From picard@ext.jussieu.fr Tue Dec 9 04:39:10 1997 Received: from shiva.jussieu.fr for picard@ext.jussieu.fr by www.ccl.net (8.8.3/950822.1) id EAA15331; Tue, 9 Dec 1997 04:31:55 -0500 (EST) Received: from idf.ext.jussieu.fr (idf.ext.jussieu.fr [134.157.81.129]) by shiva.jussieu.fr (8.8.8/jtpda-5.2.9) with ESMTP id KAA16713 for ; Tue, 9 Dec 1997 10:31:45 +0100 (CET) Received: from [134.157.11.19] by idf.ext.jussieu.fr (8.8.5/jtpda-4.0) with SMTP; Tue, 9 Dec 1997 10:31:42 +0100 (MET) Date: Tue, 9 Dec 1997 10:31:42 +0100 (MET) X-Sender: picard@idf.ext.jussieu.fr Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" To: chemistry@www.ccl.net From: picard@ext.jussieu.fr (G.S. PICARD) Subject: basis sets for Nb Dear Netters, I actually perform calculations on Nb compounds using ECPs and their corresponding optimized valence basis sets. Therefore, I would like to use several large valence basis sets for Nb (including polarization and diffuse functions). Does anybody have any references on such basis sets ? thank you for your help. Gerard S. PICARD , Directeur de Recherche au C.N.R.S., LABORATOIRE D'ELECTROCHIMIE ET DE CHIMIE ANALYTIQUE, Unite de Recherche associee au C.N.R.S. no 216, Equipe : "PROCESSUS INTERFACIAUX & REACTIVITE EN MILIEUX IONIQUES LIQUIDES" 11 rue Pierre et Marie Curie - 75231 Paris cedex 05 - FRANCE. Tel : (33) 1.55.42.63.89. Fax : (33) 1.44.27.67.50. WWW Home Page : http://alcyone.enscp.jussieu.fr/Pages/LECA/GP/ From dino@iris.inc.bme.hu Tue Dec 9 07:39:10 1997 Received: from nic.bme.hu for dino@iris.inc.bme.hu by www.ccl.net (8.8.3/950822.1) id HAA16279; Tue, 9 Dec 1997 07:31:37 -0500 (EST) Received: from ns.bme.hu (ns.bme.hu [152.66.116.1]) by nic.bme.hu (8.8.8/8.8.8) with ESMTP id NAA17523 for <@nic.bme.hu:chemistry@www.ccl.net>; Tue, 9 Dec 1997 13:31:30 +0100 (MET) Received: from iris.inc.bme.hu (iris.inc.bme.hu [152.66.63.5]) by ns.bme.hu (8.8.5/8.8.5) with SMTP id NAA08601 for <@nic.bme.hu:chemistry@www.ccl.net>; Tue, 9 Dec 1997 13:31:30 +0100 (MET) Received: by iris.inc.bme.hu (950413.SGI.8.6.12/920502.SGI.AUTO) for chemistry@www.ccl.net id NAA19804; Tue, 9 Dec 1997 13:30:09 +0100 From: "Szieberth Denes" Message-Id: <9712091330.ZM19802@iris.inc.bme.hu> Date: Tue, 9 Dec 1997 13:30:08 +0000 X-Mailer: Z-Mail (3.2.3 08feb96 MediaMail) To: chemistry@www.ccl.net Subject: irc summary Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Here' s a summary of the replies I've got to my question concerning Gaussian IRC calculations. Thanks for all who have replied. My original question was: > Several of my internal reaction coordinate -following optimum searches > run into the same problem: from a TS which has a negative frequency > apparently corresponding to a dissociation, the resulting minima in both > direction are the same. Has anyone experienced similar problems? Could it > be a problem of the IRC method implemented in gaussian? I'd be grateful for > any kind of help. ---------------------------------------------------------------------- ---------------------------------------------------------------------- did you use something like opt=(tight,ts)? if you don't use tight then g94 sometimes think that forward and reverse are the same direction. actually what happens is that your ts search doesn't end up at the ts itself but only near the ts. if the distance of your apparent ts structure to the actual ts is smaller than the stepsize of the irc than the irc will not cross the barrier and 'roll back'. in this case this is neither a bug in g94 nor a bug in your input but just an insufficient optimisation. hope this helps Ernst-Udo Wallenborn Phys. Chem. Lab --------------------------------------------------------- Hello! I am not sure this is the case with your calculations, but the symptoms are the same; sorry if this is wrong. If you have a molecule with some symmetry (for example, Cs), and the vibrational vector corresponding to the imaginary frequency is not of full-symmetric representation (A" for Cs), then your structures should be the same. Could it be your problem? I saw you saying "apparently correcponding to a dissociation". I assume you looked up your vector; please let me know when you find out what was the problem -- I am extensively using IRC, so I want to be aware of any possible bug. Thanks. Dmitry Khoroshun dima@euch4e.chem.emory.edu ---------------------------------------------------------------------- 1) I've found the IRC routines in G94 to work properly only from version D1 onward (see CCL archives for more information on this issue). For earlier versions the answer might be: upgrade. 