From chemistry-request@www.ccl.net Fri Oct 30 01:22:00 1998 Received: from hartree.chem.kuleuven.ac.be (hartree.quantchem.kuleuven.ac.be [134.58.49.1]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id BAA04234 Fri, 30 Oct 1998 01:21:58 -0500 (EST) Received: from localhost (steven@localhost) by hartree.chem.kuleuven.ac.be (8.9.0/8.9.0) with ESMTP id IAA09902 for ; Fri, 30 Oct 1998 08:23:08 +0100 Date: Fri, 30 Oct 1998 08:23:07 +0100 (NFT) From: Steven Creve To: Computational Chemistry List Subject: CCL:Re:CC versus CI In-Reply-To: <199810291659.LAA29281@chemistry.mps.ohio-state.edu> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 29 Oct 1998, David Heisterberg wrote: > >Suppose we perform a CI calculation with all determinants that are > >normally generated by CCSD (exp(T1+T2)). Would the obtained result be > >worse or better than the CCSD value? > > Doesn't exp(T1) in fact generate all possible excitations? The CI > case would thus be full CI and therefore exact. I think the coupled > cluster wave operator, as long as it includes singles, must generate > all possible excitations, but most of them have their coefficients > determined as functions of the coefficients of lower order excitations. > > Dave Heisterberg > Truncated CCSD does *not* generate all excitations. By making use of the hausdorff comutator expansion, one can show that the exp-series terminates exactly after a certain number of orders. steven -------------------------------------------------------------------------- Steven Creve steven.creve@chem.kuleuven.ac.be Labo Quantumchemie Celestijnenlaan 200F 3001-HEVERLEE tel: (32) (16) 32 73 93 BELGIUM fax: (32) (16) 32 79 92 From chemistry-request@www.ccl.net Fri Oct 30 02:08:37 1998 Received: from hera.wku.edu (root@hera.wku.edu [161.6.20.1]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id CAA04562 Fri, 30 Oct 1998 02:08:36 -0500 (EST) Received: from adler by hera.wku.edu with local (Exim 1.82 #1) id 0zZ8fl-00046z-00; Fri, 30 Oct 1998 01:08:37 -0600 From: Allen Adler To: chemistry@www.ccl.net, adler@hera.wku.edu Subject: Re: CCL:CC versus CI Message-Id: Date: Fri, 30 Oct 1998 01:08:37 -0600 Let me apologize in advance. I'm walking into this discussion in the middle and understand neither the terminology, the acronymns nor the context. However, I registered some recognition upon reading Steven Creve's remark: >Truncated CCSD does *not* generate all excitations. By making use of the >hausdorff comutator expansion, one can show that the exp-series terminates >exactly after a certain number of orders. in regard to the expression CCSD (exp(T1+T2)). Being a mathematician, I usually run across the Campbell-Baker-Hausdorff formula in connection with Lie groups and Lie algebras. So now I'm trying to guess whether (1) Steven Creve is saying, in effect, that T1 and T2 are operators that generate a Lie algebra which is nilpotent. (2) David Heisterberg was, in effect, guessing that the algebra was either abelian or 2-step nilpotent. Anyway, now I'm curious to know what it actually is that is being discussed, on the off chance that I might understand it. Specifically, (a) what are T1,T2? (b) what is the definition of the expression CCSD (exp(T1+T2))? (c) where is the result on the nilpotence of the Lie algebra proved? (d) what do all the acronyms stand for? (e) in what readily accessible work is all this explained? I may be able to get my hands on a copy of the 1989 edition of Ostland and Sczabo. Is it there? Naively, Allan Adler adler@hera.wku.edu From chemistry-request@www.ccl.net Fri Oct 30 05:12:49 1998 Received: from mserv.itpa.lt (mserv.itpa.lt [193.219.53.20]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id FAA05215; Fri, 30 Oct 1998 05:12:38 -0500 (EST) Received: from localhost (tamulis@localhost) by mserv.itpa.lt (8.9.1/8.9.1) with SMTP id LAA12101; Fri, 30 Oct 1998 11:09:00 +0100 (CET) Date: Fri, 30 Oct 1998 11:09:00 +0100 (CET) From: Arvydas Tamulis To: gwaltney@qtp.ufl.edu, cramer@maroon.tc.umn.edu, ck@ws2.theochem.uni-hannover.de, husuter@cscs.ch, frj@dou.dk, Matthew.Harbowy@tjlus.sprint.com, acp37@rs1.rrz.uni-koeln.de, CHEMISTRY@www.ccl.net, CHEMISTRY-REQUEST@www.ccl.net, MAILSERV@www.ccl.net Subject: Fluorescence emission