Molcas Forum
Support and discussions for Molcas and OpenMolcas users and developers
You are not logged in.
- Topics: Active | Unanswered
Announcement
Please note: The forum's URL has changed. The new URL is: https://molcasforum.univie.ac.at. Please update your bookmarks!
You can choose an avatar and change the default style by going to "Profile" → "Personality" or "Display".#1 2025-11-04 19:45:19
- Lukhmanul Hakeem k
- Member
- Registered: 2024-02-05
- Posts: 45
Clarification on Orbital Selection and Rotation in CASSCF Calculation
Dear Sir,
I hope you are doing well. I have a small doubt regarding the CASSCF calculation I performed with the following input:
&GATEWAY
Symmetry = Z
Basis set
C.ANO-L-VDZP
...
&SEWARD
&SCF
OCCUPIED = 47 10
ITERATIONS = 40
&RASSCF
Symmetry = 1
Spin = 1
NACTEL = 12 0 0
INACTIVE = 47 4
RAS2 = 1 11
LumOrb
&GRID_ITIn this calculation, I did not edit the .Scforb file or save it as an INPORB file. I simply chose a (12,12) active space that includes 6 occupied and 5 unoccupied orbitals from symmetry 2, and 1 unoccupied orbital from symmetry 1.
My doubt is: since I did not manually choose the orbitals, does OpenMolcas automatically select the next unoccupied orbital (e.g., orbital 48) from the .Scforb file as the first active orbital for symmetry 1?
When I visualize the orbitals, I noticed that orbital 48 has a weak isosurface, while orbitals 50 and 54 show stronger orbital density.
I also see a warning in the RASSCF output:
RASSCF iterations: Energy and convergence statistics
----------------------------------------------------
Iter CI SX CI RASSCF Energy max ROT max BLB max BLB Level Ln srch Step QN Walltime
iter iter root energy change param element value shift minimum type update hh:mm:ss
Nr of preliminary CI iterations: 5
###############################################################################
###############################################################################
### ###
### ###
### WARNING: Large orbital rotation. ###
### ###
### ###
###############################################################################
###############################################################################
Molecular orbital 11 of symmetry 2 MO space 2 weight is 0.458684
1 1 24 0 -761.21507191 0.00E+00 5.27E-01* 7 17 2 3.36E-02* 0.47 0.00 SX NO 0:01:06
2 5 22 0 -761.25908116 -4.40E-02* 2.79E-01* 48 50 1 1.65E-02* 0.46 0.00 SX NO 0:00:34
3 5 20 0 -761.27078931 -1.17E-02* 3.88E-01* 48 50 1 1.55E-02* 0.46 0.00 SX NO 0:00:34
4 5 20 0 -761.28157133 -1.08E-02* -3.42E-01* 48 50 1 1.46E-02* 0.45 0.00 SX NO 0:00:32
5 6 15 0 -761.29054797 -8.98E-03* -3.14E-01* 48 50 1 1.40E-02* 0.45 2.21 QN YES 0:00:33
6 6 11 0 -761.29673482 -6.19E-03* -1.13E-01* 48 50 1 1.31E-02* 0.45 1.50 LS YES 0:00:33
7 4 11 0 -761.29761411 -8.79E-04* 6.17E-02 48 50 1 1.27E-02* 0.45 1.58 QN YES 0:00:32
8 6 11 0 -761.29849564 -8.82E-04* 1.20E-01* 48 50 1 1.22E-02* 0.45 2.44 QN YES 0:00:39
9 6 11 0 -761.30015209 -1.66E-03* 1.31E-01* 48 50 1 1.09E-02* 0.45 2.13 QN YES 0:01:07
10 5 11 0 -761.30184661 -1.69E-03* -1.12E-01* 48 50 1 8.89E-03* 0.45 2.04 QN YES 0:00:40
11 6 11 0 -761.30333804 -1.49E-03* 1.11E-01* 48 50 1 6.44E-03* 0.45 2.03 QN YES 0:00:34
12 6 11 0 -761.30456308 -1.23E-03* 1.02E-01* 48 51 1 -5.13E-03* 0.45 2.08 QN YES 0:00:36
13 6 11 0 -761.30553086 -9.68E-04* 7.88E-02 48 51 1 -6.39E-03* 0.45 2.02 LS YES 0:00:34
14 6 10 0 -761.30584800 -3.17E-04* 5.36E-02 48 51 1 -7.37E-03* 0.45 1.32 QN YES 0:00:37
15 6 10 0 -761.30631890 -4.71E-04* 1.01E-01* 48 51 1 -6.55E-03* 0.45 2.29 QN YES 0:00:33
16 6 10 0 -761.30690452 -5.86E-04* -6.32E-02 48 51 1 -4.57E-03* 0.45 1.56 QN YES 0:00:33
17 5 10 0 -761.30717031 -2.66E-04* -4.94E-02 48 51 1 -3.04E-03* 0.45 1.91 LS YES 0:00:35
18 6 10 0 -761.30723902 -6.87E-05* 1.31E-02 48 51 1 -1.86E-03* 0.45 1.08 QN YES 0:00:39
19 6 10 0 -761.30737139 -1.32E-04* 1.77E-02 48 51 1 -1.50E-03* 0.45 2.46 LS YES 0:00:37
20 6 10 0 -761.30748142 -1.10E-04* 1.55E-02 48 49 1 -1.09E-03* 0.45 1.58 QN YES 0:01:14