2) To help the IRC it sometimes helped me to take big stepsizes in the beginning. Otherwise the IRC 'got stuck' around the TS. I have not encounterd your problem yet, but might it be related to the flatness of the PES around the TS? I look forward to a summary of the answers (hoping that some real IRC-gurus will also respond) Best regards, Han ______________________________________________________________________ Han Zuilhof | E-mail: ZUILHOF@SG1.OC.WAU.NL Laboratory of Organic Chemistry | FAX : 31-317-484914 Department of Biomolecular Sciences | phone : 31-317-482367 Wageningen Agricultural University | Dreijenplein 8 | 6703 HB Wageningen | "Excite a photochemist!" The Netherlands | ---------------------------------------------------------------------- I had a similar problem last year, which was never completely resolved as I moved before completing the project. You may be able to resolve this by optimizing with "verytight" convergence criteria, and using a small step size for the IRC search. You could also try contacting the developer of the IRC method, Prof. Schlegel, at hbs@sun.chem.wayne.edu Jaime Martell, Ph.D. P.O. Box 2317 Camille and Henry Dreyfus Fellow Batesville, AR 72503-2317 Biology and Chemistry Division Phone: 870-698-4688 Lyon College Fax: 870-698-4622 -------------------------------------------------------------------------- --------------------------------------------------------------------------- Eventually the usage of bigger stepsize, and in some cases simply the application of cartesian coordinates instead of the default mass-weighted coordinates have solved my problem. Denes Szieberth Dep. of Inorg. Chem. Technical University of Budapest Hungary dino@goliat.eik.bme.hu From chem8@york.ac.uk Tue Dec 9 08:10:48 1997 Received: from pump2.york.ac.uk for chem8@york.ac.uk by www.ccl.net (8.8.3/950822.1) id HAA16303; Tue, 9 Dec 1997 07:34:40 -0500 (EST) Received: from ebor.york.ac.uk (chem8@ebor.york.ac.uk [144.32.129.242]) by pump2.york.ac.uk (8.8.7/8.8.7) with SMTP id MAA24759; Tue, 9 Dec 1997 12:33:58 GMT Date: Tue, 9 Dec 1997 12:33:57 +0000 (GMT) From: John Waite To: "G.S. PICARD" cc: chemistry@www.ccl.net Subject: Re: CCL:basis sets for Nb In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi Gerrard, Try the below: CONTRACTED GTO BASIS FOR Nb STATE: 6D J.WAITE, THE NATIONAL HELLENIC RESEARCH FOUNDATION, ORGANIC CHEMISTRY INSTITUTE, VAS. KONSTANTINOU 48, ATHENS 116-35, GREECE EXPONENTS CONTRACTION COEFFICIENTS S S 9 1.0 2.596691005103630D+06 5.358116800858631D-05 3.923197781086183D+05 3.820844971788148D-04 8.974980977489868D+04 2.217679232489908D-03 2.559274259710869D+04 7.885484698343958D-03 8.475461936399899D+03 3.272314568765415D-02 3.132363613415850D+03 7.329863363453792D-02 1.252706012755098D+03 2.421923437194168D-01 5.314677476532034D+02 2.555472936082273D-01 2.395647012780761D+02 5.128897331050493D-01 S S 3 1.0 1.292359938977227D+02 -2.731573045052698D-01 5.498477092545167D+01 6.155795692141395D-01 1.949743450788932D+01 6.276786746406200D-01 S S 2 1.0 7.577247986046905D+00 3.591191920557027D-01 3.097600614864375D+00 6.730469618995687D-01 S S 2 1.0 1.466514788328507D+00 -4.719582301894687D-01 2.916763814515443D-01 -6.293315672661724D-01 S S 2 1.0 8.978643784453145D-02 3.930698040067477D-01 3.066469781932725D-02 6.538293069924649D-01 P P 8 1.0 1.388505254682974D+04 3.450362436273742D-04 3.292622810964910D+03 3.006090736978063D-03 1.069127544272230D+03 1.643771929173906D-02 4.074847231099110D+02 6.524332516168243D-02 1.717410356584460D+02 1.876948661761543D-01 7.716888813854490D+01 