spectra calculations In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Netters, I met problem how to calculate the fluorescence emission spectra for organic molecules in my interesting Light Emitting Diodes (LED) field. The absorption spectra I more or less well calculating by CNDO/S, ab initio CIS, ZINDO/S-CI programes (the last is the best) but how to calculate the emission spectra of as example Coumarine 515 molecule stacked with Carbazole molecule in the LED layers? I have the good agreement of absorption - experimental and my calculated spectra of carbazole molecule. After the absorption carbazole molecule gives electron to coumarine molecule and this molecule emitt light. How to simulate quantum chemically the fluorescence emission spectra in such a organic electron acceptor molecules, which methods are most suitable, which programs are the best? Thanking your in advance. With best regards. Yours sincerely, Arvydas Tamulis TEMPORARY ADDRESSES UNTIL 07th of November, 1998: Work: dr. Arvydas TAMULIS c\o Prof. Jean-Michel NUNZI Groupe Composants Organiques DTA-LETI-DEIN, Service de Physique Electronique CEA Saclay, F91191 Gif sur Yvette, FRANCE Phone: 0033-169088543 e-mail: TAMULIS@ITPA.lt Home: 52, rue de Madrid, F91190 Gif-sur-Yvette, FRANCE ----------------------------------------------- Arvydas Tamulis Doctor of Natural Sciences, senior research fellow PERMANENT ADDRESSES AFTER 11th OF NOVEMBER, 1998: Institute of Theoretical Physics and Astronomy, Theoretical Molecular Electronics Research Group, A. Gostauto 12, Vilnius 2600, Lithuania e-mail: TAMULIS@ITPA.lt; WEBsite: http://www.itpa.lt/~tamulis/ fax: +(370-2)-225361 or +(370-2)-224694 Phone: +(370-2)-620861 Home address: Didlaukio 27-40, Vilnius 2057, Lithuania Phone: +(370-2)-778743 From chemistry-request@www.ccl.net Fri Oct 30 06:07:33 1998 Received: from stinch10.csmtbo.mi.cnr.it (stinch10.csmtbo.mi.cnr.it [155.253.8.10]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id GAA05423 Fri, 30 Oct 1998 06:07:31 -0500 (EST) Received: (from davide@localhost) by stinch10.csmtbo.mi.cnr.it (8.8.8/8.8.8) id LAA04390 for CHEMISTRY@www.ccl.net; Fri, 30 Oct 1998 11:57:41 +0100 (MET) Date: Fri, 30 Oct 1998 11:57:41 +0100 (MET) From: Davide Proserpio Message-Id: <199810301057.LAA04390@stinch10.csmtbo.mi.cnr.it> To: CHEMISTRY@www.ccl.net Subject: ISI and citation After Le Bail message concerning the ISI's most cited chemists , I would like to suggest to read also a recent paper published on New J. Chemistry concerning : Sense and Nonsense of science citation analyses : comments on the monopoly position of ISI and citation inaccuracies. Risk of possible misuses and biased citation and impact data by Jan Reedijk N.J. Chem. 1998 pag 767-770 Note: I just read the paper, I have no connection with the author. At the present NJC is freely downladable from the internet (for all 1998) http://www.rsc.org/is/journals/current/newjchem/njccon.htm Davide ------------------------------------------------------------------- Dr. Davide M. Proserpio Dipartimento di Chimica Strutturale e Stereochimica Inorganica Universita' di Milano, Via Venezian, 21 - 20133 Milano, Italy phone +39-2-70635120 fax 70635288 - davide@csmtbo.mi.cnr.it ------------------------------------------------------------------- From chemistry-request@www.ccl.net Fri Oct 30 07:14:39 1998 Received: from ssmain.uniss.it (ssmain.uniss.it [193.205.8.2]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id HAA05644 Fri, 30 Oct 1998 07:14:33 -0500 (EST) Received: from frodo.agraria.uniss.it (frodo.agraria.uniss.it [193.205.10.80]) by ssmain.uniss.it with SMTP (8.8.8/8.8.8) id NAA26005 for ; Fri, 30 Oct 1998 13:12:55 +0100 (MET) Message-ID: <3639ADA6.4632@antas.agraria.uniss.it> Date: Fri, 30 Oct 1998 13:14:30 +0100 From: bruno manunza X-Mailer: Mozilla 3.01Gold (Win95; I) MIME-Version: 1.0 To: "'chemistry@www.ccl.net'" Subject: all image convention References: <01BE01BC.2E9D6920.johnw@msi.