From chemistry-request@ccl.net Wed May 28 23:22:13 2003 -0400 Return-Path: Received: from sccrmhc02.attbi.com (sccrmhc02.attbi.com [204.127.202.62]) by ccl.net (8.11.6+Sun/8.11.6/OSC 2.1) with ESMTP id h4T3MCt17541 for ; Wed, 28 May 2003 23:22:12 -0400 (EDT) Received: from c1353359a (12-208-133-89.client.attbi.com[12.208.133.89]) by attbi.com (sccrmhc02) with SMTP id <2003052903221100200al6r3e>; Thu, 29 May 2003 03:22:11 +0000 Reply-To: From: "Mark Thompson" To: Cc: Subject: bug-fix release for ArgusLab 3.1 Date: Wed, 28 May 2003 20:18:06 -0700 Message-ID: <000a01c32590$ec504df0$0300a8c0/at/attbi.com> We've just uploaded a bug-fix release for ArgusLab 3.1 Several users reported the settings file version sometimes being read incorrectly upon startup. This release is free for licensed users and can be installed over an existing ArgusLab 3.1 installation. To link directly to the download page click on: http://www.planaria-software.com/download_3_0.htm ================================= Mark Thompson, Ph.D. Planaria Software PO Box 55207 Seattle, WA 98155 FAX: 206-440-3305 ArgusLab is available at: http://www.arguslab.com ================================= From chemistry-request@ccl.net Thu May 29 01:19:23 2003 Received: from sccrmhc03.attbi.com (sccrmhc03.attbi.com [204.127.202.63]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T5JMgC026856 for ; Thu, 29 May 2003 01:19:22 -0400 Received: from c1353359a (12-208-133-89.client.attbi.com[12.208.133.89]) by attbi.com (sccrmhc03) with SMTP id <20030529051921003004d5cre>; Thu, 29 May 2003 05:19:22 +0000 Reply-To: From: "Mark Thompson" To: "Geoff Hutchison" , Cc: Subject: CCL: UV spectrum calculation (new question) Date: Wed, 28 May 2003 22:15:16 -0700 Message-ID: <000001c325a1$4a9384d0$0300a8c0.-at-.attbi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) In-Reply-To: Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Geoff, Great, I'll look at your paper. I've obtained quite reasonable results with my ZINDO/RPA while specifying the active space (e.g. 10 occupied, 10 virtual orbitals). The calculations do take longer than ZINDO/CIS, but for typical organic molecules of ~50 atoms, it's still just seconds to do the whole calculation on a commodity PC. I've obtained good results on the bacteriochlorophyll-b dimer from the R. viridis xray structure for both UV/vis absorption and rotary strength with good agreement between length and velocity operators for the transition moments. It's remarkable really, that ZINDO mimics a complete basis set in some sense since the length-velocity operator agreement is exact only in the limit of a complete basis. In any case, we're working on our own ZINDO/RPA parameterization and we will slipstream the parameters into the nearest ArgusLab build when they are available. In the meantime, one can run ZINDO/RPA with ArgusLab 3.1 by selecting the "calculate rotary strengths" button. This turns on both ZINDO/CIS and ZINDO/RPA for the spectroscopic calculation of the molecule (the CIS results are given first in the output and the RPA come after that). In our 3.5 release (nearing completion), all the spectroscopic results can be viewed more cleanly with our new property control. The UV/vis calculation setup is also cleaner with respect to delineating RPA vs. CIS Mark ================================= Mark Thompson, Ph.D. Planaria Software PO Box 55207 Seattle, WA 98155 FAX: 206-440-3305 ArgusLab is available at: http://www.arguslab.com ================================= > -----Original Message----- > From: Computational Chemistry List [mailto:chemistry-request.-at-.ccl.net]On > Behalf Of Geoff Hutchison > Sent: Wednesday, May 28, 2003 3:50 PM > To: sergiusz.kwasniewski.-at-.luc.ac.be > Cc: chemistry.-at-.ccl.net > Subject: CCL:UV spectrum calculation (new question) > > > > I'm curious how well INDO/S-RPA works against INDO/S-CIS for large > > systems. > > I only remember one article by M. Zerner quite a long time ago.. > > A comparison of ZINDO/RPA vs. ZINDO/CIS on various conjugated oligomers > is included as part of my larger paper (which also compares HF/CIS, > HF/RPA, TDDFT/CIS and TDDFT/RPA): > > J. Phys. Chem. A (2002) 106(44) p. 10596-10605. > > At the time, the only ZINDO/RPA I could put my hands on was contributed > by Sergei Tretiak and Shaul Mukamel, which didn't let me set an active > space. (In fact, it uses the Lanczos method for a full active space.) A > parameterization for RPA calculations or a smaller active space would > probably help. > > > Probably the calculation takes a lot longer (Hamiltonian matrix is > > four times > > as big since mixing of particle-hole and hole-particle excitations > > occurs). > > I think with an efficient implementation, it's not really an issue. > (It's a semiempirical calc. after all!) I never did timing tests, but I > can relate that up to 60+ heavy atoms, ZINDO/RPA calculations still > only took something on the order of minutes (on an 800MHz Pentium III, > so faster times nowadays). > > > Are there any problems with numerical instabilities with the > > implemented RPA scheme in a INDO/S-RPA calculation ? > > I'll second Dr. Thompson's comments. I haven't run any extremely > "strange" molecules, but haven't seen any instabilities. > > Cheers, > -Geoff > > -- > -Geoff Hutchison > Ratner/Marks Groups (847) 491-3295 > Northwestern Chemistry http://www.chem.northwestern.edu/ > > > > -= This is automatically added to each message by mailing script =- > To send e-mail to subscribers of CCL put the string CCL: on your > Subject: line > and send your message to: CHEMISTRY.-at-.ccl.net > > Send your subscription/unsubscription requests to: > CHEMISTRY-REQUEST.-at-.ccl.net > HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs > > If your mail is bouncing from CCL.NET domain send it to the maintainer: > Jan Labanowski, jkl.-at-.ccl.net (read about it on CCL Home Page) > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > > > > > > From chemistry-request@ccl.net Thu May 29 02:09:21 2003 Received: from sungod.ccs.yorku.ca (mailrelay.yorku.ca [130.63.236.104]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T69LgC027527 for ; Thu, 29 May 2003 02:09:21 -0400 Received: from curl.gkcl.yorku.ca (curl.gkcl.yorku.ca [130.63.232.224]) by sungod.ccs.yorku.ca (8.12.9/8.12.8) with ESMTP id h4T69KuF011115 for ; Thu, 29 May 2003 02:09:20 -0400 (EDT) Received: (from chan@localhost) by curl.gkcl.yorku.ca (8.10.0/8.10.0) id h4T69KN13455 for chemistry*at*ccl.net; Thu, 29 May 2003 02:09:20 -0400 (EDT) From: Wai-To Chan Message-Id: <200305290609.h4T69KN13455*at*curl.gkcl.yorku.ca> Subject: Re: calculating atomic charges with AIMPAC To: chemistry*at*ccl.net Date: Thu, 29 May 2003 02:09:19 -0400 (EDT) X-Mailer: ELM [version 2.5 PL3] MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< I currently have an undergraduate student who is attempting to compare charges of methylated benzene and ferrocene systems using the AIMPAC software. In the case of hexamethyl benzene and the ferrocene analogues he is getting very large negative charges on some the ring carbons (-5 to -20) and small positive charges on iron (0.5) Can anyone tell me if there is a problem with the program or if the ring system can not be treated with this method? >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> I remember reading about the tendency of AIM changes to be larger in magnitude than expected on the basis of chemical intuition. There is a paper by Kenneth Wiberg from the early 90s in J. comp. chem. about the comparison of several atomic-charge partitioning scheme including Bader's you may find useful. You may also consider using the Natural atomic charges method if you want faster and less 'surprising' results. Wai-To Chan From chemistry-request@ccl.net Thu May 29 01:58:13 2003 Received: from wyoming.com (WYOMING.COM [206.54.112.2]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T5wDgC027382 for ; Thu, 29 May 2003 01:58:13 -0400 Received: from [204.227.221.209] (account madscientist HELO wyoming.com) by wyoming.com (CommuniGate Pro SMTP 4.0.6) with ESMTP id 35390406 for chemistry*at*ccl.net; Wed, 28 May 2003 23:58:12 -0600 Message-ID: <3ED5A166.3BAE2D51*at*wyoming.com> Date: Wed, 28 May 2003 23:57:59 -0600 From: Philip Stortz Reply-To: madscientist*at*wyoming.com X-Mailer: Mozilla 4.75C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; U; PPC) X-Accept-Language: en MIME-Version: 1.0 To: chemistry*at*ccl.net Subject: Re: CCL:orbitals Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The idea of orbitals seems reasonable, but obviously there have been a lot of simplifications in the already hugely complex programs written for ccl. I would also argue, that the programs currently in use as well as the parameters for each element have been aggressively tweaked to match lab results. Clearly current ccl code has enough degrees of freedom to model nearly any thing else that's currently modeled by subtly changing parameters and/or code. CCL programs are useful tools only because of this "tweaking", though any tool is better than no tool usually. obviously correlation between calculation and result does not prove a theory "true", particularly when you have feedback between experiment and model development. also, most programs do seem to do better on large systems, probably because of the tendency for errors to cancel each other to some extent. one of my favorite examples: there was a period during the late 60's and early seventies when the length of ladies skirts very, very closely matched sunspot activity, for a whole 11 year cycle! One might have concluded that fashion can predict solar activity or vice versa, and developed the appropriate predicative code, at least until the trend changed. As i said, it seems to me that modern ccl code could be modified for nearly any predictive activity and the constant tweaking means that it must approach consistency with experimental results, no matter whether the original idea has any basis in "reality" or not. Reality is in the mind of the beholder, as long as you agree with the majority, otherwise you still think you understand "reality", but others will think you foolish or insane. Agreement with experimental data supports a theory, until that experimental data can no longer be accommodated into the existing model or someone comes up with a simpler or more accurate theory (even if more complicated, or fundamentally less "real"). Finally, as of last year (I believe it was last year, sorry i don't have a reference), fractional electrons exist, at least in liquid helium near absolute zero, where they were able to induce i believe at least 6 different species of charge carrier with different charge to mass ratios. It wouldn't surprise me if this partially explains (assuming you take the experimental observation to be "real") Pi bonds. And considering it's place on the periodic table (another highly useful but imperfect tool) i'd suspect this may account for some of the problems modeling lithium compounds. It might even relate to superconductivity, which is observed, but still poorly understood i believe. I only wish there were good code for inorganic crystal electronic spectrum calculations, but that's not where most people see the money, so the code base hasn't been tweaked in that direction extensively. Still, i'd prefer i guide who happens to be right over no guidance at all, it provides that "comfort" we all seek, and of course a way to shift the blame. ("it's not my fault we wasted billions, that was what the program said to do, blame the developers") elewars wrote: > > 2003 May 28 > > Are MOs physically real? This is a meaningful question only if there is some > experiment or observation that could provide an answer _yes_ or _no_. Is there, > at least in principle, such an experiment or observation? ----------- -- Philip Stortz -- To be nobody but yourself when the whole world is trying its best night and day to make you everybody else is to fight the hardest battle any human being will ever fight. -- E.E. Cummings From chemistry-request@ccl.net Thu May 29 02:05:38 2003 Received: from wyoming.com (WYOMING.COM [206.54.112.2]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T65bgC027496 for ; Thu, 29 May 2003 02:05:38 -0400 Received: from [204.227.221.209] (account madscientist HELO wyoming.com) by wyoming.com (CommuniGate Pro SMTP 4.0.6) with ESMTP id 35392087 for chemistry*at*ccl.net; Thu, 29 May 2003 00:05:36 -0600 Message-ID: <3ED5A324.B8CC04AE*at*wyoming.com> Date: Thu, 29 May 2003 00:05:24 -0600 From: Philip Stortz Reply-To: madscientist*at*wyoming.com X-Mailer: Mozilla 4.75C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; U; PPC) X-Accept-Language: en MIME-Version: 1.0 To: chemistry*at*ccl.net Subject: CCL: Crystal building tools References: <000001c325a1$4a9384d0$0300a8c0*at*attbi.com> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To help with modeling electronic spectra of crystals, is there a good package out there to build the initial crystal with at least a couple dozen unit cells if i know the crystal structure? I've tried this with Cache a few years ago, and the drawing tool was not very useful and very painful for building 3d crystals, bonds often wound up in odd places where i didn't mean to put them and it was very hard to view the bonds to see these problems. Thanks again. -- Philip Stortz -- To be nobody but yourself when the whole world is trying its best night and day to make you everybody else is to fight the hardest battle any human being will ever fight. -- E.E. Cummings From chemistry-request@ccl.net Thu May 29 02:01:05 2003 Received: from wyoming.com (WYOMING.COM [206.54.112.2]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T614gC027437 for ; Thu, 29 May 2003 02:01:05 -0400 Received: from [204.227.221.209] (account madscientist HELO wyoming.com) by wyoming.com (CommuniGate Pro SMTP 4.0.6) with ESMTP id 35391074 for chemistry*at*ccl.net; Thu, 29 May 2003 00:01:03 -0600 Message-ID: <3ED5A212.9753567*at*wyoming.com> Date: Thu, 29 May 2003 00:00:51 -0600 From: Philip Stortz Reply-To: madscientist*at*wyoming.com X-Mailer: Mozilla 4.75C-CCK-MCD {C-UDP; EBM-APPLE} (Macintosh; U; PPC) X-Accept-Language: en MIME-Version: 1.0 To: chemistry*at*ccl.net Subject: CCL: Electronic Spectra of inorganic