21 6 10 0 -761.30755281 -7.14E-05* -1.44E-02 48 49 1 -9.78E-04* 0.45 1.42 QN YES 0:00:34
22 5 9 0 -761.30758819 -3.54E-05* -2.21E-02 48 49 1 -8.90E-04* 0.45 1.72 QN YES 0:00:39
23 6 9 0 -761.30761790 -2.97E-05* 1.78E-02 48 49 1 -7.87E-04* 0.45 1.57 QN YES 0:00:38
24 6 8 0 -761.30763426 -1.64E-05* -1.26E-02 48 49 1 -7.07E-04* 0.45 1.76 QN YES 0:00:39
25 6 9 0 -761.30764465 -1.04E-05* -7.63E-03 48 49 1 -6.41E-04* 0.45 1.58 QN YES 0:00:33
26 6 9 0 -761.30765253 -7.88E-06* -7.63E-03 48 49 1 -5.73E-04* 0.45 1.69 QN YES 0:00:33
27 6 9 0 -761.30766059 -8.06E-06* -1.05E-02 48 49 1 -4.83E-04* 0.45 1.81 QN YES 0:00:34
28 7 9 0 -761.30766966 -9.07E-06* -1.18E-02 48 49 1 -3.57E-04* 0.45 2.16 LS YES 0:00:36
29 7 9 0 -761.30767226 -2.59E-06* -3.37E-03 48 54 1 4.29E-04* 0.45 0.83 QN YES 0:00:34
30 6 9 0 -761.30767878 -6.52E-06* -4.97E-03 48 54 1 3.45E-04* 0.45 2.50 LS YES 0:00:33
31 5 8 0 -761.30768424 -5.46E-06* 7.29E-03 48 54 1 2.22E-04* 0.45 1.49 QN YES 0:00:42
32 7 8 0 -761.30768792 -3.67E-06* -6.04E-03 48 50 1 -1.34E-04* 0.45 1.41 QN YES 0:00:34
33 6 8 0 -761.30768958 -1.66E-06* 4.50E-03 9 26 2 -9.67E-05 0.45 1.98 QN YES 0:00:34
34 7 8 0 -761.30769065 -1.07E-06* 3.87E-03 9 20 2 5.45E-05 0.45 1.30 QN YES 0:00:38
35 5 7 0 -761.30769107 -4.23E-07* -1.90E-03 9 20 2 4.00E-05 0.45 1.27 QN YES 0:00:33
36 5 7 0 -761.30769119 -1.15E-07* 4.92E-04 48 59 1 2.50E-05 0.45 1.14 QN YES 0:00:39
37 1 6 0 -761.30769121 -2.48E-08* -2.04E-04 48 54 1 -1.85E-05 0.45 1.51 QN YES 0:00:35
38 3 6 0 -761.30769122 -1.28E-08* -4.70E-04 48 54 1 -1.21E-05 0.45 1.21 QN YES 0:00:39
39 1 6 0 -761.30769123 -4.00E-09 8.57E-05 48 54 1 -7.91E-06 0.45 1.11 QN YES 0:00:34
Convergence after 39 iterations
40 1 6 0 -761.30769123 -1.79E-09 8.57E-05 48 54 1 -4.44E-06 0.45 1.11 QN YES 0:00:36
++ Final optimization conditions:
------------------------------
Average CI energy -761.30769123
RASSCF energy for state 3 -761.27892150
Super-CI energy -0.00000000
RASSCF energy change 0.00000000
Max change in MO coefficients 1.518E-04
Max non-diagonal density matrix element 8.571E-05
Maximum BLB matrix element -4.435E-06
(orbital pair 48, 54 in symmetry 1)
Norm of electronic gradient 4.821E-02
--Does this mean that OpenMolcas automatically performs orbital rotations during the optimization and can replace orbitals (e.g., 48) with higher orbitals (e.g., 54) if they better describe the active space?
I would be very grateful if you could please clarify this behavior.
Thank you very much for your time and guidance.
Kind regards,
Lukhmanul Hakeem K.
Last edited by Lukhmanul Hakeem k (2025-11-04 19:54:40)
Offline
#2 2025-11-05 08:59:47
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,165
Re: Clarification on Orbital Selection and Rotation in CASSCF Calculation
The active orbitals will be no. 48 in irrep 1 and nos. 5-15 in irrep 2 from the input orbitals, no questions asked. The input orbitals will be the SCF orbitals in this case (automatically selected).
The shape of the orbitals will change during the optimization, and orbital 48 may look quite different at the beginning and at the end, it may end up being closer to some other initial orbital.
Offline
#3 2025-11-05 11:13:54
- Lukhmanul Hakeem k
- Member
- Registered: 2024-02-05
- Posts: 45
Re: Clarification on Orbital Selection and Rotation in CASSCF Calculation
Dear Sir,
Thank you for your explanation. I have one more doubt.
In the RASSCF output, I found the following lines:
Max non-diagonal density matrix element 8.571E-05
Maximum BLB matrix element -4.435E-06
(orbital pair 48, 54 in symmetry 1)
Norm of electronic gradient 4.821E-02
What does the line “orbital pair 48, 54 in symmetry 1” mean?
Does it indicate that the final shapes of orbitals 48 are similar to orbital 54?
Kind regards,
Lukhmanul Hakeem K.
Offline
#4 2025-11-05 14:30:10
- Ignacio
- Administrator
- From: Uppsala
- Registered: 2015-11-03
- Posts: 1,165
Re: Clarification on Orbital Selection and Rotation in CASSCF Calculation
It's the maximum off-diagonal element, which should approach zero on convergence. It doesn't tell you much.
Offline