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Can anybody give some explanation about the 'all image convention'? It looks like some technique which should be used when the cut-off radius is larger tha the half cell edge. Is it? Thanks Bruno -- Dr Bruno Manunza DISAABA Univ. of Sassari V.le Italia 39 07100 Sassari ITALY phone 39 79 229215 fax 39 79 229276 e_mail bruno@antas.agraria.uniss.it http://antas.agraria.uniss.it From chemistry-request@www.ccl.net Fri Oct 30 09:02:04 1998 Received: from chemistry.mps.ohio-state.edu (chemistry.mps.ohio-state.edu [128.146.33.22]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id JAA06441 Fri, 30 Oct 1998 09:02:03 -0500 (EST) Received: (from dheister@localhost) by chemistry.mps.ohio-state.edu (8.8.8/8.8.5) id JAA26118; Fri, 30 Oct 1998 09:01:46 -0500 (EST) Date: Fri, 30 Oct 1998 09:01:46 -0500 (EST) From: David Heisterberg Message-Id: <199810301401.JAA26118@chemistry.mps.ohio-state.edu> To: Steven.Creve@chem.kuleuven.ac.be, chemistry@www.ccl.net Subject: Re: CCL:CC versus CI >Truncated CCSD does *not* generate all excitations. By making use of the >hausdorff comutator expansion, one can show that the exp-series terminates >exactly after a certain number of orders. I think we're talking about two different things. If |F> is the reference state, I'm thinking that exp (T1+T2) |F> implicitly generates all possible excitations. But exp (-T1-T2) H exp (T1+T2) terminates so only a subset of excited configuration coefficients are independently determined. In the energy expression though, aren't the effects of the higher-excited configurations included, at least implicitly, with coefficients that are functions of those that were determined by the CC equations? Dave Heisterberg From jkl@ccl.net Thu Oct 29 18:43:09 1998 Return-Path: Received: from leona.harvard.edu (daizadeh@leona.harvard.edu [128.103.96.68]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id SAA20003 for ; Thu, 29 Oct 1998 18:43:08 -0500 (EST) Received: from localhost (daizadeh@localhost) by leona.harvard.edu (8.8.7/8.8.7) with SMTP id SAA31218 for ; Thu, 29 Oct 1998 18:46:09 -0500 Date: Thu, 29 Oct 1998 18:46:09 -0500 (EST) From: Iraj Daizadeh X-Sender: daizadeh@leona.harvard.edu To: jkl@ccl.net Subject: Summary: BioInformatics. (fwd) Message-ID: Hello. I did not get a wide response concerning Bioinformatics software. I am slightly disappointed, but I believe it to be understandable. My hypothesis is this: Computational Chemistry has recently split into the following three sub-areas: Computational Chemistry / | \ Traditional Biogenomics Combinatorial Bioinformatics By Traditional, I mean: quantum chemistry, dynamics, etc... By Biogenomics/Bioinformatics: I mean: genome/protein database analysis By Combinatorial, I mean: molecular diversity, molecular similarity, etc.. There are of course ``interdisciplinary'' fields as well: quantum molecular similarity, etc... I have been ``brought'' up in the left-most branch of the above tree, viz. ``traditional'' computational chemistry. My dissertation, entitled ``Theoretical Study of Electron Transfer Reactions in Biological and Chemical Systems'', examined long-distance electron transfer reactions using semi-empirical, ab initio, and molecular dynamics techniques (and their combinations). My post-doctoral work is in Combinatorial Chemistry, my interest lies more in BioGenomics, however. The question I submitted to the CCL was meant to introduce me (and through this summary hopefully others) to the software resources that are available in the biogenomics area of research. >From the few responses to my query, I see that maybe the Computational Chemistry Communittee (or at least members of this list) is focused on the more traditional areas of research. Yet, when I explore www.altavista.com with the keyword bioinformatics, I find many, many hits that correspond to software packages (free or commercial). [There is a chance that few people actually took the time to read my email!] The importance of this field has been quickly realized. Journals such as Science and Nature are loaded with employment opportunities for Bioinformatics experts. Indeed, there is a strong trend for those in industry to invest in Bioinformatics. A recent article was posted in C&EN: (Oct. 19, 1998 pp.29-33 ``Bioinformatics begins to buzz'' Ronald S. Rogers); and also in the notebook section of The Scientist ( ``...But don't forget about bioinformatics'' September 28,1998 pp 31) attribute to this phenomena. The