From ahocquet@logatome.micronet.fr Fri May 10 16:10 EDT 1996 Received: from logatome.micronet.fr for ahocquet@logatome.micronet.fr by www.ccl.net (8.7.1/950822.1) id QAA01159; Fri, 10 May 1996 16:10:13 -0400 (EDT) Received: from ppp02.micronet.fr (ppp02.micronet.fr [194.51.75.134]) by logatome.micronet.fr (8.7.5/8.7.3) with SMTP id WAA08647; Fri, 10 May 1996 22:10:01 +0200 (MET DST) Date: Fri, 10 May 1996 22:10:01 +0200 (MET DST) Message-Id: <199605102010.WAA08647@logatome.micronet.fr> X-Sender: ahocquet@mail.Micronet.fr X-Mailer: Windows Eudora Version 1.4.4 Mime-Version: 1.0 Content-Transfer-Encoding: 8bit To: chemistry@www.ccl.net From: ahocquet@micronet.fr (Alexandre HOCQUET) Subject: Cyclohexa1,4diene summary : from Boyd and Lipkowitz to us dumb users Cc: hocquet@cas.lca.espci.fr Content-Type: text/plain; charset="iso-8859-1" Content-Length: 18022 Status: R Cyclohexa1,4diene summary : from Boyd and Lipkowitz to us dumb users Dear CCLers, Under that title you will find the summary of the answers to the message "another dumb question of molecular mechanics". The initial question is first listed, then the contributions of the cclers. First, it appears that Cyclohexa1,4diene experimentally exists in the two conformations according to experimental results from Xray and NMR cited in Boyd and Lipkowitz, although the authors comments are brief. According to Eliel's "Stereochemistry of organic compounds", Cyclohexa1,4diene planar form is the most stable but the potential well is "shallow". Idid not see further comments. The reference cited therein is a book "conformational analysis of cyclohexadiene..." from Lipkowitz and Rabideau, which i could not reach. Although semiempirical methods unambiguously yield planar conformation, molecular mechanics vary greatly according to the software used .... even with the (apparently) same forcefields ! In some cases, planar form is considered to be the global minimum, in others it is the boat form. If frequency vibrations are calculed, some find a true minimum, others a saddle point. If, as Eliel says, the potentiel well is shallow, then we can imagine that according to the force field parameterization, or the minimization technique, the PES, as seen by the force field may correspond to a minimum, a saddle point or even not a minimum at all. As far as Hyperchem is concerned, a parameterization bug has been enlightened and the scientific staff is being resolving it. Finally, two people gave results and references concerning...... cyclohexa1,3diene, the conjugated one. They are listed at the end of this summary _________________________________________________________________ In their molecular mechanics principles review (J. Chem. Ed. , 59, 269-274, 1982) Boyd and Lipkowitz present the conformations of cyclohexa-1,4-diene as an example of supremacy of molecular mechanics upon hand held molecular models : the dreiding model predicts boat geometry whereas molecular mechanics calculate it planar. I do believe that this example is demonstrative, presenting molecular mechanics to students as a sophisticated hand held model. Unfortunately, as i tried to minimize cyclohexa-1,4-diene with the MM+ force field provided by Hyperchem, the results show that the minimum (9.10 kcal.mol-1) corresponds to a torsion angle between 1,2,3 and 4 carbons of 26°, in a boat conformation. The constrained planar structure corresponds to 9.73 kcal.mol-1 and goes back to the former conformation when constrain is removed. Besides, An AM1 semiempirical calculation does find the minimum to be planar. So my questions are : Does the MM+ forcefield, or the parameters provided in Hyperchem, lack the accuracy needed to perform this calculation ? Does any references about molecular mechanics calculations of this molecule exist ? Does anybody know the references about calculations, Xray and NMR results that Boyd and Lipkowitz cite with no references (or, to be honnest, the reference cited was "in press" : Lipkowitz et al. in J. Org. Chem.) ? Am i doing something wrong ? Thanks in advance for any answers (if somebody is interested, i will summarize the answers) ____________________________________________________________________________ _____ Hi I tried this minimization with Hyperchem 4.5 