crystals Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Time for my annual hopeful question, is there currently a program that calculates electronic spectra of inorganic crystals (and their surfaces) with any accuracy? I know zindo/indo works good for organics, but what about for inorganic crystals? Does mopac do this well, i know fujitsu was working on it? I'm hoping someone on this list has had good luck with such problems as it's often hard to tell from what vendors claim just how good or useful a program may be. I'm hoping to buy software in a couple of months that will hopefully let me test fewer samples to find what i'm looking for. Even if not absolutely accurate, something that reasonably predicted the change in spectrum from one inorganic material to another to help steer the research would be very, very helpful. Thank you in advance for any help you can provide. Re: spam problems (listmom), you might look into spamcop.net, they have a fairly good system with a blacklist of open relays and several addresses that are designed to be spam magnets so they can detect other spam. I'll use their service myself if i ever start getting more than one or two spams a month, i've been very lucky as far as spam goes. I've also aggressively reported spam, which does seem to help (spam cop will let anyone do that for free). -- Philip Stortz -- To be nobody but yourself when the whole world is trying its best night and day to make you everybody else is to fight the hardest battle any human being will ever fight. -- E.E. Cummings From chemistry-request@ccl.net Thu May 29 04:36:22 2003 Received: from dove2.cf.ac.uk (dove2.cf.ac.uk [131.251.1.170]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T8aMgC031620 for ; Thu, 29 May 2003 04:36:22 -0400 Received: from thor.cf.ac.uk ([131.251.0.1]) by dove2.cf.ac.uk with esmtp (Exim 4.05) id 19LItX-0003C7-00; Thu, 29 May 2003 09:36:19 +0100 Received: from localhost (sacjap@localhost) by thor.cf.ac.uk (8.8.8/8.6.12) with ESMTP id JAA19095; Thu, 29 May 2003 09:36:19 +0100 (BST) Date: Thu, 29 May 2003 09:36:18 +0100 (BST) From: Jamie Platts X-X-Sender: sacjap@thor To: "Dr. Julia Metzker" cc: chemistry~at~ccl.net Subject: Re: CCL:calculating atomic charges with AIMPAC In-Reply-To: <5.2.1.1.0.20030528150904.00b7b488~at~jm45.imap.york.ac.uk> Message-ID: References: <5.2.1.1.0.20030528150904.00b7b488~at~jm45.imap.york.ac.uk> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Julia - have you checked the quality of the integrations? This is usually best done with the number reported as "L" (I think the fourth quantity output by proaimv). This is the value of the Laplacian integrated over the atomic basin, which should vanish - values of 10^-3 are borderline OK, lower is obviously better. Also check that the total populations add up to the no. electrons in your system. For a system like ferrocene, with rings (and cages?), you might need to tweak some of the surface-tracing paramaters - see the top of the proaimv.f code. If this isn't the source of your problem, you could try the integration methods in Gaussian (disappointing in my experience) or Popelier's MORPHY, which is generally very robust. Hope this helps - Jamie > I currently have an undergraduate student who is attempting to compare > charges of methylated benzene and ferrocene systems using the AIMPAC > software. In the case of hexamethyl benzene and the ferrocene analogues he > is getting very large negative charges on some the ring carbons (-5 to -20) > and small positive charges on iron (0.5) Can anyone tell me if there is a > problem with the program or if the ring system can not be treated with this > method? > > Thank you, > Julia Metzker > University of York > Department of Chemistry From chemistry-request@ccl.net Thu May 29 05:56:24 2003 Received: from maties3.sun.ac.za (maties3.sun.ac.za [146.232.128.43]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T9uMgC001351 for >ccl.net>; Thu, 29 May 2003 05:56:23 -0400 Received: from stbex02.stb.sun.ac.za ([146.232.20.17] helo=STBEVS01.stb.sun.ac.za) by maties3.sun.ac.za with esmtp (Exim 4.10) id 19LK8z-0008Ln-00; Thu, 29 May 2003 11:56:21 +0200 X-MimeOLE: Produced By Microsoft Exchange V6.0.6334.0 content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: CCL: Re:Eigenvalues of hessian Date: Thu, 29 May 2003 11:56:21 +0200 Message-ID: <21F5207AECE530428EBD1EBF78DA95B77944AE<>STBEVS01.stb.sun.ac.za> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: CCL: Re:Eigenvalues of hessian Thread-Index: AcMlyI6BnLRXKy0KSaK8Y0qNk94dAQ== From: "Dillen Jan >sun.ac.za>" >sun.ac.za> To: >ccl.net> Cc: >stud-mailer.uni-marburg.de> X-Scanner: exiscan for exim4 (http://duncanthrax.net/exiscan/) *19LK8z-0008Ln-00*CrAi6RM/EP6* Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h4T9uOgC001352 Dear Daniel, I think that you should specify more clearly what system you are calculating (e.g. isolated molecule, crystal, use of restraints/constraints, etc.) If you do a full energy minimisation on an isolated molecule, then you MUST have 3N-6 (or 3N-5 for a linear molecule) zero frequencies if the frequency calculation is done from a Hessian matrix calculated with Cartesian derivatives, and the system was properly minimised. If the energy surface is flat, you may have to make the convergence criteria more stringent in order to reach the minimum more closely. Each negative (imaginary) frequency implies a maximum along the corresponding eigenvector, each additional zero an inflection point. For crystals the number of non-zero frequencies depends on how the calculation is done and/or what the space group of the system is. If space group symmetry is ignored, 3 (translational) modes are zero. If space group symmetry is taken into account, a centro-symmetrical space group will have no zero frequencies, whereas for the others it depends on whether the symmetry elements fix the origin along each of the 3 axes. If you are using special features, e.g. rigid groups, fixing some internal coordinates, etc., there is no general rule and the number of zero-frequencies must be worked out based on the actual number of degrees of freedom of the system. If you are using a force field with additional "goodies" like lone pairs, dummy atoms, or other particles with undefined mass, the answer depends on how these artefacts are projected out. Groeten Jan Dillen > Does anybody of you know, why there are more than 3N-6 nonzero > eigenvalues of the hessian? I am especially interested in the question, > whether this error is due to the errors made by differentiation or > whether they are due to an error within the method itself. I usually get > only 3 zero eigenvalues. No matter if the hessian is calculated on a > stationary point or not. From chemistry-request@ccl.net Thu May 29 06:05:45 2003 Received: from real.cyf-kr.edu.pl (real.cyf-kr.edu.pl [149.156.2.10]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TA5igC001612 for >ccl.net>; Thu, 29 May 2003 06:05:44 -0400 Received: from kinga.cyf-kr.edu.pl (kinga.cyf-kr.edu.pl [149.156.2.32]) by real.cyf-kr.edu.pl (8.11.6/8.11.6) with ESMTP id h4TA0uX01970 for >ccl.net>; Thu, 29 May 2003 12:00:56 +0200 X-RAV-AntiVirus: This e-mail has been scanned for viruses on host: real.cyf-kr.edu.pl Received: (from mykrol@localhost) by kinga.cyf-kr.edu.pl (8.9.3/8.9.3) id MAA11808; Thu, 29 May 2003 12:00:53 +0200 (METDST) Date: Thu, 29 May 2003 12:00:50 +0200 (METDST) From: Marcin Krol >cyf-kr.edu.pl> To: chemistry<>ccl.net Subject: DCCM in CHARMM Message-ID: >kinga.cyf-kr.edu.pl> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, I'd like to calclulate dynamic cross-corelation matrices in CHARMM. So firstly, I oriented all trajectiory frames on the crystal structre (merge orient keyword). Then I calculated average structure from the oriented trajectory and defined times series as fluctuation of all CA atoms (each CA atoms represents motions of the whole aminoacid). When I had times series for all CA atoms I started to calculate /sqrt*sqrt where Q(a) is one time series and Q(b) another. In principle I should get what I want, but the values I obtain are between 0.8-1.0 no matter what time series are chosen. It looks like motions of all aminoacids in my protein are highly correlated, which is not very probable. What can be the reason for it? I have oriented the whole trajectory on the crystal structure so translations and rotations of the center of mass should be removed. Thanks in advance for any suggestions how to obtain correct DCCM in CHARMM Marcin Krol Below is the charmm script I use. 1. To remove trans and rot of the center of mass open crystal struture read crystal structure coor copy comp open input traj files open output traj files merge firstu 41 nunit 24 skip 500 outputu 71 nfile 200 sele segid (name of my protein) end orient sele type CA end (so I should get a new traj files which contain only protein atoms and frames are oriented on crystal structure CA atoms) 2. Caluculate DCCM open crystal structure read crystal structure open oriented traj files coor dyna comp firsu 41 nunit 5 correl maxt 1050 maxseries 433 maxatoms 433 cutnb 2. set res 1 label enterdccm enter a@res fluc r sele resi @res .and. type CA end traj first 41 nunit 5 incr res by 1 if res lt 433 goto eneterdccm set res 1 label one set res label two mantime a@res fluct a@resn incr resn by 1 if resn lt 433 goto two incr res by 1 if res lt 433 goto one stop end From chemistry-request@ccl.net Thu May 29 06:49:52 2003 Received: from web20513.mail.yahoo.com (web20513.mail.yahoo.com [216.136.174.44]) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TAnqgC003114 for >ccl.net>; Thu, 29 May 2003 06:49:52 -0400 Message-ID: <20030529104951.55845.qmail<>web20513.mail.yahoo.com> Received: from [212.144.165.241] by web20513.mail.yahoo.com via HTTP; Thu, 29 May 2003 03:49:51 PDT Date: Thu, 29 May 2003 03:49:51 -0700 (PDT) From: Noriaki Ikuta >yahoo.com> Subject: CCL: ic50 values in dtp aids dataset To: chemistry<>ccl.net Cc: noriaki_ikuta<>yahoo.com MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii I have three questions regarding the DTP AIDS Antiviral Screening dataset. http://dtp.nci.nih.gov/docs/aids/aids_data.html i) The explanation for the IC50 is "concentration of compound that inhibits the growth of uninfected cells by 50%" I do not understand this. I throught it would be of interest to inhibit the virus or at least the growth of infected cells? ii) How is the conclusion active/inactive drawn from those IC50 values? I have included a few example entries with ConcUnit = M in the mail. HiConc - the highest concentration used in the dose response Flag - when different experiments are averaged, "=" indicated that all the experiments reached IC50 ">" indicates that at least one of the experiments don't actually reach an EC50 and the highest tested concentration is used for the value of IC50 IC50 - concentration of compound that inhibits the growth of uninfected cells by 50% Conclusion - CA active, CM moderately active, CI inactive NSC Log10HiConc Flag Log10IC50 Conclusion 646607 -5.4 > -5.42 CA 646607 -4.5 = -5.29 CA 646607 -3.7 = -5.79 CA 646609 -4.5 = -5.78 CA 646609 -3.7 = -5.61 CA 646610 -4.5 = -5.77 CA 646610 -3.7 = -5.88 CA 639762 -3.7 = -4.75 CM 639762 -3.6 = -4.88 CM 639763 -4.4 = -4.63 CM 639765 -3.9 = -4.69 CM 639766 -4.6 > -4.79 CM 626162 -3.7 < -7.2 CI 629266 -3.7 < -7.2 CI 630958 -3.9 = -4.74 CI 630959 -3.2 = -3.85 CI 639915 -4.7 > -4.7 CI 639975 -3.9 > -5.05 CI 640064 -5 > -5 CI 640064 -3.7 > -3.7 CI iii) Are there papers available explaining this dataset in depth? Noriaki __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com From chemistry-request@ccl.net Thu May 29 05:59:42 2003 Received: from Stud-Mailer.Uni-Marburg.DE (Stud-Mailer.Uni-Marburg.DE [137.248.9.7]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T9xfgC001466 for >ccl.net>; Thu, 29 May 2003 05:59:42 -0400 Received: from charly.chemie.uni-marburg.de (pc15160.Chemie.Uni-Marburg.DE [137.248.152.201]) by Stud-Mailer.Uni-Marburg.DE (8.12.8p1/8.12.8) with ESMTP id h4T9w6Jr081172 for >ccl.net>; Thu, 29 May 2003 11:58:06 +0200 Subject: RE: Eigenvalues of hessian From: "Daniel R. Rohr" >students.uni-marburg.de> To: ccl >ccl.net> Content-Type: text/plain; charset=ISO-8859-1 X-Mailer: Evolution/1.0 (Preview Release) Date: 29 May 2003 11:54:00 +0200 Message-Id: <1054202040.25209.3.camel@charly> Mime-Version: 1.0 X-MailScanner-Information: see http://www.uni-marburg.de/hrz/services/spamcheck/ X-MailScanner: Found to be clean Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h4T9xggC001467 Hi CCLer Maybe I should clearify my question: I do know that in the Hessians calculated, there is usually translation and rotation. I want to avoid speaking of frequencies, because we are talking about Hessians which are not necessarily calculated at stationary points. That#s why I talk of the eigenvalues. In an excat method there should be exactly 3N-6 nonzeroeigenvalues and 6 exact zeros. ANY method I tested (HF, DFT and MP2) with ANY Program I tested (Gaussian, Turbomole, Jaguar) gives me 3N nonzero Eigenvalues. The question is now. Why are there too many nonzero Eigenvalues? In my eyes there are two possible reasons: 1. The method is ok but there are errors made in the derivations of the method. Maybe neglecting terms in the differentiation. Or maybe asumptions that hold true only for stationary points. 