question is why are the molecular biologists and the computer scientists having all the fun! Why has the Computational Chemistry Communittee been so slow to respond to these new branches of study when they deal in a most direct way to what is happening on a molecular level. I feel as though our contributions to this field may be enormous! Your thoughts on any of the above statements would be acknowledged. Sincerely, Iraj. XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX The outline of this summary is as follows: I. Main query. II. Responses from the communittee. Sincerely, Iraj. I) XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX Hello. I am interested in obtaining codes (written in any language) that are free or relatively free (< $500.00) for members in academia that can visualize and manipulate Genetic Database Information (nucleic acid sequences, etc...) for problems in BioInformatics. I will summarize all respones as is traditional on the CCL. Sincerely, Iraj. Iraj Daizadeh, Ph.D. Harvard University Department of Chemistry and Chemical Biology 12 Oxford Street, Box #35 Cambridge, MA 02138-2902 Phone: 617-495-2654 Fax: 617-495-1823 email: daizadeh@lhasa.harvard.edu II) XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX Hi, Have a look at AntheProt and AntheNuc: ftp://ibcp.fr/pub/ There are also lists of PD software at http://mbisg2.sbc.man.ac.uk/soft4.html http://www.ebi.ac.uk/biocat/ - there's a lot of it out there; I'll be very interested to read your summary. Good luck, Nick ***************************************************************** Nick Rhodes, Chemical Information Systems Research Group, Department of Information Studies, University of Sheffield, Regent Court, 211 Portobello Street, Sheffield, South Yorkshire, S1 4DP, United Kingdom. XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX You may want to try CuraTools, our sequence analysis package which is freely available for use on the World Wide Web. Visit "http://www.curagen.com" and choose "CuraTools". Andreas Windemuth CuraGen Corporation 555 Long Wharf Drive New Haven, CT 06511 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX Hello, Iraj! One of my colleagues forwarded me your question on CCL. You may be interested in the Silicon Graphics MineSet product, which has been used by Incyte Pharmaceuticals to visualize genomic databases. You can see, for example, our "MineSet for Discovery Research Report" for various examples of how MineSet has been used in bioinformatics and cheminformatics: http://www.sgi.com/chembio/resources/mineset/index.html The specific example of Incyte Pharmaceuticals is here: http://www.sgi.com/chembio/resources/mineset/bioinfo/genomic_database/index.html I am afraid though that the price of MineSet is beyond what you specified. You can, however, obtain a 30 day free evaluation license for MineSet: http://www.sgi.com/Products/Evaluation/#mineset May I ask what you are interested in being able to do? Alan Robinson (alan@ebi.ac.uk) at the European Bioinformatics Insitute may be able to give you some pointers. -- Haruna :) -- XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX From chemistry-request@www.ccl.net Fri Oct 30 10:33:57 1998 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id KAA09748 Fri, 30 Oct 1998 10:33:57 -0500 (EST) Received: from main.amu.edu.pl (root@main.amu.edu.pl [150.254.65.2]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id KAA04679 for ; Fri, 30 Oct 1998 10:33:29 -0500 (EST) Received: from main.amu.edu.pl (jolanala@main.amu.edu.pl [150.254.65.2]) by main.amu.edu.pl (8.8.8/8.8.8) with SMTP id QAA14962 for ; Fri, 30 Oct 1998 16:33:07 +0100 (MET) Date: Fri, 30 Oct 1998 16:33:07 +0100 (MET) From: Jolanta Latosinska To: chemistry@ccl.net Subject: Re: CCL:HYPERFINE vs G TENSOR CALCULATION CONFUSION (fwd) In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear all, A few days ago there was a very helpfull tip from Prof. Omaley concerning hyperfine vs G tensor calculations. I tried to calculate hyperfine tensor using IOp keyword, but I obtained only isotropic part (Fermi contact Analysis). Does anybody know how to calculate anisotropic part of this tensor? Thanks in advance, Jolanta From chemistry-request@www.ccl.net