using MM+: I started with the model builder which gave a boat conformation and then with MM+ and obtained the non planar structure as a flattened boat and Energy=8.977 Kcal/mol (Gradient 0.00079). I then took that "optimized" structure and reminimized it with PCModel for Windows 5.13 where it immediately gave the planar structure! I went back to Hyperchem redrew the structure "PLANAR" and added H's WITHOUT using the model builder. I minimized this structure and now I obtained the planar conformation with Energy= 3.0488 Kcal/mol (Gradient 0,00083), i.e the real minimun. I rechecked the "non planar minimun" that HyperChem found, and in both PCModel AND Hyperchem it has a very high torsion energy component, however for some strange reason Hyperchem does not reflect that in the Gradient. I made some other tests and found that Hyperchem reached another boat type minimun with Energy=10.94!! This structure rapidly converged to the planar real minimun with PCModel. I never found differences between the parameters in Hyperchem and PCModel, maybe this is a bug in Hyperchem optimizers. As I see it, optimizers are the weakest point of Hyperchem (an implementation of BFGS would be great!) Gerardo Gerardo Burton e-mail: burton@quimor.qo.fcen.uba.ar Departamento de Quimica Organica Facultad de Ciencias Exactas y Naturales Universidad de Buenos Aires Tel/FAX 54 1 788-6915 _____________________________________________________________________________ Hola Alexandre revisando mis "pruebas" encontre que en aquellas en que Hyperchem dio el ciclohexadieno plano con energia de aprox 3 kcal/mol estaba usando el "default parameters", resultado de agregar H sin usar el Model Builder. Sin embargo si a esa estructura plana se le hace "BUILD/Calculate Types" y se optimiza normalmente mantiene la conformacion plana usando la opcion de carga/carga en MM+. Con "bond dipoles" en Hyperchem 4.5 sigue dando plana pero en Hyperchem 3.0 se curva ligeramente. Te envio el archivo .hin a continuacion correspondiente a la estructura plana optimizada con MM+ para que lo pruebes. Gerardo ;Cyclohexadiene planar conformer ;Energy 9.61 kcal/mol Grad 0.00093 ;MM+ Charge/Charge interaction forcefield mm+ sys 0 view 40 0.2308208 55 15 0.08100456 -0.02507711 0.9963982 0.7050237 0.7080831 -0.03949571 -0.7045423 0.7056837 0.07503794 -0.1855706 -0.265454 -54.1561 mol 1 atom 1 - C C4 - 0 -0.8421649 1.854515 0.05424488 4 2 s 6 s 7 s 8 s atom 2 - C C3 - 0 -2.06936 0.9877079 0.01674857 3 1 s 3 d 9 s atom 3 - C C3 - 0 -2.027517 -0.3502076 -0.04506508 3 2 d 4 s 10 s atom 4 - C C4 - 0 -0.7485514 -1.138567 -0.08400068 4 3 s 5 s 11 s 12 s atom 5 - C C3 - 0 0.4786466 -0.2717942 -0.04571453 3 4 s 6 d 13 s atom 6 - C C3 - 0 0.4368019 1.066121 0.01611715 3 5 d 1 s 14 s atom 7 - H H - 0 -0.8610181 2.464536 0.9870859 1 1 s atom 8 - H H - 0 -0.8640453 2.546919 -0.8191209 1 1 s atom 9 - H H - 0 -3.04832 1.497608 0.04210231 1 2 s atom 10 - H H - 0 -2.972697 -0.9203283 -0.06959744 1 3 s atom 11 - H H - 0 -0.7297372 -1.747683 -1.017434 1 4 s atom 12 - H H - 0 -0.726633 -1.831819 0.7886948 1 4 s atom 13 - H H - 0 1.457608 -0.7817122 -0.07074484 1 5 s atom 14 - H H - 0 1.381982 1.636223 0.04099056 1 6 s endmol 1 _______________________________________________________________________________ Muchas gracias por tu contesta veloz, Gerardo. Estoy haciendo esfuerzos desperados para alcanzar este minimum plano del 1,4 dihidrobenzeno, pero no puedo conseguirlo. Con metodo semiempirico AM1 o PM3, la conformacion plana esta imediata, pero incluso con una estructura inicial obtenida por esta manera, MM+ regresa a una conformacion "boat". Traté de seguir tu metodo con el "builder" de Hyperchem, pero no funciono : construi un benzeno para obtener un ciclo plano y , despues de anadir hidrogenos y cambiar los ordenes de enlace, cambié los tipos de atomos de CA a C4 y C3. Finalmente, en el libro "stereochemistry of organic compounds" de Eliel y Wilen, he leido que "the potential well corresponding to the planar minimum is very shallow", entonces, parece como una deficiencia del "optimiser" Saludos Alexandre Hocquet Laboratoire de Chimie Analytique ESPCI 10, Rue Vauquelin 75231 Paris Cedex 05 France tel : 33 1 40794420 fax : 33 1 40794425 email : ahocquet@micronet.fr ou