2. The method is not ok. This means that I do get different Energies for different orientations of my molecule (As is the case in DFT) Of course it is also possible that there is a mixture of both. I would like to know which error is the bigger one? And maybe get a reference for this question. Another point which I do not understand: Some answers state that there is only little translation and rotation since the frequencies are close to zero. Is the absolute value of an eigenvalue a measure for the quantity (whaterver that means) of its eigenvector???? And if that is the case why is there "more" translation and rotation at nonstationary points? I get "frequencies" (this is eigenvalues of the Hessian) between 400 and 850 cm^-1 for Water with HF 6-31G** (Gaussian calculation): Low frequencies --- -410.8545 -0.0007 -0.0006 0.0017 489.4045 853.6142 Low frequencies --- 1891.8502 3808.4304 3946.8117 These are of course the frequencies calculated BEFORE projection of translation and rotation. I hope you can help me on that Daniel Rohr Hi Daniel, Your question is a bit tricky in that the answer depends on the method you are using. If you are calculating Hartree-Fock or MP2 hessians then I think you should get 6 zero eigenvalues and 3N-6 non-zero eigenvalues at a stationary point. Even if the hessian is obtained by numerical differentiation of gradients this should come out allright, otherwise the code is broken. If you are doing a DFT calculation life typically gets a bit more complicated. If the DFT hessian is implemented as a proper second derivative of the energy expression with respect to the nuclear coordinates you should still get 6 zero eigenvalues. However this requires that the code takes the 2nd derivatives of the Becke weights (A.D. Becke, J.Chem.Phys. 88 (1988) 2547-2553) into account as well. Some codes don't bother calculating these terms in which case the strict translational invariance of the hessian is violated. This typically expresses itself in 3 additional non-zero but small valued eigenvalues. If you are calculating the vibrational modes by numerical differentiation of DFT gradients then there may still be a problem if your molecule has symmetry. Many quantum chemistry packages will reorientate your molecule such that the main axes of ineratia coincide with the Cartesian axes of your coordinate system. This means that different points in the numerical differentiation algorithm may be calculated at different orientations. However, the DFT quadratures are not always invariant for rotations. For example if you want to do calculations on H2 and you are going to use 6 points per H atom all placed on the Cartesian axes than the energy you'll get for the following two orientations will not be the same even if the bond distance is: H H | \ H H Again this will lead to additional non-zero but small eigenvalues. In this case it may help to switch the use of symmetry off, so that all points are calculated in exactly the same orientation. As you have given very little details about exactly which calculation you are doing I can not say for sure which of the above reasons is the most like cause. However I think it is most likely one of these and with some experimentation and a good look at the documentation for your program you should be able to find out what is happening. Best wishes, Huub van Dam -----Original Message----- From: Daniel R. Rohr To: ccl Sent: 5/28/03 3:20 PM Subject: CCL:Eigenvalues of hessian Hi CCLers Does anybody of you know, why there are more than 3N-6 nonzero eigenvalues of the hessian? I am especially interested in the question, whether this error is due to the errors made by differentiation or whether they are due to an error within the method itself. I usually get only 3 zero eigenvalues. No matter if the hessian is calculated on a stationary point or not. Thanks for your help Daniel Rohr -- ----------------------------------------------------------------------- Daniel Rohr Fachbereich Chemie der Philipps-Universitdt Tel.: +49-6421-28-25686 Hans-Meerwein-Stra_e Fax.: +49-6421-28-25566 D-35043 Marburg eMail: rohrd<>stud-mailer.uni-marburg.de Germany ----------------------------------------------------------------------- -= This is automatically added to each message by mailing script =- To send e-mail to subscribers of CCL put the string CCL: on your Subject: line and send your message to: CHEMISTRY<>ccl.net Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST<>ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs If your mail is bouncing from CCL.NET domain send it to the maintainer: Jan Labanowski, jkl<>ccl.net (read about it on CCL Home Page) -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ -- ----------------------------------------------------------------------- Daniel Rohr Fachbereich Chemie der Philipps-Universitdt Tel.: +49-6421-28-25686 Hans-Meerwein-Stra_e Fax.: +49-6421-28-25566 D-35043 Marburg eMail: rohrd<>stud-mailer.uni-marburg.de Germany ----------------------------------------------------------------------- From chemistry-request@ccl.net Thu May 29 04:12:10 2003 Received: from pado.krict.re.kr (pado.krict.re.kr [203.250.2.9]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4T8C8gC030824 for ; Thu, 29 May 2003 04:12:09 -0400 Received: from anakin ([203.250.4.112]) by pado.krict.re.kr (v3smtp 8.11.6.8/8.11.2) with ESMTP id h4T8DJW13674; Thu, 29 May 2003 17:13:19 +0900 From: =?utf-8?B?6rmA7J6l67Cw?= To: "'Guosheng Wu'" , Subject: RE: orbitals Date: Thu, 29 May 2003 17:12:32 +0900 Message-ID: <000501c325ba$1061b750$7004facb@anakin> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.2627 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 In-Reply-To: <20030528224544.55154.qmail=at=web10406.mail.yahoo.com> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h4T8CAgC030825 * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * Jangbae Kim Researcher Bio-Organic Science Division Korea Research Institute of Chemical Technology (KRICT) P.O.Box 107, Yuseong, Taejeon, 305-600, Korea TEL : +82-42-860-7067 FAX : +82-42-861-0307 E-mail : jbkim=at=krict.re.kr * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * > -----Original Message----- > From: Computational Chemistry List [mailto:chemistry-request=at=ccl.net] On > Behalf Of Guosheng Wu > Sent: Thursday, May 29, 2003 7:46 AM > To: chemistry=at=ccl.net > Subject: CCL:orbitals > > --- Rene Fournier - 1999-07-07 wrote: > > Leif Laaksonen makes an important point: only observables > > are "real" in QM. So the question arises: is it possible to > > define orbitals as QM observables. It entirely depends on > > how you define the word "orbital". If one says orbitals are > > one-electron functions and nothing more, then of course orbitals > > can be made to be QM observables. For example, just integrate > > an exact squared wavefunction over all electron coordinates > > but one to get the electron density, then take the square > > root, multiply by a phase factor and you get "density orbitals". > > [G Hunter, Int J Quantum Chem 9 (1975) 237]. They may not be > > very useful as orbitals, but they are "real" in the sense of > > QM observables. > > It's not clear for me how you define "QM observable"? > > MO is certainly not "observable", if "observable" is defined as something > can be measured directly or indirectly by an experiment. For example, > electronic density, ionization energy, and so on. > > The issue is related with the phase factor you mentioned: > From experimental data, one can get |MO|^2 at a certain position, > but here "MO" is not unique: MO = |MO|*[cos(theta)+i*sin(theta)], > where "theta" can be any arbitary value. > This may be traced back to the Uncertainty Principles. > > Similarly to my original post, 0, 1, 2,... are countable (defined as > non-negative integers). However, if one do the square of 1, +1 and -1 > will come out, not unique --- 2 directions in the 1-D space; > Do the square of -1, we get +i and -i, 2 direction in the 2-D space... > It seems we can safely say that -1 & i are not countable!? > > By the way, does it really matter if "MO" is observable or not in any > sense? > Maybe useful for teaching. > > Guosheng > > > > --- Rene Fournier - 1999-07-07 wrote: > > > Leif Laaksonen makes an important point: only observables > > > are "real" in QM. So the question arises: is it possible to > > > define orbitals as QM observables. It entirely depends on > > > how you define the word "orbital". If one says orbitals are > > > one-electron functions and nothing more, then of course orbitals > > > can be made to be QM observables. For example, just integrate > > > an exact squared wavefunction over all electron coordinates > > > but one to get the electron density, then take the square > > > root, multiply by a phase factor and you get "density orbitals". > > > [G Hunter, Int J Quantum Chem 9 (1975) 237]. They may not be > > > very useful as orbitals, but they are "real" in the sense of > > > QM observables. > > > Orbitals are always defined as one-electron functions, > > > but THAT does not make them "unreal". The electron density > > > is a one electron function and very real for any N-electron > > > system, not only for the H atom. > > > One "ab initio" way to define orbitals is to start from > > > exact wavefunctions and calculate the overlap of a N-electron > > > wavefunction (neutral) and the ground and various excited > > > states a (N-1)-electron wavefunction (ion) (Dyson orbitals are > > > defined in such a way I think). I'm sure there are definitions > > > that will make such orbitals unique, but does that make them > > > QM observables? I don't think so, but I'd like to hear from > > > the CCL community on that. There are other ways to define orbitals > > > > from exact wavefunctions, I'm sure: are any of them QM observables? > > > As for Kohn-Sham orbitals, they may not be real individually > > > but the N lowest energy ones squared and summed give the true > > > electron density (an observable) and the negative of the KS HOMO > > > energy is exactly equal to the first IP (another observable) so > > > there are at least some bits of reality built in KS orbitals! > > > > > > Here's some relevant articles: > > > > > > 1) Kohn, Becke, Parr, J Phys Chem 100 (1996) 12974 > > > 2) "Kohn-Sham Orbitals and Orbital Energies: Fictitious Constructs but > > > Good Approximations All the Same" > > > by S. Hamel, P. Duffy, M.E. Casida, and D.R. Salahub > > > Journal of Electron Spectroscopy and Related Phenomena (2002?) > > > 3) Chong, Gritsenko, Baerends, J Chem Phys 116 (2002) 1760 > > > > > > -------------------------------------------------------------------- > > > | Rene Fournier | Office: 303 Petrie | > > > | Chemistry Dpt, York University| Phone: (416) 736 2100 Ext. 30687 | > > > | 4700 Keele Street, Toronto | FAX: (416)-736-5936 | > > > | Ontario, CANADA M3J 1P3 | e-mail: renef=at=yorku.ca | > > > -------------------------------------------------------------------- > > > | http://www.chem.yorku.ca/profs/renef/ | > > > -------------------------------------------------------------------- > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > __________________________________ > > Do you Yahoo!? > > Yahoo! Calendar - Free online calendar with sync to Outlook(TM). > > http://calendar.yahoo.com > > > > > __________________________________ > Do you Yahoo!? > Yahoo! Calendar - Free online calendar with sync to Outlook(TM). > http://calendar.yahoo.com > > > -= This is automatically added to each message by mailing script =- > To send e-mail to subscribers of CCL put the string CCL: on your Subject: > line > and send your message to: CHEMISTRY=at=ccl.net > > Send your subscription/unsubscription requests to: CHEMISTRY- > REQUEST=at=ccl.net > HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs > > If your mail is bouncing from CCL.NET domain send it to the maintainer: > Jan Labanowski, jkl=at=ccl.net (read about it on CCL Home Page) > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ > > From chemistry-request@ccl.net Thu May 29 08:56:31 2003 Received: from srvus010.unitedcatalysts.com (srvus010.unitedcatalysts.com [208.23.162.8] (may be forged)) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TCuVgC005753 for ; Thu, 29 May 2003 08:56:31 -0400 Received: by lvlxch02.ntdomain.americas.sc-world.com with Internet Mail Service (5.5.2653.19) id ; Thu, 29 May 2003 08:54:30 -0400 Message-ID: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF1{at}lvlxch02.ntdomain.americas.sc-world.com> From: "Shobe, Dave" To: "'CCL'" Subject: RE: calculating atomic charges with AIMPAC Date: Thu, 29 May 2003 08:54:29 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-2d4cf94c-77cb-47a7-bd49-dee6013c6e1a" This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------=_NextPartTM-000-2d4cf94c-77cb-47a7-bd49-dee6013c6e1a Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C325E1.71401400" ------_=_NextPart_001_01C325E1.71401400 Content-Type: text/plain; charset="iso-8859-1" But a charge of -20 almost certainly is the result of assigning the electrons of one atom's basin to another atom. Something which occurs as a result of the program failing to find the boundary between them. At any rate, I would try a 2nd AIM calculation program if one is available, before giving up. E.g., I often find Gaussian's AIM code halts with a (usually cryptic) error message, but for the same molecule AIM2000 may complete the calculation with no problem. -----Original Message----- From: Wai-To Chan [mailto:chan{at}curl.gkcl.yorku.ca] Sent: Thursday, May 29, 2003 2:09 AM To: chemistry{at}ccl.net Subject: CCL:calculating atomic charges with AIMPAC <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< I currently have an undergraduate student who is attempting to compare charges of methylated benzene and ferrocene systems using the AIMPAC software. In the case of hexamethyl benzene and the ferrocene analogues he is getting very large negative charges on some the ring carbons (-5 to -20) and small positive charges on iron (0.5) Can anyone tell me if there is a problem with the program or if the ring system can not be treated with this method? >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> I remember reading about the tendency of AIM changes to be larger in magnitude than expected on the basis of chemical intuition. There is a paper by Kenneth Wiberg from the early 90s in J. comp. chem. about the comparison of several atomic-charge partitioning scheme including Bader's you may find useful. You may also consider using the Natural atomic charges method if you want faster and less 'surprising' results. Wai-To Chan -= This is automatically added to each message by mailing script =- To send e-mail to subscribers of CCL put the string CCL: on your Subject: line and send your message to: CHEMISTRY{at}ccl.net Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST{at}ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs If your mail is bouncing from CCL.NET domain send it to the maintainer: Jan Labanowski, jkl{at}ccl.net (read about it on CCL Home Page) -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ------_=_NextPart_001_01C325E1.71401400 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: calculating atomic charges with AIMPAC