3.615831879236470D-01 3.617402120636559D+01 3.846160243554936D-01 1.715529346491350D+01 1.580090996267060D-01 P P 3 1.0 8.004658890066210D+00 6.952938858469084D-01 3.744749205598500D+00 2.144149245389234D-01 1.553912258357026D+00 1.873942595381799D-01 P P 2 1.0 6.797808290623076D-01 9.408905897208810D-01 2.670164012927867D-01 7.541229978838121D-02 D D 5 1.0 1.888236044576220D+02 1.728511885422323D-02 5.553357277836770D+01 1.122134700802013D-01 2.014537156643630D+01 3.371595115378190D-01 7.786872712100237D+00 4.950979536340579D-01 3.016509977875910D+00 2.814409523988776D-01 D D 2 1.0 8.302472356168830D-01 6.144111756819535D-01 2.146790698009120D-01 5.496697453380328D-01 ENERG= -0.3747659010507674D+04, V.T.= -0.20157572D+01, P.E.=-.7437181504D+04, K.E.=0.3689522494D+04, Nb (18S13P7D) SCF OPT + 1F MP2 OPT. 0,6 Nb 0.0 0.0 0.0 Nb 0 S 1 1.0 2.596691005103630D+06 1.0 S 1 1.0 3.923197781086183D+05 1.0 S 1 1.0 8.974980977489868D+04 1.0 S 1 1.0 2.559274259710869D+04 1.0 S 1 1.0 8.475461936399899D+03 1.0 S 1 1.0 3.132363613415850D+03 1.0 S 1 1.0 1.252706012755098D+03 1.0 S 1 1.0 5.314677476532034D+02 1.0 S 1 1.0 2.395647012780761D+02 1.0 S 1 1.0 1.292359938977227D+02 1.0 S 1 1.0 5.498477092545166D+01 1.0 S 1 1.0 1.949743450788932D+01 1.0 S 1 1.0 7.577247986046904D+00 1.0 S 1 1.0 3.097600614864375D+00 1.0 S 1 1.0 1.466514788328507D+00 1.0 S 1 1.0 2.916763814515443D-01 1.0 S 1 1.0 8.978643784453144D-02 1.0 S 1 1.0 3.066469781932725D-02 1.0 P 1 1.0 1.388505254682974D+04 1.0 P 1 1.0 3.292622810964910D+03 1.0 P 1 1.0 1.069127544272230D+03 1.0 P 1 1.0 4.074847231099110D+02 1.0 P 1 1.0 1.717410356584460D+02 1.0 P 1 1.0 7.716888813854490D+01 1.0 P 1 1.0 3.617402120636559D+01 1.0 P 1 1.0 1.715529346491350D+01 1.0 P 1 1.0 8.004658890066210D+00 1.0 P 1 1.0 3.744749205598500D+00 1.0 P 1 1.0 1.553912258357026D+00 1.0 P 1 1.0 6.797808290623075D-01 1.0 P 1 1.0 2.670164012927867D-01 1.0 D 1 1.0 1.888236044576220D+02 1.0 D 1 1.0 5.553357277836770D+01 1.0 D 1 1.0 2.014537156643630D+01 1.0 D 1 1.0 7.786872712100236D+00 1.0 D 1 1.0 3.016509977875910D+00 1.0 D 1 1.0 8.302472356168830D-01 1.0 D 1 1.0 2.146790698009120D-01 1.0 F 1 1.0 5.190640946385990D-01 1.0 **** SCF TOTAL ENERGY = -3753.52062441463022 DIPOLE MOMENTS: 0.00000000000000 0.00000000000000 0.00000000000000 MP2 ENERGY = -3753.55672435023007 Good Computing, John Dr. John Waite, e-mail: chem8@york.ac.uk * or The National Hellenic Research Foundation,* john@john1.eie.gr Organic and Pharaceutical Chemistry Institute, phone: ++30-1-7229811-5 X 1-810 Vas. Konstantinou 48, phone: ++30-1-7247913(secrtry. Mary) Athens 116-35, fax: ++30-1-7247913 Greece or NCRS "Democritos", phone: ++30-1-6513112-5 X219 * c/o Dr. G.Kordas, e-mail john@john.nrcps.ariadne-t.gr Material Science Institute, Aghia Paraskevi, Attikis, Athens 153-10, Greece On Tue, 9 Dec 1997, G.S. PICARD wrote: > Dear Netters, > > I actually perform calculations on Nb compounds using ECPs and their > corresponding optimized valence basis sets. > Therefore, I would like to use several large valence basis sets for Nb > (including polarization and diffuse functions). > > Does anybody have any references on such basis sets ? > > thank you for your help. > > Gerard S. PICARD , Directeur de Recherche au C.N.R.S., > LABORATOIRE D'ELECTROCHIMIE ET DE CHIMIE ANALYTIQUE, > Unite de Recherche associee au C.N.R.S. no 216, > Equipe : "PROCESSUS INTERFACIAUX & REACTIVITE EN MILIEUX IONIQUES LIQUIDES" > 11 rue Pierre et Marie Curie - 75231 Paris cedex 05 - FRANCE. > Tel : (33) 1.55.42.63.89. > Fax : (33) 1.44.27.67.50. > WWW Home Page : http://alcyone.enscp.jussieu.fr/Pages/LECA/GP/ > > > > -------This is added Automatically by the Software-------- > -- Original Sender Envelope Address: picard@ext.jussieu.fr > -- Original Sender From: Address: picard@ext.jussieu.fr > 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 > > From rassolov@chem.nwu.edu Tue Dec 9 10:39:11 1997 Received: from mercury.chem.nwu.edu for rassolov@chem.nwu.edu by www.ccl.net (8.8.3/950822.1) id KAA17142; Tue, 9 Dec 1997 10:17:56 -0500 (EST) Received: from [129.105.14.220] (potassium.chem.nwu.edu [129.105.14.220]) by mercury.chem.nwu.edu (8.8.5/8.8.5) with ESMTP id JAA23721; Tue, 9 Dec 1997 09:17:30 -0600 (CST) Message-Id: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Tue, 9 Dec 1997 09:17:40 -0600 To: chemistry@www.ccl.net From: "Vitaly A. Rassolov" Subject: correction on 6-31G(**) definition Cc: jsl@virgil.ruc.dk Dear CCLers: I want to apologize for a very silly mistake in the 6-31G set definition in my yesterday posting. Rather than "6 basis functions for each core orbital" I had to write "6 primitive exponents in the single contraction for each core orbital". By the way, a single star after "G" refers to the single polarization function of higher angular momentum added to the 6-31G set. That would be "d" function for Li through Ar, and "f" afterwards. For H and He 6-31G and 6-31G* are identical. A second star refers to the "p" type function on H and He, and no change for the rest of atoms. This means that 6-31G* and 6-31G** for Li through Zn are identical. I am grateful to Jens Spanget-Larsen from Roskilde University for pointing out my mistake. Vitaly Rassolov Department of Chemistry tel. (847) 491-3423 Northwestern University 2145 Sheridan Rd. rassolov@chem.nwu.edu Evanston, IL 60208-3113 From PHogue@space.honeywell.com Tue Dec 9 12:52:27 1997 Received: from az76ms01.az76.honeywell.com for PHogue@space.honeywell.com by www.ccl.net (8.8.3/950822.1) id MAA17964; Tue, 9 Dec 1997 12:10:41 -0500 (EST) Received: by az76ms01.az76.honeywell.com with SMTP (Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63) id <01BD048B.D8CE83E0@az76ms01.az76.honeywell.com>; Tue, 9 Dec 1997 10:18:52 -0700 Message-ID: From: "Hogue, Pat (AZ76)" To: "'CCL'" Subject: Palladium Catalysis Date: Tue, 9 Dec 1997 10:18:50 -0700 X-Mailer: Microsoft Exchange Server Internet Mail Connector Version 4.0.994.63 Encoding: 27 TEXT Greetings CCL-ers: I need a recommendation on software to model, and any experience people may have, for the following problem: An alloy of Pd, Pt, Au and Ag rolls against an alloy of Au, Pt, and Ag in perfluoroisopropylether, also called perfluoropolyalkylether [PFPAE], the fluid is not degassed so it probably contains lots of dissolved oxygen. Minor additions of Cu and Zn are present to cause dispersion strengthening of these alloys. The manufacturer's metallurgist says Cu ands Zn are present as intermetallic (spinodal) phases, so free Cu or free Zn are not available to diffuse throughout the alloy and do nasty things like react with oxygen. As time goes by the rolling electrical contact resistance begins to become somewhat erratic and steadily rises. During over night quite conditions the resistance continues to rise. My first impression is that surface chemistry (probably palladium catalyzed decomposition of PFPAE) may result in the formation of an insulating "friction polymer" e.g., Teflon or some other polymer containing oxygen, fluorine, and carbon. Of course oxidation of surface elements is probably also part of the cause for increased resistance. Are there programs that look at surface catalysis? Any suggestions for software or experience will be appreciated. From fgonzale@lauca.usach.cl Tue Dec 9 14:39:22 1997 Received: from ralun.usach.cl for fgonzale@lauca.usach.cl by www.ccl.net (8.8.3/950822.1) id OAA19130; Tue, 9 Dec 1997 14:22:51 -0500 (EST) Received: from lauca.usach.cl (lauca.usach.cl [158.170.64.28]) by ralun.usach.cl (8.8.5/8.8.8) with ESMTP id QAA02049 for ; Tue, 9 Dec 1997 16:23:17 -0300 (CDT) Received: from danilo (kekule.usach.cl [158.170.16.6]) by lauca.usach.cl (8.6.12/8.6.12) with ESMTP id QAA04885 for ; Tue, 9 Dec 1997 16:29:23 -0300 Message-ID: <348DA8E5.8982F3F8@lauca.usach.cl> Date: Tue, 09 Dec 1997 16:24:07 -0400 From: Danilo Gonzalez Organization: Universidad de Santiago de Chile X-Mailer: Mozilla 4.01 [en] (WinNT; I) MIME-Version: 1.0 To: "chemistry@www.ccl.net" Subject: Summary: Xmol on O2 SGI X-Priority: 3 (Normal) Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Hi!! I received many answer about the following question: *********************************************************************** I trying to run XMOL on a SGI O2, but when I run it, an error message "Wrong Architecture" appears. Can anybody help me to solve this problem? *********************************************************************** In summary, XMOL not run in IRIX 6.X, the solution in only run XMOL in machine with IRIX 5.X, and run via remote display.... ;^( Other possibility is to use other alternative program how VMD ay way, I put it all the answer received.... Thanks again! ***************** Xmol 1.3.1 has been compiled under old IRIX (even before 5.3) in the COFF executable format. Newer IRIX-es (starting with 6.1) no longer support this format; the executable standard is now ELF. You have to go back to IRIX 5.3 to run xmol. To make sure type this command: file xmol Hope this helps, Robert Fraczkiewicz University of Texas Medical Branch ************************* Dear Fdo Danilo Gonzalez Nilo, According to my experience, this is because the difference in the IRIX version. You know SGI O2 runs IRIX6.3 which is very unique. So, try to compile XMOL on SGI O2 again. Good luck. Arthur _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ _/ Arthur Wang Doctoral Candidate _/ _/ Molecular Design Lab _/ _/ Institute of Physical Chemistry, Peking University _/ _/ Beijing 100871, P.R.China _/ _/ _/ _/ E-mail: arthur@ipc.pku.edu.cn _/ _/ Tel: 86-10-62751490 Fax: 86-10-62751725 _/ _/ WWW: http://www.ipc.pku.edu.cn/moldes/arthur/home.html _/ _/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/_/ ********************************** Hi, The short answer is that you can't run Xmol on an O2. The SGI version of Xmol uses the COFF executable format which is not supported by versions of IRIX after 5.3 . Last I heard Xmol was not still being developed so it is unlikely that it will become available for new machines and operating system versions. Dave Young youngd2@mail.auburn.edu ------------------------------------------------------------------------ Live long and maintain marketable job skills. ------------------------------------------------------------------------ *********************** Hi Danilo, the problem we all have is that there has been a big change in the operating system of SGI. The result is that executable programs which had been compiled to the old COFF format are no longer supported since IRIX 6.2 (or 6.0 ?). As you will probably have IRIX 6.4 (or eventually 6.2) you can no longer execute programs in this old COFF format. By the way, you can check your version by entering "/sbin/uname -r" and will probably get an output like "6.2" or "6.4". Unfortunately the XMol crew doesn't want to spread their program's source code (which one could compile for nearly any operating system that is derived >from UNIX). Additionally, they do not recompile their old version of XMol to the new ELF executable format. In order to work around this problem, we use an old SGI workstation running on IRIX 4.0.5 for executing this program and do this with the option "-display=:0.0". This is by now the only way to get at least the graphics output and control panels piped to our new SGI INDIGO2, where we can use this program. Your problem is now that you'd need an SGI machine that runs on an old IRIX... Hope this helps (or at least answers your question), Best Regards, Michael ********************************************** * Dr. Michael Schimeczek * * AK Boche * * FB Chemie der Philipps-Universitaet * * Hans-Meerwein-Str. * * D-35032 Marburg/Lahn * * email: schimi@sg1512.chemie.uni-marburg.de * ********************************************** ************************ My guess is you are using the wrong executable on the O2. Either find an executable for the O2 or recompile XMOL on the O2 itself. -- Jeffrey L. Nauss, PhD Telephone: 513-556-0148 Dir. Molec. Model. Serv. Fax: 513-556-9239 Department of Chemistry e-mail: Jeffrey.Nauss@UC.Edu University of Cincinnati URL http://www.che.uc.edu/~nauss *********************** Hi, usually you get an error message like this if you try to run an MIPS4 (R8K/R10K) executable on a