Fri Oct 30 10:45:28 1998 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id KAA11022 Fri, 30 Oct 1998 10:45:28 -0500 (EST) Received: from Amnesix.UQSS.UQuebec.ca (Amnesix.UQSS.UQuebec.CA [192.77.51.5]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with ESMTP id KAA05112 for ; Fri, 30 Oct 1998 10:45:27 -0500 (EST) From: jianhui_wu@IAF.UQUEBEC.CA Received: from poste.iaf.uquebec.ca by Amnesix.UQSS.UQuebec.ca with SMTP (1.37.109.24/15.6) id AA236702326; Fri, 30 Oct 1998 10:45:26 -0500 Received: from ccMail by poste.iaf.uquebec.ca (ccMail Link to SMTP R8.10.00) id AA909762450; Fri, 30 Oct 98 10:47:34 -0500 Message-Id: <9810309097.AA909762450@poste.iaf.uquebec.ca> X-Mailer: ccMail Link to SMTP R8.10.00 Date: Fri, 30 Oct 98 10:51:05 -0500 To: Subject: Discover, water evaporate Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Content-Description: "cc:Mail Note Part" Dear Colleagues, I do some MD on small molecule solvated with water box using Discover. By the end of calculations, it seems that quite a few water molecules `evaporate` outside the box. I wonder whether this will affect the result. Presumably, the images of the `evaporate waters` are still inside the box. Therefore, it does not matter. Is that right? If it does matter, how to avoid it? (By the way, I used cell-based or group-based method, Rcut < 1/2 box size). Any help will be appreciated. (Please reply to my address. I will summarize the results) Regards, JianHui Wu Institute Armond-Frappier Laval, Canada email: JianHui_Wu@IAF.Uquebec.CA From chemistry-request@www.ccl.net Thu Oct 29 03:24:45 1998 Received: from dndhost.ruc.dk (dndhost.ruc.dk [130.225.220.22]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id DAA00962 Thu, 29 Oct 1998 03:24:44 -0500 (EST) Received: from virgil.ruc.dk (virgil.ruc.dk [130.225.220.110]) by dndhost.ruc.dk (8.8.8+DND/8.8.8) with ESMTP id JAA06494; Thu, 29 Oct 1998 09:24:40 +0100 (MET) Received: from VIRGIL/SpoolDir by virgil.ruc.dk (Mercury 1.40); 29 Oct 98 09:25:01 +0100 Received: from SpoolDir by VIRGIL (Mercury 1.40); 29 Oct 98 09:24:52 +0100 From: "Jens Spanget-Larsen" Organization: Roskilde Universitetscenter To: "Eun Joo Lee" Date: Thu, 29 Oct 1998 09:24:36 +0100 Subject: CCL:G:reduced mass CC: chemistry@www.ccl.net Priority: normal X-mailer: Pegasus Mail for Windows (v2.23) Message-ID: <3ED2C422286@virgil.ruc.dk> Eun Joo Lee: > I would like to know how the gaussian program calculate the > reduced mass when you submit the job for frequency calculation of molecule. > I wonder how 3n-6 # of reduced mass is calculated instead of 3n # of reduced > mass.(n is number of atoms) > It could be very simple question, but still it puzzles me. The normal vibrational modes are defined by a set of orthonormal vectors _ _ 1/2 Q = >_ c q = >_ c m x r i ir i i ir i i where q are the massweighted atomic displacement coordinates, m are the atomic masses and x are the cartesian displabcement coordinates. The reduced mass for normal mode r is defined by _ 2 -1 M = 1/( >_ c m ) r i ir i The inverse square root of M is simply the length of the cartesian displacement vector X for unit displacement along normal coordinate Q: -1/2 |X | = |Q | M r r r The cartesian displacement vectors printed by Gaussain are normalized to unit length = 1 angstrom. The proper length is obtained by multiplication of the printed coordinates by the inverse square root of the reduced mass. Yours, Jens >--< =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= JENS SPANGET-LARSEN Phone: +45 4674 2000 (RUC) Department of Chemistry +45 4674 2710 (direct) Roskilde University (RUC) Fax: +45 4674 3011 P.O.Box 260 E-Mail: JSL@virgil.ruc.dk DK-4000 Roskilde, Denmark http://www.rub.ruc.dk/dis/chem/psos/ =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= From chemistry-request@www.ccl.net Fri Oct 30 06:46:54 1998 Received: from congo.cherwell.com (congo.cherwell.com [193.82.249.202]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id GAA05555 Fri, 30 Oct 1998 06:46:53 -0500 (EST) Received: from wharfe.cherwell.com ([193.82.249.214]) by congo.cherwell.com (Post.Office MTA v3.1.2 release (PO205-101c) ID# 0-0U10L2S100V35) with SMTP id AAB464 for ; Fri, 30 Oct 1998 11:46:17 +0000 Message-Id: <3.0.5.32.19981030114558.008037f0@cherwell.com> X-Sender: adam@cherwell.com X-Mailer: QUALCOMM Windows Eudora Light Version 3.0.5 (32) Date: Fri, 30 Oct 1998 11:45:58 +0000 To: chemistry@www.ccl.net From: Adam Hodgkin Subject: Release of ChemSymphony Beans 1.1 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Dear InternetChemist Cherwell Scientific is pleased to announce today the release of ChemSymphony Beans 1.1. For full details see: http://www.chemsymphony.com/ The new release has many important features which broaden the range of the toolkit. In particular we now have a 3D sketching tool, we have a simplified Marquee component, we now offer comprehensive JavaDocs on the public classes and the www site is deeper, broader and better indexed. We are simultaneously releasing 5 ChemSymphony JavaBeans under a Lite licence. We invite you to download these FREE beans if you are curious to find out what these software components can do for Chemistry. Adam --- --- Adam Hodgkin | e-mail: adam@cherwell.com Chairman | Phone: +44 (0)1865 784810 Cherwell Scientific Publishing | Fax: +44 (0)1865 784801 Oxford OX4 4GA, UK | URL: http://www.cherwell.com --- --- From chemistry-request@www.ccl.net Fri Oct 30 10:53:52 1998 Received: from cliff.uottawa.ca (root@cliff.uottawa.ca [137.122.6.65]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id KAA12063 Fri, 30 Oct 1998 10:53:49 -0500 (EST) Received: from oreo.science.uottawa.ca (oreo.science.uottawa.ca [137.122.43.15]) by cliff.uottawa.ca (8.8.8/8.8.7) with ESMTP id KAA83150 for ; Fri, 30 Oct 1998 10:53:03 -0500 Received: from OREO/SpoolDir by oreo.science.uottawa.ca (Mercury 1.43); 30 Oct 98 10:53:50 GMT-5 Received: from SpoolDir by OREO (Mercury 1.43); 30 Oct 98 10:53:45 GMT-5 Received: from erik (137.122.43.165) by oreo.science.uottawa.ca (Mercury 1.43); 30 Oct 98 10:53:39 GMT-5 Sender: boufer@uottawa.ca Message-ID: <3639DCC0.41C6@cennas.nhmfl.gov> Date: Fri, 30 Oct 1998 10:35:28 -0500 From: boufer Organization: None X-Mailer: Mozilla 2.02 (X11; I; AIX 1) MIME-Version: 1.0 To: "Comp. Chem. List" Subject: Frequency analysis in G-9n Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Greetings all, Can anybody explain why in the results of Gaussian, the values labeled as : Low Frequencies -- are in general different from those listed below in the part corresponding to the frequency analysis. Thanks for your time and help. -- Ahmed From chemistry-request@www.ccl.net Fri Oct 30 11:01:39 1998 Received: from octane.kemi.aau.dk (octane.kemi.aau.dk [130.225.22.149]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id LAA12946 Fri, 30 Oct 1998 11:01:38 -0500 (EST) Received: from octane.kemi.aau.dk (localhost.kemi.aau.dk [127.0.0.1]) by octane.kemi.aau.dk (950413.SGI.8.6.12/950213.SGI.AUTOCF) via SMTP id RAA17178; Fri, 30 Oct 1998 17:01:37 -0800 Sender: roberson@kemi.aau.dk Message-ID: <363A6170.41C6@kemi.aau.dk> Date: Fri, 30 Oct 1998 17:01:36 -0800 From: MJR Organization: Center for Metal Catalysed Reactions X-Mailer: Mozilla 3.01SGoldC-SGI (X11; I; IRIX64 6.4 IP30) MIME-Version: 1.0 To: David Heisterberg CC: CHEMISTRY@www.ccl.net Subject: Re: CCL:CC versus CI References: <199810301401.JAA26118@chemistry.mps.ohio-state.edu> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit David Heisterberg wrote: > > >Truncated CCSD does *not* generate all excitations. By making use of the > >hausdorff comutator expansion, one can show that the exp-series terminates > >exactly after a certain number of orders. > > I think we're talking about two different things. If |F> is the reference > state, I'm thinking that exp (T1+T2) |F> implicitly generates all possible > excitations. But exp (-T1-T2) H exp (T1+T2) terminates so only a subset of > excited configuration coefficients are independently determined. In the > energy expression though, aren't the effects of the higher-excited configurations > included, at least implicitly, with coefficients that are functions of those > that were determined by the CC equations? > > Dave Heisterberg It looks like we are getting lost in the terminological thicket. CCSD generates only single and double excitations, which are fed into the calculation via the cluster operator whose expansion contains terms