But a charge of -20 almost certainly is the result of = assigning the electrons of one atom's basin to another atom.  = Something which occurs as a result of the program failing to find the = boundary between them. 

At any rate, I would try a 2nd AIM calculation = program if one is available, before giving up.  E.g., I often find = Gaussian's AIM code halts with a (usually cryptic) error message, but = for the same molecule AIM2000 may complete the calculation with no = problem.



-----Original Message-----
From: Wai-To Chan [mailto:chan{at}curl.gkcl.yorku.ca]
Sent: Thursday, May 29, 2003 2:09 AM
To: chemistry{at}ccl.net
Subject: CCL:calculating atomic charges with = AIMPAC


<<<<<<<<<<<<<<<<= ;<<<<<<<<<<<<<<<<<
I currently have an undergraduate student who is = attempting to compare
charges of methylated benzene and ferrocene systems = using the AIMPAC
software.  In the case of hexamethyl benzene = and the ferrocene analogues he
is getting very large negative charges on some the = ring carbons (-5 to -20)
and small positive charges on iron (0.5) Can anyone = tell me if there is a
problem with the program or if the ring system can = not be treated with this
method?
>>>>>>>>>>>>>>>>= ;>>>>>>>>>>>>>>>


   I remember reading about the tendency of = AIM changes to be larger in magnitude
than expected on the basis of chemical intuition. = There is a paper by
Kenneth Wiberg from the early 90s in J. comp. chem. = about the comparison of
several atomic-charge partitioning scheme including = Bader's you may find useful.
You may also consider using the Natural atomic = charges method if you want
faster and less 'surprising' results.

Wai-To Chan



-=3D This is automatically added to each message by = mailing script =3D-
To send e-mail to subscribers of CCL put the string = CCL: on your Subject: line
and send your message to:  = CHEMISTRY{at}ccl.net

Send your subscription/unsubscription requests to: = CHEMISTRY-REQUEST{at}ccl.net
HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs

If your mail is bouncing from CCL.NET domain send it = to the maintainer:
Jan Labanowski,  jkl{at}ccl.net (read about it on = CCL Home Page)
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-= +-+-+-+-+




------_=_NextPart_001_01C325E1.71401400-- ------=_NextPartTM-000-2d4cf94c-77cb-47a7-bd49-dee6013c6e1a-- From chemistry-request@ccl.net Thu May 29 08:46:35 2003 Received: from srvus010.unitedcatalysts.com (srvus010.unitedcatalysts.com [208.23.162.8] (may be forged)) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TCkZgC005459 for ; Thu, 29 May 2003 08:46:35 -0400 Received: by lvlxch02.ntdomain.americas.sc-world.com with Internet Mail Service (5.5.2653.19) id ; Thu, 29 May 2003 08:44:19 -0400 Message-ID: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF0{at}lvlxch02.ntdomain.americas.sc-world.com> From: "Shobe, Dave" To: "'Renxiao Wang'" , chemistry{at}ccl.net Subject: RE: A question on heat of formation Date: Thu, 29 May 2003 08:44:17 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-45783012-3329-4d2f-8144-10177a06720f" This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------=_NextPartTM-000-45783012-3329-4d2f-8144-10177a06720f Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C325E0.04B18270" ------_=_NextPart_001_01C325E0.04B18270 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Most people do the following. They find the lowest-energy = conformations of the molecules involved, and use the energy at those geometries to = calculate delta-H. The computational chemist should of course use any prudent, reasonable means to ensure that these are indeed the *global* = conformational minima. However... I think to really "do it right" one has to calculate all the low-lying conformations and determine the relative statistical-mechanical = populations of each conformation, and from that the average enthalpy per molecule = (or per mole). =20 Usually there are enough other sources of error that the correct = procedure has little advantage over the first approach, at least for enthalpy. = But the corresponding "correct" procedure is often done for entropy, where population of low-lying conformational isomers can make a significant difference in the outcome. --David Shobe S=FCd-Chemie Inc. phone (502) 634-7409 fax (502) 634-7724 email dshobe{at}sud-chemieinc.com Don't bother flaming me: I'm behind a firewall. -----Original Message----- From: Renxiao Wang [mailto:renxiao{at}med.umich.edu] Sent: Wednesday, May 28, 2003 8:19 PM To: chemistry{at}ccl.net Subject: CCL:A question on heat of formation Hi, CCLers! Should the heat of formations of two enantiomers of a chiral compound = be identical? I think the answer is yes. Please point out some references for this.=20 A disturbing fact is that, no matter you use QM or MM to calculate the heat of formation for a given molecule, the number is conformation-depentdent. I guess this is because of the steric energy part is conformation-dependent. So, if I really want to compare the stability of two molecules using heat of formation, how may I calculate it right? Best regards, Renxiao Wang renxiao{at}med.umich.edu -=3D This is automatically added to each message by mailing script =3D- To send e-mail to subscribers of CCL put the string CCL: on your = Subject: line and send your message to: CHEMISTRY{at}ccl.net Send your subscription/unsubscription requests to: = CHEMISTRY-REQUEST{at}ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs=20 If your mail is bouncing from CCL.NET domain send it to the maintainer: Jan Labanowski, jkl{at}ccl.net (read about it on CCL Home Page) -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+= ------_=_NextPart_001_01C325E0.04B18270 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: A question on heat of formation

Most people do the following.  They find the = lowest-energy conformations of the molecules involved, and use the = energy at those geometries to calculate delta-H.  The = computational chemist should of course use any prudent, reasonable = means to ensure that these are indeed the *global* conformational = minima.

However...

I think to really "do it right" one has to = calculate all the low-lying conformations and determine the relative = statistical-mechanical populations of each conformation, and from that = the average enthalpy per molecule (or per mole). 

Usually there are enough other sources of error that = the correct procedure has little advantage over the first approach, at = least for enthalpy.  But the corresponding "correct" = procedure is often done for entropy, where population of low-lying = conformational isomers can make a significant difference in the = outcome.

--David Shobe
S=FCd-Chemie Inc.
phone (502) 634-7409
fax     (502) 634-7724
email  dshobe{at}sud-chemieinc.com

Don't bother flaming me: I'm behind a = firewall.




-----Original Message-----
From: Renxiao Wang [mailto:renxiao{at}med.umich.edu]<= /FONT>
Sent: Wednesday, May 28, 2003 8:19 PM
To: chemistry{at}ccl.net
Subject: CCL:A question on heat of formation


Hi, CCLers!

Should the heat of formations of two enantiomers of a = chiral compound be
identical? I think the answer is yes. Please point = out some references
for this.

