[molpro-user] BUG report: MCSCF VB output
Michael Nix
M.G.Nix at leeds.ac.uk
Tue Jul 16 17:32:06 BST 2013
Dear all,
Thanks for the responses and advice. On closer inspection it seems that the VB actually works fine and outputs all the state energies correctly after the CI vectors at the end of the calculation. However, higher up in the output, where each state is given individually in the form:
-------------
Results for state 1.1
=====================
!MCSCF STATE 1.1 Energy -76.082673308319
Nuclear energy 8.90325418
Kinetic energy 75.93525551
One electron energy -122.39794837
Two electron energy 37.41202088
Virial ratio 2.00194136
!MCSCF STATE 1.1 Dipole moment 0.00000000 0.00000016 0.76955895
Dipole moment /Debye 0.00000000 0.00000042 1.95589563
-----------------
There seems to be a problem. The parameters reported for each state identical and are in fact those of the state averaged wavefunction (energy matches exactly the SA_VB_CAS energy). Therefore i think this is just a bug in the output code that reads the results of the VB and prints them in the output file. This only seems to affect variational VB called from the MULTI routine.
Unfortunately it also affects the internal variables so calling the variable names subsequently to make a table of energies does not work either. Perhaps this could go on the bug list?
Many thanks,
Mike
-----Original Message-----
From: molpro-user-bounces at molpro.net [mailto:molpro-user-bounces at molpro.net] On Behalf Of molpro-user-request at molpro.net
Sent: 16 July 2013 15:53
To: molpro-user at molpro.net
Subject: Molpro-user Digest, Vol 60, Issue 21
Send Molpro-user mailing list submissions to
molpro-user at molpro.net
To subscribe or unsubscribe via the World Wide Web, visit
http://www.molpro.net/mailman/listinfo/molpro-user
or, via email, send a message with subject or body 'help' to
molpro-user-request at molpro.net
You can reach the person managing the list at
molpro-user-owner at molpro.net
When replying, please edit your Subject line so it is more specific than "Re: Contents of Molpro-user digest..."
Today's Topics:
1. CASVB multiple states (Michael Nix)
----------------------------------------------------------------------
Message: 1
Date: Tue, 16 Jul 2013 15:50:57 +0100
From: Michael Nix <M.G.Nix at leeds.ac.uk>
To: "'molpro-user at molpro.net'" <molpro-user at molpro.net>
Subject: [molpro-user] CASVB multiple states
Message-ID:
<20F23C738817FE44A28276050BC5787502F47CC3AF2F at HERMES8.ds.leeds.ac.uk>
Content-Type: text/plain; charset="us-ascii"
Dear Molpro users,
Does anyone know if it is possible to get several states from a fully variational VB wavefunction calculation. I have tried with water as a test case:
-------------INPUT-----------
hf
{multi
wf,10,1,0
state,2
locorb
}
{multi
wf,10,1,0
state,2
vb
}
------------------------------
The calculation runs seemingly ok, but the energies of both the states from the VB calculation are identical (see output below)
--------OUPUT-------------
1PROGRAM * MULTI (Direct Multiconfiguration SCF) Authors: P.J. Knowles, H.-J. Werner (1984) S.T. Elbert (1988)
Number of closed-shell orbitals: 1 ( 1 )
Number of active orbitals: 6 ( 6 )
Number of external orbitals: 34 ( 34 )
State symmetry 1
Number of electrons: 8 Spin symmetry=Singlet Space symmetry=1
Number of states: 2
Number of CSFs: 105 (225 determinants, 225 intermediate states)
1PROGRAM * CASVB (Valence bond MCSCF) Authors: T. Thorsteinsson, D.L. Cooper (1996-2006)
Number of active electrons : 8
active orbitals : 6
Total spin : 0.0
State symmetry : 1
CI vector dump at record 3500.2
Molecular orbitals read from record 2140.2 Type=MCSCF/LOCALIZED (state averaged)
EXTRA SYMMETRY OF MOS IN SYMMETRY 1: 1 1 1 1 2 1 1 1 1 2 1 1 1 1 2 2 1 1 1 1 1 2 1 2 1 1 2 1 2 1 2 1 1 1 1 1 2 2 1 1
1
Wavefunction dump at record 2141.2
Convergence thresholds 0.10E-01 (gradient) 0.10E-05 (energy) 0.10E-02 (step length)
Weight factors for state symmetry 1: 0.50000 0.50000
Number of orbital rotations: 244 ( 6 Core/Active 34 Core/Virtual 0 Active/Active 204 Active/Virtual)
Total number of variables: 694
ITER. MIC NCI NEG ENERGY(VAR) ENERGY(PROJ) ENERGY CHANGE GRAD(0) GRAD(ORB) GRAD(CI) STEP TIME
1 140 0 0 -76.05206612 -76.07266226 -0.02059614 0.15314356 0.00193678 0.00000000 0.81D+00 0.26
2 27 0 0 -76.08094348 -76.08230019 -0.00135672 0.04037377 0.00000422 0.00000000 0.13D+00 0.35
3 27 0 0 -76.08258017 -76.08264194 -0.00006177 0.01194359 0.00000761 0.00000000 0.23D-01 0.42
4 27 0 0 -76.08266110 -76.08266831 -0.00000722 0.00429805 0.00000304 0.00000000 0.76D-02 0.48
5 27 0 0 -76.08267124 -76.08267245 -0.00000121 0.00167699 0.00000131 0.00000000 0.33D-02 0.53
6 27 0 0 -76.08267295 -76.08267317 -0.00000021 0.00067493 0.00000054 0.00000000 0.14D-02 0.57
7 28 0 0 -76.08267326 -76.08267329 -0.00000004 0.00027557 0.00000022 0.00000000 0.62D-03 0.60
** WVFN **** CONVERGENCE REACHED, FINAL GRADIENT: 0.11D-03
CASVB -- summary of results :
-----------------------------
Spatial VB configurations
-------------------------
Conf. => Orbitals
1 => 1 1 2 2 3 4 5 6
Number of VB configurations : 1
VB structures : 2
VB determinants : 6
== Starting optimization - step 1 ========
Energy-based optimization (Evb).
