Re: Difficult optimization in MOPAC
I thank everyone who sent me suggestions on how to optimize the geometry
of C6H5-COO radical using RHF techniques. The best suggestion was given
by Dr John McKelvey. His solution to the problem is given in the following
ARC file from MOPAC 93. The GNORM was reduced to below the pre-specified
limit of 0.01, which is definitive proof of the validity of the optimization.
About 5-7 days had been spent on attempting the optimization, without any
success.
Within a day of posting the problem, I had received the correct answer. This
demonstrates the usefulness of the Net, and the skill of its users.
Jimmy Stewart
SUMMARY OF PM3 CALCULATION
MOPAC 93.00
C7 H5 O2
Thu Jan 27 09:28:47 1994
GNORM=0.01 GRADIENTS PRECISE PM3 SHIFT=50 EF C.I.=(3,2)
C6H5-COO(.)
GEOMETRY OPTIMISED USING EIGENVECTOR FOLLOWING (EF).
SCF FIELD WAS ACHIEVED
HEAT OF FORMATION = -10.726466 KCAL = -44.87954 KJ
ELECTRONIC ENERGY = -6178.829220 EV STATE: DOUBLET B2
CORE-CORE REPULSION = 4687.367953 EV
GRADIENT NORM = 0.009496
DIPOLE = 4.39069 DEBYE SYMMETRY: C2v
NO. OF FILLED LEVELS = 22
AND NO. OF OPEN LEVELS = 1
CONFIGURATION INTERACTION WAS USED
IONIZATION POTENTIAL = 10.354278 EV
HOMO (SOMO) LUMO (EV) = -10.439 ( -7.018) -0.921
MOLECULAR WEIGHT = 121.115
SCF CALCULATIONS = 17
COMPUTATION TIME = 12 MINUTES AND 10.633 SECONDS
FINAL GEOMETRY OBTAINED CHARGE
GNORM=0.01 GRADIENTS PRECISE PM3 SHIFT=50 EF C.I.=(3,2)
C6H5-COO(.)
C 0.00000000 0 0.0000000 0 0.0000000 0 0 0 0 -0.0563
C 2.78147112 1 0.0000000 0 0.0000000 0 1 0 0 -0.1536
C 1.39164098 1 60.0624509 1 0.0000000 0 1 2 0 -0.1172
C 1.39164180 1 60.0622252 1 179.9992100 1 1 2 3 -0.1172
C 1.39738087 1 60.0816056 1 -0.0001412 1 2 1 3 -0.0343
C 1.39737820 1 60.0816087 1 -179.9989763 1 2 1 3 -0.0343
H 1.09548486 1 179.9966649 1 -103.5372856 1 1 2 3 0.1071
H 1.09537441 1 120.0286616 1 -179.9985171 1 3 1 2 0.1124
H 1.09537460 1 120.0285221 1 179.9985854 1 4 1 2 0.1124
H 1.09678871 1 120.1471184 1 179.9997238 1 5 2 1 0.1195
H 1.09678839 1 120.1472596 1 -179.9983554 1 6 2 1 0.1195
C 1.47024274 1 179.9997536 1 3.4455962 1 2 1 3 0.4514
O 1.26492843 1 130.2209371 1 -179.9902413 1 12 2 5 -0.2548
O 1.26492930 1 130.2194911 1 0.0089828 1 12 2 5 -0.2548
From billg%scg. at.at at.at ccl.net Thu Jan 27 22:45:00 1994
Received: from mail.barrnet.net for billg%scg. at.at at.at ccl.net
by www.ccl.net (8.6.4/930601.1506) id WAA08525; Thu, 27 Jan 1994 22:41:48 -0500
Received: from fujitsu1.fai.com by mail.barrnet.net (5.67/1.37)
id AA18612; Thu, 27 Jan 94 19:41:34 -0800
Received: from scg.scg.fai.com by fujitsu.fai.com (4.1/SMI-4.1)
id AA10706; Thu, 27 Jan 94 19:41:13 PST
Received: by scg.scg.fai.com (4.1/SMI-4.1)
id AA05161; Thu, 27 Jan 94 19:41:51 PST
Date: Thu, 27 Jan 94 19:41:51 PST
From: billg at.at scg.fai.com (Marketing)
Message-Id: <9401280341.AA05161 at.at scg.scg.fai.com>
To: admiraal at.at bio.vu.nl
Subject: Perturbational MO References
Cc: chemistry at.at ccl.net
>I need information on using perturbuation theory (? I don't
>know if this is the correct name)
>for modelling interactions between organic molecules.
>Thanks, Alex
Here are three references on the subject of perturbational
molecular orbital (PMO) theory.
1) Dewar, M. J. S., "The Molecular Orbital Theory of
Organic Chemistry"; McGraw--Hill: New York, 1969.
2) Fukui, K. Acc. Chem. Res., 1971, vol. 4, p. 57.
3) Fleming, I. Frontier Orbitals and Organic Chemical
Reactions"; Wiley-Interscience: London.
I believe there is also a good reference by L. Salem
which I cannot recall.
PMO theory had its halcyon days about 20 years ago
before computational chemistry codes and platforms were
widely available. Today variational semiempirical (e.g., MNDO
PM3, etc.) codes are often used instead since they can
locate transition states (at least in principle) and
even entire reaction pathways. Yet, for sheer insight into
many types of concerted reactions, PMO theory offers a lot of
bang for the buck!
*********************************************************************
* William A. Glauser, Ph.D. | Internet: billg at.at scg.fai.com *
* Computational Chemist | Voice: (610) 436-8125 *
* SuperComputer Group | FAX: (610) 430-1526 *
* Fujitsu America, Inc. |-----------------------------------*
* 1564 McDaniel Drive | *
* West Chester, PA 19380 | *
*********************************************************************