A disturbing fact is that, no matter you use QM or MM = to calculate the
heat of formation for a given molecule, the number = is
conformation-depentdent. I guess this is because of = the steric energy
part is conformation-dependent. So, if I really want = to compare the
stability of two molecules using heat of formation, = how may I calculate
it right?

Best regards,

Renxiao Wang
renxiao{at}med.umich.edu



-=3D This is automatically added to each message by = mailing script =3D-
To send e-mail to subscribers of CCL put the string = CCL: on your Subject: line
and send your message to:  = CHEMISTRY{at}ccl.net

Send your subscription/unsubscription requests to: = CHEMISTRY-REQUEST{at}ccl.net
HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs

If your mail is bouncing from CCL.NET domain send it = to the maintainer:
Jan Labanowski,  jkl{at}ccl.net (read about it on = CCL Home Page)
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-= +-+-+-+-+




------_=_NextPart_001_01C325E0.04B18270-- ------=_NextPartTM-000-45783012-3329-4d2f-8144-10177a06720f-- From chemistry-request@ccl.net Thu May 29 09:45:33 2003 Received: from mailgate.tripos.com (mailgate.tripos.com [192.160.145.7]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TDjXgC007456 for ; Thu, 29 May 2003 09:45:33 -0400 Message-Id: <200305291345.h4TDjXgC007456{at}server.ccl.net> From: "Paul Hawkins" To: "Noriaki Ikuta" Cc: Date: Thu, 29 May 2003 09:44:44 -0400 Subject: RE: ic50 values in dtp aids dataset MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit In-Reply-To: <20030529104951.55845.qmail{at}web20513.mail.yahoo.com> X-Verify: Valid sender address in headers Noriaki, I suspect the statement "concentration of compound that inhibits the growth of uninfected cells by 50%" is a typo. IC50 is the concentration of a compound required to reduce the readout of the assay by 50%, whether you are measuring enzyme activity, receptor activation or cell growth/division. The classification into the different activity classes is somewhat arbitrary. Generally speaking a compound having an IC50 greater than 10^-5 is not considered active as the doses required to achieve those concentrations in an animal (let alone a human) would be huge. For comparison most anti-HIV drugs on the market now have IC50's in this sort of assay around 10^-6 and have IC50's against their target receptor (the protease or reverse transcriptase mostly) in the 10^-9 range. There may also be a known correlation between IC50's in this assay and doses required to cause remission of infection in an animal model of HIV in humans and again the maximum acceptable dose may translate to an IC50 in this assay of 10^-5. I am unaware of any papers on this dataset in particular though a good many studies have used this dataset for various purposes. One word of warning Log10HiConc should always be around 10x greater than the IC50 so as the Log10HiConc approaches the reported IC50 the IC50 result becomes less and less reliable. Paul Hawkins Applications Scientist Tripos Inc 1699 South Hanley Road St. Louis MO 63144 Ph: 617-899-4151 E-mail: phawkins{at}tripos.com www.tripos.com -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request{at}ccl.net]On Behalf Of Noriaki Ikuta Sent: Thursday, May 29, 2003 6:50 AM To: chemistry{at}ccl.net Cc: noriaki_ikuta{at}yahoo.com Subject: CCL:ic50 values in dtp aids dataset I have three questions regarding the DTP AIDS Antiviral Screening dataset. http://dtp.nci.nih.gov/docs/aids/aids_data.html i) The explanation for the IC50 is "concentration of compound that inhibits the growth of uninfected cells by 50%" I do not understand this. I throught it would be of interest to inhibit the virus or at least the growth of infected cells? ii) How is the conclusion active/inactive drawn from those IC50 values? I have included a few example entries with ConcUnit = M in the mail. HiConc - the highest concentration used in the dose response Flag - when different experiments are averaged, "=" indicated that all the experiments reached IC50 ">" indicates that at least one of the experiments don't actually reach an EC50 and the highest tested concentration is used for the value of IC50 IC50 - concentration of compound that inhibits the growth of uninfected cells by 50% Conclusion - CA active, CM moderately active, CI inactive NSC Log10HiConc Flag Log10IC50 Conclusion 646607 -5.4 > -5.42 CA 646607 -4.5 = -5.29 CA 646607 -3.7 = -5.79 CA 646609 -4.5 = -5.78 CA 646609 -3.7 = -5.61 CA 646610 -4.5 = -5.77 CA 646610 -3.7 = -5.88 CA 639762 -3.7 = -4.75 CM 639762 -3.6 = -4.88 CM 639763 -4.4 = -4.63 CM 639765 -3.9 = -4.69 CM 639766 -4.6 > -4.79 CM 626162 -3.7 < -7.2 CI 629266 -3.7 < -7.2 CI 630958 -3.9 = -4.74 CI 630959 -3.2 = -3.85 CI 639915 -4.7 > -4.7 CI 639975 -3.9 > -5.05 CI 640064 -5 > -5 CI 640064 -3.7 > -3.7 CI iii) Are there papers available explaining this dataset in depth? Noriaki __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com -= This is automatically added to each message by mailing script =- To send e-mail to subscribers of CCL put the string CCL: on your Subject: line and send your message to: CHEMISTRY{at}ccl.net Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST{at}ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs If your mail is bouncing from CCL.NET domain send it to the maintainer: Jan Labanowski, jkl{at}ccl.net (read about it on CCL Home Page) -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ From chemistry-request@ccl.net Thu May 29 09:57:54 2003 Received: from srvus010.unitedcatalysts.com (srvus010.unitedcatalysts.com [208.23.162.8] (may be forged)) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TDvrgC007809 for ; Thu, 29 May 2003 09:57:54 -0400 Received: by lvlxch02.ntdomain.americas.sc-world.com with Internet Mail Service (5.5.2653.19) id ; Thu, 29 May 2003 09:55:48 -0400 Message-ID: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF5{at}lvlxch02.ntdomain.americas.sc-world.com> From: "Shobe, Dave" To: "'Daniel R. Rohr'" , ccl Subject: RE: Eigenvalues of hessian Date: Thu, 29 May 2003 09:55:48 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-9290d070-73cb-4de6-80e7-a42000a4c926" This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------=_NextPartTM-000-9290d070-73cb-4de6-80e7-a42000a4c926 Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C325EA.0207DAB0" ------_=_NextPart_001_01C325EA.0207DAB0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable What I said earlier was made on the assumption of being at a = stationalry point. =20 I have always heard that a freq calculation is unreliable at = non-stationary points, but I don't remember (and perhaps I never knew) the reason for = this warning. =20 --David Shobe S=FCd-Chemie Inc. phone (502) 634-7409 fax (502) 634-7724 email dshobe{at}sud-chemieinc.com Don't bother flaming me: I'm behind a firewall. -----Original Message----- From: Daniel R. Rohr [mailto:rohrd{at}students.uni-marburg.de] Sent: Thursday, May 29, 2003 5:54 AM To: ccl Subject: CCL:Eigenvalues of hessian Hi CCLer Maybe I should clearify my question: I do know that in the Hessians calculated, there is usually translation and rotation. I want to avoid speaking of frequencies, because we are talking about Hessians which are not necessarily calculated at stationary points. That#s why I talk of the eigenvalues. In an excat method there should be exactly 3N-6 nonzeroeigenvalues and = 6 exact zeros. ANY method I tested (HF, DFT and MP2) with ANY Program I tested (Gaussian, Turbomole, Jaguar) gives me 3N nonzero Eigenvalues. The question is now. Why are there too many nonzero Eigenvalues? In my eyes there are two possible reasons: 1. The method is ok but there are errors made in the derivations of the method. Maybe neglecting terms in the differentiation. Or maybe asumptions that hold true only for stationary points. 2. The method is not ok. This means that I do get different Energies = for different orientations of my molecule (As is the case in DFT) Of course it is also possible that there is a mixture of both. I would like to know which error is the bigger one? And maybe get a reference for this question. Another point which I do not understand: Some answers state that there is only little translation and rotation since the frequencies are close to zero. Is the absolute value of an eigenvalue a measure for the quantity (whaterver that means) of its eigenvector???? And if that is the case why is there "more" translation and rotation at nonstationary points? I get "frequencies" (this is eigenvalues of the Hessian) between 400 and 850 cm^-1 for Water with HF 6-31G** (Gaussian calculation): Low frequencies --- -410.8545 -0.0007 -0.0006 0.0017 489.4045 853.6142 Low frequencies --- 1891.8502 3808.4304 3946.8117 These are of course the frequencies calculated BEFORE projection of translation and rotation. I hope you can help me on that Daniel Rohr Hi Daniel, Your question is a bit tricky in that the answer depends on the method you are using. =20 If you are calculating Hartree-Fock or MP2 hessians then I think you should get 6 zero eigenvalues and 3N-6 non-zero eigenvalues at a stationary point. Even if the hessian is obtained by numerical differentiation of gradients this should come out allright, otherwise the code is broken. If you are doing a DFT calculation life typically gets a bit more complicated. If the DFT hessian is implemented as a proper second derivative of the energy expression with respect to the nuclear coordinates you should still get 6 zero eigenvalues. However this requires that the code takes the 2nd derivatives of the Becke weights (A.D. Becke, J.Chem.Phys. 88 = (1988) 2547-2553) into account as well. Some codes don't bother calculating these terms in which case the strict translational invariance of the hessian is violated. This typically expresses itself in 3 additional non-zero but small valued eigenvalues.=20 If you are calculating the vibrational modes by numerical differentiation of DFT gradients then there may still be a problem if your molecule has symmetry. Many quantum chemistry packages will reorientate your = molecule such that the main axes of ineratia coincide with the Cartesian axes of your coordinate system. This means that different points in the numerical differentiation algorithm may be calculated at different orientations. However, the DFT quadratures are not always invariant for rotations. = For example if you want to do calculations on H2 and you are going to use 6 points per H atom all placed on the Cartesian axes than the energy you'll get for the following two orientations will not be the same even if the bond distance is: H H | \ H H Again this will lead to additional non-zero but small eigenvalues. In this case it may help to switch the use of symmetry off, so that all points are calculated in exactly the same orientation.=20 As you have given very little details about exactly which calculation you are doing I can not say for sure which of the above reasons is the most like cause. However I think it is most likely one of these and with some experimentation and a good look at the documentation for your program you should be able to find out what is happening. Best wishes, Huub van Dam -----Original Message----- From: Daniel R. Rohr To: ccl Sent: 5/28/03 3:20 PM Subject: CCL:Eigenvalues of hessian Hi CCLers Does anybody of you know, why there are more than 3N-6 nonzero eigenvalues of the hessian? I am especially interested in the question, whether this error is due to the errors made by differentiation or whether they are due to an error within the method itself. I usually = get only 3 zero eigenvalues. No matter if the hessian is calculated on a stationary point or not. Thanks for your help Daniel Rohr --=20 ----------------------------------------------------------------------- Daniel Rohr Fachbereich Chemie der Philipps-Universitdt Tel.: +49-6421-28-25686 Hans-Meerwein-Stra_e Fax.: +49-6421-28-25566 D-35043 Marburg eMail: rohrd{at}stud-mailer.uni-marburg.de Germany ----------------------------------------------------------------------- To send e-mail to subscribers of CCL put the string CCL: on your Subject: line Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST{at}ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs --=20 ----------------------------------------------------------------------- Daniel Rohr Fachbereich Chemie der Philipps-Universitdt Tel.: +49-6421-28-25686 Hans-Meerwein-Stra_e Fax.: +49-6421-28-25566 D-35043 Marburg eMail: rohrd{at}stud-mailer.uni-marburg.de Germany ----------------------------------------------------------------------- -=3D This is automatically added to each message by mailing script =3D- To send e-mail to subscribers of CCL put the string CCL: on your = Subject: line and send your message to: CHEMISTRY{at}ccl.net Send your subscription/unsubscription requests to: = CHEMISTRY-REQUEST{at}ccl.net HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs=20 If your mail is bouncing from CCL.NET domain send it to the maintainer: Jan Labanowski, jkl{at}ccl.net (read about it on CCL Home Page) -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+= ------_=_NextPart_001_01C325EA.0207DAB0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: Eigenvalues of hessian

What I said earlier was made on the assumption of = being at a stationalry point. 