hocquet@cas.espci.fr ou hocquet@ens-cachan.fr ____________________________________________________________________ I just tried 1,4-cyclohexadiene using SPARTAN. Usint the MM+ force field I get a strain energy o +1.08. Kcals/mol and an HF (AM1) of 16.998 Kcal/mol. Does this fit with your calculations? JIm Macmillan James G. Macmillan Department of Chemistry University of Northern Iowa Cedar Falls, IA 50614 E-Mail: macmillan@uni.edu Ph: (319)273-2476 FAX; (319)273-7127 ========================================================================== Where good wine flows, so does good conversation. _________________________________________________________________ Dear Alexandre, this is not a specific answer to your problem, but illustration of similar (?) problem. I tried the MM2 force field in CS Chem3D Pro, which predicts 1,4-cyclohexadiene to be planar. on the other hand, I also built a carefully-constructed flattened version of cyclohexane to illustrate how molecular mechanics finds local energy-minima (eg boat form for cyclohexane if the initail geometry is not near the global minimum). run 1: CS Chem3D Pro found the boat conformer run 2: I rotated the initial geometry (same bond lengths and angles as run 1) CS Chem3D Pro then found the chair conformer! In this case I suspect some numerical round-off instability. For research work, I'd recommend (eg) QCPE-type source code. In your case, do you know if it is MM1 , MM2, or MM3 force field? These are 3 very similar, but slightly different parameterisations. Or is MM+ HyperChem's own parameterisation? Could that be the problem? Kieran ---------------------------------------------------------- Dr Kieran F Lim Biol. and Chemical Sciences ph: + [61] (52) 27-2146 Deakin University fax: + [61] (52) 27-1040 Geelong VIC 3217 email: lim@deakin.edu.au AUSTRALIA WWW: http://www.deakin.edu.au/~lim/ ____________________________________________________________________ We had occasion to look at the 1,3-cyclohexadiene problem with AM1 mentioned by Alexandre HOCQUET. It is really interesting and a fine pedagocgical example for students. HOCQUET's conclusion that AM1 gives the wrong conformation, the planar conformation, is not true. However, this is a commonly observed result. The calculation EASILY FINDS THE WRONG CONFORMATION. The wrong conformation is not an energy minimum but rather a maximum at the transition state. That is what makes it so interesting. This is nicely evident by calculating and viewing the IR spectrum, using for example the CAChe system. The IR spectrum of the planar structure contains an absorption at a NEGATIVE frequency. Inspection of the nature of the vibration indicates the distortion necessary to find the proper nonplanar conformation which is lower in energy. James E. Gano, Director Instrumentation Center in Arts and Sciences Bowman-Oddy Laboratories University of Toledo Toledo, Ohio 43606 Instrumentation Center : http://www.icenter.utoledo.edu Department of Chemistry: http://www.chem.utoledo.edu 419-530-7847 (419-530-2104) 419-530-4033 (FAX) _________________________________________________________________ An obscure bug has been found in the MM+ force field for HyperChem. The effect is that under certain circumstances, HyperChem reads a set of parameters for one torsional angle calculation, and uses the same parameters for another torsional angle as well, instead of getting new parameters. Further details are available on the HyperChem mailing list and on our WWW page (http://www.hyper.com). We would like to thank Alexandre Hocquet for bringing the problem to our attention. We expect to be issuing corrected versions of the MM+ backend to all of our users to correct the problem. Further details will be available soon on our mailing list and WWW site. Joel ------------ Joel Polowin, Ph.D. Manager, Scientific Support Email to: polowin@hyper.com Hypercube Inc, 419 Phillip St, Waterloo, Ont, Canada N2L 3X2 (519)725-4040 Info requests to: info@hyper.com Support questions to: support@hyper.com Email group: Send "subscribe hyperchem" to hyperchem-request@hyper.com WWW: http://www.hyper.com/ ____________________________________________________________________ Dear CCLers and Hyperchemers, Re: Alexandre Hocquet question on the molecular