0.010703 -0.068153 -0.016555 0.112377 -0.034
643
-0.096309 0.000000 -0.076888 0.000000
-0.004622 0.000000 0.007471 0.000000
NB: adding 8.1 in the active space solves the problem:
multi;
closed,2,1,1,0;occ,8,3,3,1;
wf,20,1,4;lquant,2;
wf,20,4,4;lquant,2;
------------------------------------------------------
Jacky LIEVIN
Universite Libre de Bruxelles
Laboratoire de Chimie Physique Moleculaire
Cpi 160/09
50 Av. F.D. Roosevelt
B-1050 Bruxelles
Belgium
Tel: 32-2-650 4089
Fax: 32-2-650 4232
------------------------------------------------------
More information about the Molpro-user
mailing list