I have always heard that a freq calculation is = unreliable at non-stationary points, but I don't remember (and perhaps = I never knew) the reason for this warning. 

--David Shobe
S=FCd-Chemie Inc.
phone (502) 634-7409
fax     (502) 634-7724
email  dshobe{at}sud-chemieinc.com

Don't bother flaming me: I'm behind a = firewall.



-----Original Message-----
From: Daniel R. Rohr [mailto:rohrd{at}students.uni-= marburg.de]
Sent: Thursday, May 29, 2003 5:54 AM
To: ccl
Subject: CCL:Eigenvalues of hessian


Hi CCLer

 Maybe I should clearify my question:
I do know that in the Hessians calculated, there is = usually translation
and rotation. I want to avoid speaking of = frequencies, because we are
talking about Hessians which are not necessarily = calculated at
stationary points. That#s why I talk of the = eigenvalues.
In an excat method there should be exactly 3N-6 = nonzeroeigenvalues and 6
exact zeros. ANY method I tested (HF, DFT and MP2) = with ANY Program I
tested (Gaussian, Turbomole, Jaguar) gives me 3N = nonzero Eigenvalues.
The question is now. Why are there too many nonzero = Eigenvalues? In my
eyes there are two possible reasons:

1. The method is ok but there are errors made in the = derivations of the
method. Maybe neglecting terms in the = differentiation. Or maybe
asumptions that hold true only for stationary = points.

2. The method is not ok. This means that I do get = different Energies for
different orientations of my molecule (As is the = case in DFT)

Of course it is also possible that there is a mixture = of both. I would
like to know which error is the bigger one? And = maybe get a reference
for this question.

Another point which I do not understand:
Some answers state that there is only little = translation and rotation
since the frequencies are close to zero. Is the = absolute value of an
eigenvalue a measure for the quantity (whaterver = that means) of its
eigenvector???? And if that is the case why is there = "more" translation
and rotation at nonstationary points? I get = "frequencies" (this is
eigenvalues of the Hessian) between 400 and 850 = cm^-1 for Water with HF
6-31G** (Gaussian calculation):

Low frequencies --- -410.8545   = -0.0007   -0.0006    0.0017  = 489.4045
853.6142
 Low frequencies --- 1891.8502 3808.4304 = 3946.8117

These are of course the frequencies calculated BEFORE = projection of
translation and rotation.

I hope you can help me on that

Daniel Rohr


Hi Daniel,

Your question is a bit tricky in that the answer = depends on the method
you
are using.
 
If you are calculating Hartree-Fock or MP2 hessians = then I think you
should
get 6 zero eigenvalues and 3N-6 non-zero eigenvalues = at a stationary
point.
Even if the hessian is obtained by numerical = differentiation of
gradients
this should come out allright, otherwise the code is = broken.

If you are doing a DFT calculation life typically = gets a bit more
complicated. If the DFT hessian is implemented as a = proper second
derivative
of the energy expression with respect to the nuclear = coordinates you
should
still get 6 zero eigenvalues. However this requires = that the code takes
the
2nd derivatives of the Becke weights (A.D. Becke, = J.Chem.Phys. 88 (1988)
2547-2553) into account as well. Some codes don't = bother calculating
these
terms in which case the strict translational = invariance of the hessian
is
violated. This typically expresses itself in 3 = additional non-zero but
small
valued eigenvalues.

If you are calculating the vibrational modes by = numerical
differentiation of
DFT gradients then there may still be a problem if = your molecule has
symmetry. Many quantum chemistry packages will = reorientate your molecule
such that the main axes of ineratia coincide with = the Cartesian axes of
your
coordinate system. This means that different points = in the numerical
differentiation algorithm may be calculated at = different orientations.
However, the DFT quadratures are not always = invariant for rotations. For
example if you want to do calculations on H2 and you = are going to use 6
points per H atom all placed on the Cartesian axes = than the energy
you'll
get for the following two orientations will not be = the same even if the
bond
distance is:

    H  H
    |   \
    H    H

Again this will lead to additional non-zero but small = eigenvalues. In
this
case it may help to switch the use of symmetry off, = so that all points
are
calculated in exactly the same orientation.

As you have given very little details about exactly = which calculation
you
are doing I can not say for sure which of the above = reasons is the most
like
cause. However I think it is most likely one of = these and with some
experimentation and a good look at the documentation = for your program
you
should be able to find out what is happening.

Best wishes,

   Huub van Dam

-----Original Message-----
From: Daniel R. Rohr
To: ccl
Sent: 5/28/03 3:20 PM
Subject: CCL:Eigenvalues of hessian

Hi CCLers

 Does anybody of you know, why there are more = than 3N-6 nonzero
eigenvalues of the hessian? I am especially = interested in the question,
whether this error is due to the errors made by = differentiation or
whether they are due to an error within the method = itself. I usually get
only 3 zero eigenvalues. No matter if the hessian is = calculated on a
stationary point or not.

Thanks for your help

Daniel Rohr

--
---------------------------------------------------------------= --------
 Daniel = Rohr           &n= bsp;           &n= bsp;            = Fachbereich Chemie der
          &nb= sp;           &nb= sp;           &nb= sp;           &nb= sp; Philipps-Universitdt
 Tel.: = +49-6421-28-25686         &= nbsp;           &= nbsp;  Hans-Meerwein-Stra_e
 Fax.: = +49-6421-28-25566         &= nbsp;           &= nbsp;  D-35043 Marburg
 eMail: = rohrd{at}stud-mailer.uni-marburg.de      &nbs= p; Germany
---------------------------------------------------------------= --------




To send e-mail to subscribers of CCL put the string = CCL: on your
Subject: line

Send your subscription/unsubscription requests = to:
CHEMISTRY-REQUEST{at}ccl.net
HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs

--
---------------------------------------------------------------= --------
 Daniel = Rohr           &n= bsp;           &n= bsp;            = Fachbereich Chemie der
          &nb= sp;           &nb= sp;           &nb= sp;           &nb= sp; Philipps-Universitdt
 Tel.: = +49-6421-28-25686         &= nbsp;           &= nbsp;  Hans-Meerwein-Stra_e
 Fax.: +49-6421-28-25566    =             =         D-35043 Marburg
 eMail: = rohrd{at}stud-mailer.uni-marburg.de      &nbs= p; Germany
---------------------------------------------------------------= --------




-=3D This is automatically added to each message by = mailing script =3D-
To send e-mail to subscribers of CCL put the string = CCL: on your Subject: line
and send your message to:  = CHEMISTRY{at}ccl.net

Send your subscription/unsubscription requests to: = CHEMISTRY-REQUEST{at}ccl.net
HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs

If your mail is bouncing from CCL.NET domain send it = to the maintainer:
Jan Labanowski,  jkl{at}ccl.net (read about it on = CCL Home Page)
-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-= +-+-+-+-+