mechanics optimized structure of cyclohexa-1,4-diene I have had a quick look at this problem using Nemesis (COSMIC MM force field) and also found that the pseudo-boat (torsion 25 degrees) is predicted to be more stable than the flat structure by approximately 0.7 kcal/mol. However if the optimizing is started from a flat structure (modifing benzene)than this is not kinked by Nemesis. /\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\ Dr. Jeff Gosper Dept. of Chemistry BRUNEL University Uxbridge Middx UB8 3PH, UK voice: 01895 274000 x2187 facsim: 01895 256844 internet/email/work: Jeffrey.Gosper@brunel.ac.uk internet/WWW: http://http1.brunel.ac.uk:8080/~castjjg Re_View's Home page (A molecular display/animation/analysis program): http://http1.brunel.ac.uk:8080/depts/chem/ch241s/re_view/re_view.htm \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ _______________________________________________________________ Alexandre Hocquet, Here is what I have found using SPARTAN. MM2 with Sybyl FF gives a planar structure of energy 1.081245 Kcal/mol. AM1 starting from Sybyl geometry is still planar with an energy of 19.733 Kcal/mol. A frequency calculation gives on imaginary vibration at 191.76 of B3u symmetry. AM1 starting from Sybyl geometry has an energy of -229.037790 hartrees and a frequency calculation does not show and imaginay vibrations although the B3u vibrationa at 125.97 looks to be the same as the imaginary on from AM1. I am not sure why there is a difference between HF ab initio and AM1 semi empirical. Jim ========================================================================== James G. Macmillan Department of Chemistry University of Northern Iowa Cedar Falls, IA 50614 E-Mail: macmillan@uni.edu Ph: (319)273-2476 FAX; (319)273-7127 ========================================================================== Where good wine flows, so does good conversation. _________________________________________________________________ 1996 May 7 Hello, There has been a discussion of the conformation of 1,3-cyclohexadiene as calculated by MM and AM1. I got these results, using Spartan: MM jobs: 1 C2 (nonplanar ring) input from the builder, opt with MM3-->C2, 9.12 kcal/mol 2 Cs (planar ring) input from the builder, opt with MM3-->C2v, 10.38 kcal/mol AM1 jobs: 3 C2 input from MM3, opt with AM1--> C2, 17.49 kcal/mol, no imaginary freqs 4 C2v (planar ring) input from MM3, opt with AM1--> C2v, 17.64 kcal/mol, 1 imag. freq, 74i cm-1 ("-74" cm-1) Now, an MM energy cannot be compared with an AM1 energy, since they refer to different things, but within each class of computation the C2 (nonplanar ring) geometry has the lower energy, and the AM1 C2 geom is a relative minimum on the potential energy surface, since it has no imaginary frequencies. The "small" value of the imaginary freq of the AM1 C2v structure shows that this AM1 transition state is "flat" (not steeply curved along the reaction coordinate), as might have been suspected from the very small activation energy, and animating the vibration reveals that the planar C2v molecule wants to twist to the C2 geometry. This little exercise shows (1) the importance of the _input_ geom; an MM job can throw an input structure into a higher sym, but a semiempirical or ab initio calc will keep the input symmetry (2) the importance of doing a freq job on an optimized structure, to see if it is a relative minimum or some kind of saddle point (e.g. a transition state=1st-order saddle point) ON THAT PARTICULAR PES, which is not necessarily the _true_ PES. Most MM programs do not yet do freq calcs, altho' I think the latest version of MM3 does. Does any one know what _experiment_ tells us about the shape of 1,3-hexadiene? E. Lewars Chemistry Dept Trent U, Ontario, CANADA -------This is added Automatically by the Software-------- -- Original Sender Envelope Address: elewars@alchemy.chem.utoronto.ca -- Original Sender From: Address: elewars@alchemy.chem.utoronto.ca CHEMISTRY@www.ccl.net: Everybody | CHEMISTRY-REQUEST@www.ccl.net: Coordinator MAILSERV@www.ccl.net: HELP CHEMISTRY or HELP SEARCH | Gopher: www.ccl.net 73 Anon. ftp: www.ccl.net | CHEMISTRY-SEARCH@www.ccl.net -- archive search Web: http://www.ccl.net/chemistry.html _________________________________________________________________ Hi, I haven't followed the whole discourse on MM and AM1 calcs on 1,3-cyclohexadiene, so I will be brief. Microwave and electron diffraction studies on 1,3-cyclohexadiene are available. The structure has C2 symmetry and is well represented at the RHF/6-31G(d) level. While in the Houk group, I and others looked at the electrocyclic reactions of substituted 1,3-cyclohexadienes and at all levels of theory employed (including AM1) these structures were staggered about the sp3-sp3 single bond. The microwave and electron diffraction references are: Oberhammer, H. ... JACS 1969, 91, 10 Dallinga, G. ... J. Mol. Struc. 1968, 1, 11 Traetteberg, M. Acta. Chem. Scand. 1968, 22, 2305 Butcher, S. S. J. Chem. Phys. 1965, 42, 1830 I hope this helps. Bert -- _________________________________________________________ | | | Bert E. Thomas IV, Ph.D. bthomas@iris.procept.com | | Procept, Inc. (617)491-1100 ext 3040 | | Cambridge, MA 02139 | |_________________________________________________________| Alexandre Hocquet Laboratoire de Chimie Analytique ESPCI 10, Rue Vauquelin 75231 Paris Cedex 05 France tel : 33 1 40794420 fax : 33 1 40794425 email : ahocquet@micronet.fr ou hocquet@cas.espci.fr ou hocquet@ens-cachan.fr From lbh08545@amatl.dmcyp.udg.mx Thu May 9 22:06 EDT 1996 Received: from amatl.dmcyp.udg.mx for lbh08545@amatl.dmcyp.udg.mx by www.ccl.net (8.7.1/950822.1) id WAA25149; Thu, 9 May 1996 22:06:01 -0400 (EDT) Received: from [200.33.130.76] by amatl.dmcyp.udg.mx (4.1/SMI-4.1) id AA04573; Thu, 9 May 96 20:08:44 CST Date: Thu, 9 May 96 20:08:41 CST Message-Id: <1.5.4.16.19960509210440.1ce70116@amatl.dmcyp.udg.mx> X-Sender: lbh08545@amatl.dmcyp.udg.mx (Unverified) X-Mailer: Windows Eudora Light Version 1.5.4 (16) Mime-Version: 1.0 To: chemistry@www.ccl.net From: "Hugo H. Lozano B." Subject: Extraction liquid-liquid software Content-Type: text/plain; charset="us-ascii" Content-Length: 215 Status: RO Dear sirs: Anybody knows about software to liquid-liquid extraction for two or more solutes (WINDOWS,DOS or any program (BAsic,pascal,Mathlab, Mathcad,Mathematica)??? Please I need your information thanks From WSMITH@msnet.mathstat.uoguelph.ca Fri May 10 15:08 EDT 1996 Received: from ccshst06.cs.uoguelph.ca for WSMITH@msnet.mathstat.uoguelph.ca by www.ccl.net (8.7.1/950822.1) id PAA00796; Fri, 10 May 1996 15:08:52 -0400 (EDT) Received: from msnet.nw.uoguelph.ca (msnet.mathstat.uoguelph.ca) by ccshst06.cs.uoguelph.ca with ESMTP (1.37.109.16/16.2) id AA062025331; Fri, 10 May 1996 15:08:51 -0400 Received: from MSNET/MAILQ by msnet.nw.uoguelph.ca (Mercury 1.21); 10 May 96 15:11:57 GMT-5 Received: from MAILQ by MSNET (Mercury 1.21); 10 May 96 15:11:48 GMT-5 From: "William R. Smith" Organization: Dept. of Math & Stats. To: chemistry-request@www.ccl.net, chemistry@www.ccl.net Date: Fri, 10 May 1996 15:11:45 EDT Mime-Version: 1.0 Content-Transfer-Encoding: 7BIT Subject: Re: CCL:Re: CCL:Free Energy of Aluminium Hydroxide species X-Confirm-Reading-To: "William R. Smith" X-Pmrqc: 1 Priority: normal X-Mailer: Pegasus Mail v3.2 (pr2) Message-Id: <2BC8082546A@msnet.nw.uoguelph.ca> Content-Type: text/plain; charset=US-ASCII Content-Length: 1245 Status: R Hi, I too posted a reply, which also appeared to have vaporized. The reply was basically that I thought such data was available in standard compilations like the JANAF tables, Gurvich tables, etc. As a related point, I'm "theoretical chemist/physicist" who is _not_ a quantum chemist (stat mech actually), I'd like to ask/challenge the group with a question: Why not use Group Contribution Methods for such calculations? Surely their accuracy is as good as or better than the current state of the art for the QM approach. If I'm wrong, pls enlighten me. > I don't know if you received my message, since the copy to CCL did not arrive. > I repost an other copy of my response to your question. > Hope you will receive this message, also don't be worried if you receive this > message two times. This is to be sure you have it. This could help you. > Frederic ...snip Best Regards, W. R. Smith Professor Dept. of Mathematics and Statistics and School of Engineering University of Guelph FAX: 519-837-0221 Guelph, Ontario Tel: 519-824-4120, ext. 3038 CANADA N1G 2W1