MIPS2 machine (R4K/R5K). In this case you need to recompile XMOL for your machine architecture. Normally a MIPS2 executable should run on a MIPS4 machine. Regards Stefan Disclaimer: All opinions expressed here are mine, not my employers. -- o---------------------------------------------------------------------------o | Dr. Stefan Guessregen Phone: (0049)-(0)-89-451030-0 | | Accelerated Discovery Services FAX: (0049)-(0)-89-451030-30 | | Tripos GmbH WWW: http://www.tripos.com | | Martin-Kollar-Str. 13, D-81829 Muenchen Email: sguess@tripos.com | o---------------------------------------------------------------------------o ***************************** Hola Tambien tenemos una SGI O2 y queremos instalar xmol. De momento xmol lo tenemos funcionando sin ningun problema en estaciones IBM RS600. No se como se resuelve tu problema, pero si lo logras te agradeceria que me lo contases. Un saludo Luis Carballeira Quimica-Fisica Universidad de Vigo, Galicia Espana ********************************* Hi ! As far as I know, xmol only runs with irix 5.3. Everything higher won't work; If there will be a version running on any irix higher than this, please tell me ! Cheers CR _________________________________________________________________ / \ | Christian Rummey | | Institut fuer Organische Chemie | | Universitaet Wuerzburg | | Computational Chemistry Group | | | | Am Hubland | email : rummey@chemie.uni-wuerzburg.de | | 97074 Wuerzburg | voice : +49-(0)931-888-4750 | | Germany | fax : +49-(0)931-888-4755 | | | | http://www-organik.chemie.uni-wuerzburg.de/mitarbeiter/rummey | \_________________________________________________________________/ *********************** There is only one problem: IRIX-6.x does not longer support COFF binaries: $ file /usr/local/bin/xmol /usr/local/bin/xmol: MIPSEB COFF executable (paged) - version 2.40 There are no chances to get a newer version of xmol ( i had a correspondence with the developers of xmol). You have to use other programs. Ciao Bernd Melchers -- Bernd Melchers | melchers@FU-Berlin.DE Freie Universitaet Berlin | "We don't write software, we compose it." AG Macromelecular Modelling - Prof. Dr. E.W. Knapp for more information see http://www.chemie.fu-berlin.de/~melchers ********************** Nobody can help you. IRIX 6.3 (which is the only operating system O2 is capable of running reliably at the moment) does not support COFF executables. The only way to make Xmol run under 6.3 is to recompile it using one of the ELF executable formats 6.3 supports. However, since Xmol is not supported or developed anymore, and the source code is not publicly available, this is not possible. Tough luck... Regards, /Serge.P ********************* I'm afraid there is not much you can do about it. As you know, part of Xmol is distributed as an executable and this has been compiled for different SGI machines, the newest of them being IRIX 5.x As far as I know, the Minnesota team who created Xmol do not plan to upgrade anymore and there we have 1.3.1.1 (if I remember well) as the latest version, which does not work on IRIX 6.x machines, thus including SGI O2. My suggestion would be: install Xmol in an older machine and then run it from your O2 setting the proper DISPLAY. If anyone else comes up with a Xmol version for IRIX 6.x please let me know, I would also be extremelly interested on it (gsastre@itq.upv.es). Cheers. German \|/ (o o) ------------------------------------------oOO-(_)-OOo----------------------- German Sastre URL: http://www.ri.ac.uk/DFRL/G.Sastre Instituto de Tecnologia Quimica e-mail: gsastre@itq.upv.es Universidad Politecnica de Valencia Phone: +34 (9)6-387-7803 Camino de Vera s/n; 46071 Valencia (Spain) Fax: +34 (9)6-387-7809 ---------------------------------------------------------------------------- oo0 0oo ********************** Hi Danilo, I think that happens because you are running IRIX 6.2 or higher. Xmol is not supported for those platforms. I have not heared that anyone was still working on it, so it seems unlikely that it will work in the near future. Cheers, Roelant **************************** Hi, the problem you have is that IRIX 6.x does not support the elf binary format any more. Since XMol is not available in the right format (coeff) you'll have to choose another molecular viewer. You can try VMD (http://www.ks.uiuc.edu/Research/vmd/), that's what I use. Cheers, Dominik -- ------------------------------------------------------------------ Dominik Horinek Institut fuer physikalische und theoretische Chemie Universitaet Regensburg Universitaetsstr. 