------_=_NextPart_001_01C325EA.0207DAB0-- ------=_NextPartTM-000-9290d070-73cb-4de6-80e7-a42000a4c926-- From chemistry-request@ccl.net Thu May 29 10:15:03 2003 Received: from mx3.trentu.ca (mx3.trentu.ca [192.75.12.4]) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TEF3gC008066 for ; Thu, 29 May 2003 10:15:03 -0400 Received: (qmail 1035 invoked by uid 504); 29 May 2003 14:15:02 -0000 Received: from elewars{at}trentu.ca by mx3.trentu.ca by uid 501 with qmail-scanner-1.14 (fsecure: 4.15/4370/2003-05-28/2003-05-27. 2003-05-27/ Clear:. Processed in 0.366398 secs); 29 May 2003 14:15:02 -0000 X-Qmail-Scanner-Mail-From: elewars{at}trentu.ca via mx3.trentu.ca X-Qmail-Scanner: 1.14 (Clear:. Processed in 0.366398 secs) Received: from unknown (HELO trentu.ca) (209.42.101.30) by mx3.trentu.ca with SMTP; 29 May 2003 14:15:01 -0000 Message-ID: <3ED616A7.F1006CBA{at}trentu.ca> Date: Thu, 29 May 2003 10:18:15 -0400 From: elewars X-Mailer: Mozilla 4.79 [en] (WinNT; U) X-Accept-Language: en MIME-Version: 1.0 To: Leif Laaksonen , chemistry{at}ccl.net Subject: Re: CCL:Orbitals References: <003801c32546$68145fd0$bb1c7b0a@ISOSIRRI> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit 2003 May 29 Hello, So you are saying that there is a one-to-one correspondence between physically real things X and QM operators X^ , the average value of the thing X for a system being the integral of the system's wavefunction over the operator. So orbitals are real only if there is, mathematically, an orbital operator PSI^--which there is not (?). Good point. EL ====== Leif Laaksonen wrote: > Hi, > > I might remember wrong from my primary courses in quantum chemistry > that only obsrvables are the expectation values from the wave > function! > > This means that while the wave function (|P>) as such is not > observable the expectation value of property X is (). > > Regards, > > -leif > > -----Original Message----- > From: Computational Chemistry List [mailto:chemistry-request{at}ccl.net] On Behalf Of Guosheng Wu > Sent: Wednesday, May 28, 2003 7:46 PM > To: chemistry{at}ccl.net > Subject: CCL:Orbitals > > > > > --- elewars wrote: > > > > > 2003 May 28 > > > > > > > > > > Are MOs physically real? This is a meaningful question only if there is some > > > > > experiment or observation that could provide an answer _yes_ or _no_. Is there, > > > > > at least in principle, such an experiment or observation? > > > > Guosheng Wu wrote: > > > > > > > MO is just a useful mathematical tool for quantum chemistry/physics, and > > > > it seems to me itself does not have much physical meaning. > > > > > > > > However, many properties from MOs are meaningful and can be observed > > > > by exmperiments: such as the square of absolute MOs (|MO|^2) corresponding > > > > to the electronic density. Things like |MO|^2 of d-orbitals were published > > > > in a Science article last one or a few years. > > > --- elewars wrote: > > > Hello, > > > > > > Thanks for your ideas. If photoel. spectra (and Koopmans theorem--in this case the > > > second MO is at the ionization limit) really measure the energy difference of two MOs, > > > then MOs exist! > > > > > > In any case, it is an interesting discussion. > > > > > > EL > > > ===== > > The properties from |MO|^2 exist does not really mean MO exists. > > When we know some properties from x^2 exist and can be measured, > this does not mean x exists and can be measured. > > For example, what does the complex number(i) mean in the real world? > Measurable? Not really, but obviously it's a useful mathematical tool > for lots of applications. Certainly (i^2=-1) means something. Right? > > Related issue for absolute energy: energy difference is measurable > by many kind of experiments, but not for the absolute energy. > > This may lead to some kind of philosophical problem, if one has > such a taste. > > Guosheng > > __________________________________ > Do you Yahoo!? > Yahoo! Calendar - Free online calendar with sync to Outlook(TM). > http://calendar.yahoo.com > > -= This is automatically added to each message by mailing script =- > To send e-mail to subscribers of CCL put the string CCL: on your Subject: line > and send your message to: CHEMISTRY{at}ccl.net > > Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST{at}ccl.net > HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs > > If your mail is bouncing from CCL.NET domain send it to the maintainer: > Jan Labanowski, jkl{at}ccl.net (read about it on CCL Home Page) > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ From chemistry-request@ccl.net Thu May 29 11:29:19 2003 Received: from red.CACheSoftware.com ([12.17.172.66]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TFTIgC010027 for ; Thu, 29 May 2003 11:29:19 -0400 Received: from dgallagher.cachesoftware.com ([192.168.110.100]) by red.CACheSoftware.com (8.11.6/8.9.3/3) with ESMTP id h4TFTHI32036; Thu, 29 May 2003 08:29:17 -0700 Message-Id: <5.2.0.9.0.20030529081742.039f0630:at:mail.cachesoftware.com> X-Sender: daveg:at:mail.cachesoftware.com (Unverified) X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9 Date: Thu, 29 May 2003 08:28:26 -0700 To: Philip Stortz From: David Gallagher Subject: Re: CCL:Crystal building tools Cc: chemistry:at:ccl.net In-Reply-To: <3ED5A324.B8CC04AE:at:wyoming.com> References: <000001c325a1$4a9384d0$0300a8c0:at:attbi.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Hi Phil, For building crystals there is a special crystal builder in CAChe complete with all 230 space groups. Indeed, the drawing tool was never intended to build crystal structures. The crystal builder, which was recently updated to handle larger systems, requires only the space group, cell parameters, and fractional coordinates, then you can specify as many unit cells as you wish. If you send us your crystal data, we'll build the crystal for you and send it back in the format you specify. David Gallagher, Fujitsu. At 12:05 AM 5/29/2003 -0600, Philip Stortz wrote: >To help with modeling electronic spectra of crystals, is there a good >package out there to build the initial crystal with at least a couple >dozen unit cells if i know the crystal structure? I've tried this with >Cache a few years ago, and the drawing tool was not very useful and very >painful for building 3d crystals, bonds often wound up in odd places where >i didn't mean to put them and it was very hard to view the bonds to see >these problems. Thanks again. > >-- >Philip Stortz -- To be nobody but yourself when the whole world is trying >its best night and day to make you everybody else is to fight the hardest >battle any human being will ever fight. -- E.E. Cummings > > >-= This is automatically added to each message by mailing script =- >To send e-mail to subscribers of CCL put the string CCL: on your Subject: line >and send your message to: CHEMISTRY:at:ccl.net > >Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST:at:ccl.net >HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs > >If your mail is bouncing from CCL.NET domain send it to the maintainer: >Jan Labanowski, jkl:at:ccl.net (read about it on CCL Home Page) >-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ============================================== David A. Gallagher, C. Chem. Vice President, Sales and Marketing CAChe Group Fujitsu America, Inc. 15244 NW Greenbrier Parkway Beaverton, OR 97006-5733, USA Direct dial: 503 746 3607 Main Switchboard: 503 746 3600 Fax: 503 531 9966 email: dgallagher:at:cachesoftware.com http://www.cachesoftware.com http://www.cachesoftware.com/staff/dg.shtml ============================================== From chemistry-request@ccl.net Thu May 29 11:16:47 2003 Received: from alcor.concordia.ca (alcor.Concordia.CA [132.205.7.51]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TFGlgC009657 for ; Thu, 29 May 2003 11:16:47 -0400 Received: from zeta (Xenon.Concordia.CA [132.205.24.92]) by alcor.concordia.ca (8.12.9/8.12.9) with ESMTP id h4TFGahC020588 for ; Thu, 29 May 2003 11:16:36 -0400 (EDT) Date: Thu, 29 May 2003 11:16:39 -0400 From: "Qadir K.Timerghazin" X-Mailer: The Bat! (v1.53d) Reply-To: "Qadir K.Timerghazin" Organization: =?ISO-8859-1?B?Q2VudHJlIGRlIFJlY2hlcmNoZSBlbiBNb2TpbGlzYXRpb24gTW9s6WN1?= =?ISO-8859-1?B?bGFpcmU=?= X-Priority: 3 (Normal) Message-ID: <96266648750.20030529111639:at:cermm.concordia.ca> To: "'CCL'" Subject: Re: CCL:calculating atomic charges with AIMPAC In-Reply-To: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF1:at:lvlxch02.ntdomain.americas.sc-world.com> References: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF1:at:lvlxch02.ntdomain.americas.sc-world.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit X-Scanned-By: MIMEDefang 2.26 (www . roaringpenguin . com / mimedefang) Which integration algorithm are you using? PROAIM is much faster than PROMEGA, but if something goes wrong, I recommend to compare the results with PROMEGA. And, of course, always check if your integrated L values are small enough (at least 10e-3) -- Qadir mailto:Qadir.Timerghazin:at:CERMM.Concordia.CA SD> But a charge of -20 almost certainly is the result of assigning the SD> electrons of one atom's basin to another atom. Something which occurs as a SD> result of the program failing to find the boundary between them. SD> At any rate, I would try a 2nd AIM calculation program if one is available, SD> before giving up. E.g., I often find Gaussian's AIM code halts with a SD> (usually cryptic) error message, but for the same molecule AIM2000 may SD> complete the calculation with no problem. SD> -----Original Message----- SD> From: Wai-To Chan [mailto:chan:at:curl.gkcl.yorku.ca] SD> Sent: Thursday, May 29, 2003 2:09 AM SD> To: chemistry:at:ccl.net SD> Subject: CCL:calculating atomic charges with AIMPAC SD> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< SD> I currently have an undergraduate student who is attempting to compare SD> charges of methylated benzene and ferrocene systems using the AIMPAC SD> software. In the case of hexamethyl benzene and the ferrocene analogues he SD> is getting very large negative charges on some the ring carbons (-5 to -20) SD> and small positive charges on iron (0.5) Can anyone tell me if there is a SD> problem with the program or if the ring system can not be treated with this SD> method? >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> SD> I remember reading about the tendency of AIM changes to be larger in SD> magnitude SD> than expected on the basis of chemical intuition. There is a paper by SD> Kenneth Wiberg from the early 90s in J. comp. chem. about the comparison of SD> several atomic-charge partitioning scheme including Bader's you may find SD> useful. SD> You may also consider using the Natural atomic charges method if you want SD> faster and less 'surprising' results. SD> Wai-To Chan SD> To send e-mail to subscribers of CCL put the string CCL: on your Subject: SD> line SD> Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST:at:ccl.net **************************************************************************** Qadir K. Timerghazin * Kadyr K. Timergazine Centre for Research * Centre de Recherche en in Molecular Modeling * Modilisation Moliculaire * Concordia University * Universiti Concordia 1455 De Maisonneuve Blvd St. West, * 1455 Bvd De Maisonneuve Ouest Montreal, Quebec, * Montrial, Quibec, CANADA H3G 1M8 * CANADA H3G 1M8 * Room: 1034-1 * Salle: 1034-1 Phone: (514) 848 3360 * Tel: (514) 848 3360 **************************************************************************** From chemistry-request@ccl.net Thu May 29 11:09:27 2003 Received: from smtp805.mail.sc5.yahoo.com (smtp805.mail.sc5.yahoo.com [66.163.168.184]) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TF9QgC009419 for ; Thu, 29 May 2003 11:09:26 -0400 Received: from cat207.cqe.northwestern.edu (HELO chem.northwestern.edu) (a-beim:at:sbcglobal.net@129.105.122.207 with plain) by smtp-sbc-v1.mail.vip.sc5.yahoo.com with SMTP; 29 May 2003 15:09:25 -0000 Date: Thu, 29 May 2003 10:09:27 -0500 Subject: Re: CCL:Electronic Spectra of inorganic crystals Content-Type: text/plain; charset=US-ASCII; format=flowed Mime-Version: 1.0 (Apple Message framework v552) Cc: chemistry:at:ccl.net To: Philip Stortz From: Geoff Hutchison In-Reply-To: <3ED5A212.9753567:at:wyoming.com> Message-Id: <8A678C09-91E7-11D7-BD27-000A95701B7A:at:chem.northwestern.edu> Content-Transfer-Encoding: 7bit X-Mailer: Apple Mail (2.552) On Thursday, May 29, 2003, at 01:00 AM, Philip Stortz wrote: > Time for my annual hopeful question, is there currently a program that > calculates electronic spectra of inorganic crystals (and their > surfaces) with any accuracy? I have little experience in these calculations, but do collaborate with a solid-state physics group that calculates semiconductor band structures all the time. The state-of-the-art AFAIK, are "screened-exchange" DFT band codes, or GW calculations. These produce band structues and optical properties of crystals and "slabs" at surfaces to a high degree of accuracy. (Hmm, I probably have papers on those somewhere around here, so please ask if you want more details. Or do a search on "Prof. Arthur Freeman" at Northwestern, and you'll turn up this group.) But you also said "trends" were OK, in which case, most DFT band structure codes should do what you want with no problem (e.g. DMol3, which I've used, FLAPW, or Wein2K, etc.) The band gap is usually given in eV directly from the calculation. Regards, -Geoff -- -Geoff Hutchison Ratner/Marks Groups (847) 491-3295 Northwestern Chemistry http://www.chem.northwestern.edu/ From chemistry-request@ccl.net Thu May 29 12:04:39 2003 Received: from srvus010.unitedcatalysts.com (srvus010.unitedcatalysts.com [208.23.162.8] (may be forged)) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TG4dgC011393 for ; Thu, 29 May 2003 12:04:39 -0400 Received: by lvlxch02.ntdomain.americas.sc-world.com with Internet Mail Service (5.5.2653.19) id ; Thu, 29 May 2003 12:02:27 -0400 Message-ID: <5CF08BBFE6DE97478C1E76E654CAF90801BC0BF7(at)lvlxch02.ntdomain.americas.sc-world.com> From: "Shobe, Dave" To: "'Guosheng Wu'" , "'Renxiao Wang'" , "'CCL'" Subject: RE: CCL:A question on heat of formation Date: Thu, 29 May 2003 12:02:26 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/mixed; boundary="----=_NextPartTM-000-901fc9be-3d98-4be5-8bcb-6563e19f507b" This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------=_NextPartTM-000-901fc9be-3d98-4be5-8bcb-6563e19f507b Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C325FB.B2E3ABF0" ------_=_NextPart_001_01C325FB.B2E3ABF0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable ---Guosheng Wu wrote: > It seems to me, dH has not much to do with propabilities of all conformations, > which is essentially entropy contribution.=20 There is a small effect on enthalpy, for temperatures above zero = Kelvin. Most people ignore it, since there are more important sources of error. = But I think the NIST/JANAF tables do take this effect into account. It may affect dH/dT more than it affects dH itself. --David Shobe S=FCd-Chemie Inc. phone (502) 634-7409 fax (502) 634-7724 email dshobe(at)sud-chemieinc.com Don't bother flaming me: I'm behind a firewall. -----Original Message----- From: Guosheng Wu [mailto:wu_guosheng2002(at)yahoo.com] Sent: Thursday, May 29, 2003 11:20 AM To: Shobe, Dave; 'Renxiao Wang' Subject: Re: CCL:A question on heat of formation Were you talking about free energy(dG), instead of heat of formation = (dH)? It seems to me, dH has not much to do with propabilities of all conformations, which is essentially entropy contribution.=20 BTW, John Pople and others had a few papers on Gn theory to get very = high quality heat of formation for a small molecule, by ab initio plus scaling strategy... but obviously not interested to most of the Med. Chem. community.=20 Guosheng --- "Shobe, Dave" wrote: > Most people do the following. They find the lowest-energy = conformations of > the molecules involved, and use the energy at those geometries to calculate > delta-H. The computational chemist should of course use any prudent, > reasonable means to ensure that these are indeed the *global* conformational > minima. >=20 > However... >=20 > I think to really "do it right" one has to calculate all the = low-lying > conformations and determine the relative statistical-mechanical populations > of each conformation, and from that the average enthalpy per molecule = (or > per mole). =20 >=20 > Usually there are enough other sources of error that the correct = procedure > has little advantage over the first approach, at least for enthalpy. = But > the corresponding "correct" procedure is often done for entropy, = where > population of low-lying conformational isomers can make a significant > difference in the outcome. >=20 > --David Shobe > S=FCd-Chemie Inc. > phone (502) 634-7409 > fax (502) 634-7724 > email dshobe(at)sud-chemieinc.com >=20 > Don't bother flaming me: I'm behind a firewall. >=20 >=20 >=20 >=20 > -----Original Message----- > From: Renxiao Wang [mailto:renxiao(at)med.umich.edu] > Sent: Wednesday, May 28, 2003 8:19 PM > To: chemistry(at)ccl.net > Subject: CCL:A question on heat of formation >=20 >=20 > Hi, CCLers! >=20 > Should the heat of formations of two enantiomers of a chiral compound = be > identical? I think the answer is yes. Please point out some = references > for this.=20 >=20 > A disturbing fact is that, no matter you use QM or MM to calculate = the > heat of formation for a given molecule, the number is > conformation-depentdent. I guess this is because of the steric energy > part is conformation-dependent. So, if I really want to compare the > stability of two molecules using heat of formation, how may I = calculate > it right? >=20 > Best regards, >=20 > Renxiao Wang > renxiao(at)med.umich.edu >=20 >=20 >=20 > -=3D This is automatically added to each message by mailing script = =3D- > To send e-mail to subscribers of CCL put the string CCL: on your = Subject: > line >=20 > Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST(at)ccl.net >=20 > HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs=20 >=20 >=20 >=20 >=20 >=20 >=20 >=20 __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com ------_=_NextPart_001_01C325FB.B2E3ABF0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: CCL:A question on heat of formation