Optimization algorithm: dFletch
Maximum number of iterations: 50
Spin basis: Kotani
===========================================
Generate Kotani spin functions.
Converged ... maximum update to coefficient: 0.258742D-07 Final Evb : -76.0826733083
Number of iterations used: 1
CASVB -- summary of results :
-----------------------------
== Wavefunction summary - step 2 =========
Orbital coefficients :
----------------------
1 2 3 4 5 6
1 -0.0911820 -0.2531360 -0.2289313 -0.3797013 0.2303814 -0.3799745
2 0.2444897 0.6787421 -0.4572484 0.6986149 0.8081230 0.3118124
3 -0.2445503 -0.6789103 0.8085066 -0.3119850 -0.4566062 -0.6986260
4 -0.9337670 -0.1135207 -0.2680998 0.2580340 0.2689622 0.2578063
5 -0.0096241 -0.0267182 -0.1119590 -0.4466633 0.0206513 0.0659489
6 0.0096069 0.0266703 0.0204725 -0.0658775 -0.1122727 0.4466551
Overlap between orbitals :
--------------------------
1 2 3 4 5 6
1 1.0000000 0.4615698 -0.0370216 0.0444443 0.0358089 0.0446562
2 0.4615698 1.0000000 -0.7673342 0.7629901 0.7661051 0.7630138
3 -0.0370216 -0.7673342 1.0000000 -0.5052765 -0.8681430 -0.6877886
4 0.0444443 0.7629901 -0.5052765 1.0000000 0.6871184 0.5877157
5 0.0358089 0.7661051 -0.8681430 0.6871184 1.0000000 0.5039956
6 0.0446562 0.7630138 -0.6877886 0.5877157 0.5039956 1.0000000
Structure coefficients :
------------------------
-0.17019896 0.98540972
Saving VB wavefunction to record : 3208.2
Evb : -76.0826733083
Chirgwin-Coulson weights of structures :
----------------------------------------
VB spin+space (norm 1.00000000) :
-0.04226258 1.04226258
VB spin only (norm 118.92355302) :
0.02896769 0.97103231
Total number of structure transformations : 2495
Total number of Hamiltonian applications : 1188
Total number of 2-electron density evaluations : 74
Total number of Hessian applications : 1077
Total number of pure orbital Hessian applications : 455
Approximate memory usage (8-byte words) : 10094
CASVB completed on 16-Jul-13 at 15:39:35 after 0.370 CPU seconds
Results for state 1.1
=====================
!MCSCF STATE 1.1 Energy -76.082673308319
Nuclear energy 8.90325418
Kinetic energy 75.93525551
One electron energy -122.39794837
Two electron energy 37.41202088
Virial ratio 2.00194136
!MCSCF STATE 1.1 Dipole moment 0.00000000 0.00000016 0.76955895
Dipole moment /Debye 0.00000000 0.00000042 1.95589563
Results for state 2.1
=====================
!MCSCF STATE 2.1 Energy -76.082673308319
Nuclear energy 8.90325418
Kinetic energy 75.93525551
One electron energy -122.39794837
Two electron energy 37.41202088
Virial ratio 2.00194136
!MCSCF STATE 2.1 Dipole moment 0.00000000 0.00000016 0.76955895
Dipole moment /Debye 0.00000000 0.00000042 1.95589563
State-averaged charge density matrix saved on record 2141.2 (density set 1)
!MCSCF expec <1.1|DMY|1.1> 0.000000164276 au = 0.000000417521 Debye
!MCSCF trans <2.1|DMY|1.1> 0.000000172427 au = 0.000000438236 Debye
!MCSCF expec <2.1|DMY|2.1> 0.000000164276 au = 0.000000417521 Debye
!MCSCF expec <1.1|DMZ|1.1> 0.769558945693 au = 1.955895625196 Debye
!MCSCF trans <2.1|DMZ|1.1> -0.552675809450 au = -1.404669783781 Debye
!MCSCF expec <2.1|DMZ|2.1> 0.769558945693 au = 1.955895625196 Debye
--------------------
The energies of the states in the preceding MO based MULTI are different and correct.
Any ideas what i'm doing wrong here?
Thanks
Mike
Dr. Mike Nix
Lecturer in Physical Chemistry
University of Leeds
LS2 9JT
+44(0)113 343 5960
m.g.nix at leeds.ac.uk
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20130716/7d436214/attachment.html>
------------------------------
_______________________________________________
Molpro-user mailing list
Molpro-user at molpro.net
http://www.molpro.net/mailman/listinfo/molpro-user
End of Molpro-user Digest, Vol 60, Issue 21
*******************************************
More information about the Molpro-user
mailing list