like T1T1T2T1T2; throwing the anticommutation machinery of second quantisation allows the energy expression to be rewritten containing only T1 and T2, so that the contributions from the higher excitations are approximately included. This works in practice because, for the kind of systems chemists are interested in, Cijkl is to a good approximation a linear combination of Cij * Ckl and the other symbols generated by permuting the labels. Szabo and Ostlund would be the reference to look in for this stuff, in all it's gory detail. Mark ********************************************************************** To reply, replace the danish robersen with the welsh roberson. ********************************************************************** From chemistry-request@www.ccl.net Fri Oct 30 12:39:14 1998 Received: from ccl.net (atlantis.ccl.net [192.148.249.4]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id MAA23411 Fri, 30 Oct 1998 12:39:13 -0500 (EST) Received: from bioc1.msi.com (bioc1.msi.com [146.202.0.2]) by ccl.net (8.8.6/8.8.6/OSC 1.1) with SMTP id MAA08461 for ; Fri, 30 Oct 1998 12:39:09 -0500 (EST) Received: by bioc1.msi.com (931110.SGI/930416.SGI) for chemistry@ccl.net id AA17634; Fri, 30 Oct 98 09:35:15 -0800 Received: from dxg-pc.msi.com(146.202.11.19) by bioc1.msi.com via smap (V2.0) id xma017626; Fri, 30 Oct 98 09:35:11 -0800 Message-Id: <3.0.32.19981030093216.0070d194@146.202.0.225> X-Sender: dgregory@146.202.0.225 X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Fri, 30 Oct 1998 09:32:17 -0800 To: jianhui_wu@IAF.UQUEBEC.CA From: Don Gregory Subject: Re: CCL:Discover, water evaporate Cc: chemistry@ccl.net (Comp. Chem. List) Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Hi Dr. Wu, You can find some helpful information on MSI's scientific support, FAQ web-site. First go to: http://www.msi.com/support/FAQ.html and select the InsightII Core area. There you will find a tutorial on creating your own solvent boxes, and within the "How to Equilibrate the box" portion of that tutorial, you'll find a discussion on how to treat the 'visual' effect of waters/images outside the box. Hope this helps! Don Gregory At 10:51 AM 10/30/98 -0500, you wrote: > >Dear Colleagues, > >I do some MD on small molecule solvated with water box using Discover. >By the end of calculations, it seems that quite a few water molecules >`evaporate` outside the box. >I wonder whether this will affect the result. Presumably, the images of the >`evaporate waters` are still >inside the box. Therefore, it does not matter. Is that right? If it does >matter, how to avoid it? >(By the way, I used cell-based or group-based method, Rcut < 1/2 box size). > >Any help will be appreciated. (Please reply to my address. I will summarize the >results) > >Regards, > >JianHui Wu >Institute Armond-Frappier >Laval, Canada >email: JianHui_Wu@IAF.Uquebec.CA > > > >-------This is added Automatically by the Software-------- >-- Original Sender Envelope Address: chemistry-request@www.ccl.net >-- Original Sender From: Address: jianhui_wu@IAF.UQUEBEC.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 > > > > Dr. Don Gregory (dgregory@msi.com) Japan Science Liaison Molecular Simulations Inc. 9685 Scranton Rd. San Diego, CA 92121 (619) 546-5331 http://www.msi.com From chemistry-request@www.ccl.net Fri Oct 30 17:19:07 1998 Received: from london.tcg.anl.gov (london.tcg.anl.gov [146.137.200.37]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id RAA11125 Fri, 30 Oct 1998 17:19:07 -0500 (EST) Received: (from shepard@localhost) by london.tcg.anl.gov (8.6.11/8.6.11) id RAA19948; Fri, 30 Oct 1998 17:19:07 -0500 Date: Fri, 30 Oct 1998 17:19:07 -0500 From: Ron Shepard Message-Id: <199810302219.RAA19948@london.tcg.anl.gov> To: Steven.Creve@chem.kuleuven.ac.be, robersen@kemi.aau.dk Subject: Re: CCL:CC versus CI Cc: CHEMISTRY@www.ccl.net >> Suppose we perform a CI calculation with all determinants that are >> normally generated by CCSD (exp(T1+T2)). Would the obtained result be >> worse or better than the CCSD value? >> >> Any comments are welcome? This problem is not phrased exactly correctly. The set of determinants generated by exp(T1) itself is the full set of determinants, so if a CI calculation were done