31, 93053 Regensburg, Deutschland email: Dominik.Horinek@chemie.uni-regensburg.de ------------------------------------------------------------------ ************************** Hello, The reason XMol doesn't work on the O2 is that is was compiled for IRIX 4. IRIX 6, which the O2s use, doesn't support those binaries any more. Since XMol is no longer updated and source isn't available, you will have to use other programs. The one I know the most (because I helped develop it) is VMD at http://www.ks.uiuc.edu/Research/vmd/ . Other good, free ones I've heard of are MOLMOL and gOpenMol, which you should be able to find with a net search program. Andrew Dalke dalke@mag.com ****************** XMOL will not run on an O2. The executable was compiled for an entirely different binary architecture on earlier SGI's. Here, we have O2s, Indigo 2s and and Indigo. The only computer that will run XMOL is the Indigo. As far as I know, there are no plans to update the executable. There has been much discussion in CCL about this; there are alternate packages for many situations. Dan Strahs ********************** -- Fdo. Danilo González Nilo University of Santiago de Chile, Faculty of Chemistry and Biology, Computational Chemistry Lab. Phone: (562) 681 2575-Anexo:799 ; Fax: (562) 681 2108 URL : http://quimbio.usach.cl/~danilo/ E-mail: fgonzale@lauca.usach.cl @@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@@ From nakiya@engin.umich.edu Tue Dec 9 15:39:14 1997 Received: from phoenix.engin.umich.edu for nakiya@engin.umich.edu by www.ccl.net (8.8.3/950822.1) id PAA19437; Tue, 9 Dec 1997 15:22:43 -0500 (EST) Received: from localhost (nakiya@localhost [127.0.0.1]) by phoenix.engin.umich.edu (8.8.7/8.8.7) with SMTP id PAA06946 for ; Tue, 9 Dec 1997 15:22:38 -0500 (EST) Date: Tue, 9 Dec 1997 15:22:38 -0500 (EST) From: Naoko Akiya To: chemistry@www.ccl.net Subject: CCL:G:Summary - open shell calculation In-Reply-To: <348DA8E5.8982F3F8@lauca.usach.cl> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Long, long time ago, I posted the following question: > I have a question regarding the open shell calculation with DFT. I am > trying to calculate the energy of a transition state structure. > Although the structure is designated as a singlet in the input, it is > "partially" open shell. How can I account for this in the calculation > using DFT methods? Below is the summary of all the responses I received. I hope you will find them useful. Sorry for not posting the summary in a timely manner. Naoko ===== The simplest approach to such a situation would be to move to a UB3LYP singlet wavefunction. This does not completely solve the problem but it does allow for alpha and beta densities to differ and thus you can approximate an open-shell singlet at one extreme and a conventional closed shell singlet with alpha = beta at the other. You will need to use GUESS=Mix for such calculations to break the spin symmetry and explore if there is a UHF singlet type solution. Also you would do well to run a STABLE=OPT calculation on the TS structure you find to see if you have a lower lying SCF solution. Should you find one you would need to re-optimize to find that TS structure so you should use a checkpoint file with this calculation to start the optimization with the right wavefunction, Guess=Read. Douglas J. Fox Director of Technical Support help@gaussian.com ===== What do you mean when you say 'partially open shell'? Do you have an unsymmetrical structure that resulted from the Jahn-Teller distortion of a symmetrical diradical? Or do you expect that your transition structure possesses