in this full space the answer would be the full-CI energy, which is the best that can be done in that finite basis. Adding the T2 operator is redundant in this context, since the determinants that it generates are also generated by the T1^2 terms in the exponential. > You ask interesting questions, Steven, and this one I think I can >answer. > > Both CISD and CCSD represent the correlated wavefunction |corr> as a >sum over configuration state functions. The difference between them is >that CI exploits the variational prinicple to minimise the energy, thus > > CI: = > >ensuring a strict upper bound to the total energy, while CC solves the >Schroedinger equation within a subspace. As a consequence, the computed > > CC: H|corr> = E|corr> > >energy may be greater than or less than the true energy. Use of the >cluster operator to parameterise |corr> results in the implicit >estimation of contributions from higher excitations ( quadruples, etc ) >so the energy in practice tends to be quite accurate. This is correct, but there are other important differences in the CI and CC energy expressions too. It is not that CC represents the wave function as an expansion in CSFs, it is that the CC represents the wave function as a nonlinear contraction of CSFs. So another important difference between CC and CI is that in CI the individual CSFs each have their own coefficient, but in CC each individual CSF has a coefficient that is constructed, in a nonlinear way, as sums of products of the T1 and T2 coefficients. Consider for example the contribution of a particular 4-fold excitation CSF in the two cases. In CI, this CSF would have its own coefficient, which would be optimized variationally. But in CC, this coefficient would be determined as a sum of T2*T2+T2*T1*T1+T1*T1*T1*T1 terms. In the CI case there would be a lot of independent parameters, which means an expensive calculation, while in the CC case the number of parameters is limited, which means a relatively cheap calculation. Then, there is the philisophical question of exactly what is in the CC wave function. If you think of the CC wave function as exp(T), then you might think that there are all possible excitations. But if you look at the way the parameters T1 and T2 are computed, then you will see that there are only 4-fold excitations included. Since the energy is computed from only T1 and T2, and since they are determined only from the 4-fold excitations, you might also say that the energy itself depends only on 4-fold excitations. Sound like a paradox? I believe that you just have to meditate on this for a while before you can make peace with it. :-) $.02 -Ron Shepard From chemistry-request@www.ccl.net Fri Oct 30 18:07:25 1998 Received: from postal.ucc.uno.edu (postal.ucc.uno.edu [137.30.1.80]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id SAA11457 Fri, 30 Oct 1998 18:07:25 -0500 (EST) Received: from slater (slater.chem.uno.edu [137.30.57.87]) by postal.ucc.uno.edu (PMDF V5.2-29 #33240) with SMTP id <0F1N008EUXEDUB@postal.ucc.uno.edu> for chemistry@www.ccl.net; Fri, 30 Oct 1998 17:03:49 -0600 (CST) Date: Fri, 30 Oct 1998 17:03:18 -0600 From: Howard Alper Subject: Parameterization Sender: heacm@slater.chem.uno.edu To: chemistry@www.ccl.net Cc: halper@uno.edu Message-id: <363A45B6.41C6@uno.edu> Organization: University of New Orleans MIME-version: 1.0 X-Mailer: Mozilla 3.0 (X11; U; AIX 1) Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7bit Hi, Several weeks ago I asked some questions about parameterization. I received several replies, and have been happily (well, sort of) plugging away at parameterizing Nitromethane. I have a further question. I am interested in using Quantum mechanical calculations to determine some of the intramolecular force constants, and to that end Craig Burkhart mentioned a paper by Hagler, Hwang, and Stockfish, which used QM calculations to parameterize (what I assume is) the DISCOVER 2nd generation force field. Does anyone know the exact reference? Thanks. Howard -- Howard E. Alper, Ph.D. Dept. of Chemistry and The Advanced Materials Research Institute University of New Orleans New Orleans, LA 70148 504-280-7216 - Helping molecules find happiness for almost a 5th of a century.