a diradical character? The first thing you may want to do is checking the restricted DFT -> unrestricted DFT stability (e.g. RB3LYP -> UB3LYP) (also called as 'triplet stability') of your solution. If the solution is unstable, the lowest eigenvalue of the stability matrix and the energy lowering for the stable optimized wavefunction can give you an idea how significant the diradical character is. Your further actions depend on these results. Best regards, Mikhail Gloukhovtsev If the RHF or R-DFT solution is triplet stable (i.e. RHF->UHF stable), using the UHF wavefunction should give the same energy as that you calculated with the RHF one. However, the triplet instability threshold can take place later when the bond breaking becomes more profound. You find a comprehensive discussion of this problem in the paper of Schlegel and Chen: First Author: CHEN-W. Other Authors: SCHLEGEL-HB*. Article Title: EVALUATION OF S-2 FOR CORRELATED WAVE-FUNCTIONS AND SPIN PROJECTION OF UNRESTRICTED MOLLER-PLESSET PERTURBATION-THEORY Journal Source: JOURNAL OF CHEMICAL PHYSICS, Volume: 101, Issue: 7, Issue Date: OCT 1, Publication Year: 1994, Pages: 5957-5968 Document Type: 3: ARTICLE. Abstract Ind.: Y Subject Cat.: UH: PHYSICS, ATOMIC, MOLECULAR & CHEMICAL. Hope this helps. Best regards, Mikhail ************************************************************************** * * * Mikhail Gloukhovtsev, Ph. D., Dr. Sci. * * Brown Laboratory, Department of Chemistry and Biochemistry * * University of Delaware, Newark, DE 19716 * * Ph. 302-831-0887 Fax (302)-831-6335 * * E-mail mng@udel.edu WWW http://planck.duch.udel.edu/~mikhail * * http://udel.edu/~mng * * * ************************************************************************** ===== It is not that simple to explan the open shell matter in a few words. Usually, when a system has any unpaired electron(s), one should use unrestricted (open shell) method. It is incorrect to work with closed shell and open shell combination. One should stick on one method throughout whole calculation. ************************************************************************ * Because the machine, the terminator, ________ ________ * * can learn the value of human life, | \ / | * * maybe we can, too. |_ \ / _| * * ------------------------------------------ __|________\/________|__ * * Ning Chen | |* * Department of Chemical Engineering | M I C H I G A N |* * University of Michigan, Ann Arbor, MI48109 |________________________|* * Phone: (313)-764-8362 (o) _| |_\ /_| |_ * * (313)-936-3186 (lab) | |\ /| | * * e-amil: ningchen@engin.umich.edu |________| \/ |________| * * URL: www-personal.engin.umich.edu/~ningchen * ************************************************************************ ===== There was recently a discussion about singlet diradicals ( open shell, i.e. e.g. 1s(1)2s(1) for helium or 1_Sigma_gerade state of molecular oxygene etc). In principle you can't do the calculation within a single determinant ansatz (in Kohn-Sham DFT you use also a single determinant of Kohn-Sham orbitals to represent the "wave function"). But you can use a trick : you calculate the energy for the triplet (t) and for the mixed state (m). The energy of the singlet state is then E(s)=2*E(m)-E(t). But this works only for energies, not for gradients. Literatur : Tom Ziegler, Arvi Rauk, Evert J. Baerends, Theoret. Chim. Acta 43 (1977) 261-271. Another problem: I was not able to do the calculation for the mixed state for molecules more complicated than He. I got very (!!!!) sincere convergence problems and wild oscillations during the course of calculation. Ciao, Heinz -- Dr. Heinz Schiffer Phone ++49-69-305-2330 Hoechst Research & Technology Fax ++49-69-305-81162 Scientific Computing, G864 Email schiffer@h1tw0036.hoechst.com 65926 Frankfurt am Main Schiffer@CRT.hoechst.com =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Naoko Akiya nakiya@engin.umich.edu University of Michigan Dept. of Chemical Engineering phone (313) 764-7121 Ann Arbor, MI 48109-2136 fax (313) 763-0459 =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=