---Guosheng Wu wrote:
> It seems to me, dH has not much to do with = propabilities of all conformations,
> which is essentially entropy contribution. =

There is a small effect on enthalpy, for temperatures = above zero Kelvin.  Most people ignore it, since there are more = important sources of error.  But I think the NIST/JANAF tables do = take this effect into account.  It may affect dH/dT more than it = affects dH itself.

--David Shobe
S=FCd-Chemie Inc.
phone (502) 634-7409
fax     (502) 634-7724
email  dshobe(at)sud-chemieinc.com

Don't bother flaming me: I'm behind a = firewall.



-----Original Message-----
From: Guosheng Wu [mailto:wu_guosheng2002@yahoo.c= om]
Sent: Thursday, May 29, 2003 11:20 AM
To: Shobe, Dave; 'Renxiao Wang'
Subject: Re: CCL:A question on heat of = formation


Were you talking about free energy(dG), instead of = heat of formation (dH)?

It seems to me, dH has not much to do with = propabilities of all conformations,
which is essentially entropy contribution.

BTW, John Pople and others had a few papers on Gn = theory to get very high quality
heat of formation for a small molecule, by ab initio = plus scaling strategy... but
obviously not interested to most of the Med. Chem. = community.

Guosheng

--- "Shobe, Dave" = <dshobe(at)sud-chemieinc.com> wrote:
> Most people do the following.  They find = the lowest-energy conformations of
> the molecules involved, and use the energy at = those geometries to calculate
> delta-H.  The computational chemist should = of course use any prudent,
> reasonable means to ensure that these are = indeed the *global* conformational
> minima.
>
> However...
>
> I think to really "do it right" one = has to calculate all the low-lying
> conformations and determine the relative = statistical-mechanical populations
> of each conformation, and from that the average = enthalpy per molecule (or
> per mole). 
>
> Usually there are enough other sources of error = that the correct procedure
> has little advantage over the first approach, = at least for enthalpy.  But
> the corresponding "correct" procedure = is often done for entropy, where
> population of low-lying conformational isomers = can make a significant
> difference in the outcome.
>
> --David Shobe
> S=FCd-Chemie Inc.
> phone (502) 634-7409
> fax     (502) = 634-7724
> email  dshobe(at)sud-chemieinc.com
>
> Don't bother flaming me: I'm behind a = firewall.
>
>
>
>
> -----Original Message-----
> From: Renxiao Wang [mailto:renxiao(at)med.umich.edu]<= /FONT>
> Sent: Wednesday, May 28, 2003 8:19 PM
> To: chemistry(at)ccl.net
> Subject: CCL:A question on heat of = formation
>
>
> Hi, CCLers!
>
> Should the heat of formations of two = enantiomers of a chiral compound be
> identical? I think the answer is yes. Please = point out some references
> for this.
>
> A disturbing fact is that, no matter you use QM = or MM to calculate the
> heat of formation for a given molecule, the = number is
> conformation-depentdent. I guess this is = because of the steric energy
> part is conformation-dependent. So, if I really = want to compare the
> stability of two molecules using heat of = formation, how may I calculate
> it right?
>
> Best regards,
>
> Renxiao Wang
> renxiao(at)med.umich.edu
>
>
>
> -=3D This is automatically added to each = message by mailing script =3D-
> To send e-mail to subscribers of CCL put the = string CCL: on your Subject:
> line
>
> Send your subscription/unsubscription requests = to: CHEMISTRY-REQUEST(at)ccl.net
>
> HOME Page: http://www.ccl.net   | Jobs Page: http://www.ccl.net/jobs
>
>
>
>
>
>
>


__________________________________
Do you Yahoo!?
Yahoo! Calendar - Free online calendar with sync to = Outlook(TM).
http://calendar.yahoo.com

------_=_NextPart_001_01C325FB.B2E3ABF0-- ------=_NextPartTM-000-901fc9be-3d98-4be5-8bcb-6563e19f507b-- From chemistry-request@ccl.net Thu May 29 11:49:20 2003 Received: from gip2.u-picardie.fr (gip2.u-picardie.fr [193.49.184.4]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TFnJgC010978 for ; Thu, 29 May 2003 11:49:20 -0400 Received: by gip2.u-picardie.fr (Postfix, from userid 33) id F379872C; Thu, 29 May 2003 17:49:14 +0200 (CEST) To: chemistry(at)ccl.net Subject: Optimization in Gamess and in Gaussian Message-ID: <1054223354.3ed62bfae75de(at)webmail.u-picardie.fr> Date: Thu, 29 May 2003 17:49:14 +0200 (CEST) From: FyD MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit User-Agent: IMP/PHP IMAP webmail program 2.2.6 Dear All, I am interested in better understanding the optimization procedure implemented in Gamess & Gaussian. In Gaussian, the optimization accuracy is controlled by the OPT keyword. Gaussian checks four parameters: the "Maximum Force", "RMS Force", "Maximum Displacement" and "RMS Displacement". From the numerous tests I ran using Gaussian, the "Maximum Displacement" and "RMS Displacement" thresholds are far MORE difficult to reach than the "Maximum Force" and "RMS Force" ones. Thus, after some optimization steps, the "Maximum Force" and "RMS Force" thresholds are usually satisfied. On the contrary, to satisfy the 2 other ones (related to the "Displacement"), this demands a certain numbers of additional optimization steps... - In GAMESS, the optimization is controlled by the OPTTOL keyword. The optimization is stopped when the "Maximum gradient" and the "RMS gradient" are below the demanded threshold value. It means (to me) that the "Maximum Displacement" and the "RMS Displacement" are NOT checked by GAMESS. Am I right ? - Since the two "Displacement" thresholds are the MOST difficult criterions to be satisfied in Gaussian and as they are NOT checked by GAMESS (from what I understand), what are the implications of such differences in the optimization accuracy in GAMESS ? Thank you very much, best regards, Francois From chemistry-request@ccl.net Thu May 29 13:00:37 2003 Received: from pump2.york.ac.uk (pump2.york.ac.uk [144.32.128.12]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TH0agC012514 for ; Thu, 29 May 2003 13:00:36 -0400 Received: from james.york.ac.uk (james.york.ac.uk [144.32.180.54]) by pump2.york.ac.uk (8.12.9/8.12.9) with ESMTP id h4TH0ZRw028571 for ; Thu, 29 May 2003 18:00:35 +0100 (BST) Message-Id: <5.2.1.1.0.20030529174915.00b17610|at|jm45.imap.york.ac.uk> X-Sender: jm45|at|jm45.imap.york.ac.uk X-Mailer: QUALCOMM Windows Eudora Version 5.2.1 Date: Thu, 29 May 2003 18:00:29 +0100 To: chemistry|at|ccl.net From: "Dr. Julia Metzker" Subject: calculating atomic charges with AIMPAC Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1"; format=flowed Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h4TH0bgC012515 In reading the responses to my previous enquiry I decided I should probably include some more information. The student working on this project has used G98 to generate the wavefunction file and is using the PROAIMV method of calculating the charges for methyl substituted benzenes and ferrocenes. He has used an all electron basis set in all of the calculation. He has had success in all cases for benzene except hexamethyl benzene where he gets very large negative charges on some of the methyl carbons and methyl hydrogens. I have looked at his wfn file and it indeed shows 45 doubly occupied molecular orbitals as expected. I have also noticed that the N value (which I am assuming is the number of electrons associated with the atom) ranges from 6.5-7.5 for the atoms with large negative charges. and the L value is relatively large (10^-1 to `0 ^-2) I have not yet tried the PROMEGA option. I have two questions, really. Can anyone tell me what the various information in the crt file means and is there a simple way to figure out what is causing the very large charges? The total charge does not add up to 0 as expected. It is equal to -35. I have included the relevant responses to my last email below Thank you, Julia Metzker Dear Dr. Metzker, It looks like you need to tightens the integration parameters in "PROAIMV" and use the "PROMEGA" option (which seldom fails, but slower of course) rather the usual option of "PROAIM" within the input files (.inp). To have an idea of the magnitudes of the charged on a Titanocene moiety, perhaps you may wish to have a look at: Bader RFW, Matta CF. Bonding to titanium. Inorg. Chem. 2001;40:5603-5611. Best of luck, Sincerely, Cherif ----------------------------------------- Chirif F. Matta, Ph.D. Chemistry Department, University of Toronto, Toronto, Ontario, Canada M5S 3H6. Telephone: (416) 978 - 3625 Fax: (416) 978 - 7580 Web Site: http://chem.utoronto.ca/~cmatta/ ------------------------------------------ I don't have any experience doing AIMPAC calcs, but I've read plenty of papers describing the calcs. The charges on C seem unrealistic. The sum of the atomic charges must be the same as the molecular charge. If you have a neutral molecule that contains some highly negatively charged atoms, it must also contain some highly positively charged atoms. You didn't mention the latter, so I suspect that something was set up incorrectly. Good luck, -Alan ==== Alan Shusterman Department of Chemistry Reed College Portland, OR Hello, Though I do not know which method you are using to get the wavefunction, I would blame first to the quality of the wavefunction than to the aimpac program. Check the boundaries of the carbon atoms to see if the atomic basin is what you expect to be. Hope it helps _-_-_-_-_-_-_-_-_-_-_-_- Carlos Silva Lspez Dept. Qummica Organica Universidade de Vigo Phone:0034 986812226 -_-_-_-_-_-_-_-_-_-_-_-_ Hi, I just read your email on the CCL. My first question is what software are you using to calculate the charges, proaimv, gaussian's implementation or Kunig's AIM200?. The second question is how did you generate the wave funcion, hf, dft, basis set. Best, f Fernando Martmn, Plexxikon 91 Bolivar Drive, Suite A Berkeley, CA 94710, USA 510.647.4014 (phone) 510.548.8014 (fax) Dear Julia, These types of systems can be problematic if the topology of the density is complex. For example, are you certain that you've successfully identified the ring and cage critical points in the ferrocene case (particularly) and assigned them to the associated bond and ring critical points, respectively? Problems in this area can lead to the "overintegration" problems that you've described. Curt Breneman RPI Chemistry --- You wrote: The charges actually do not add to 0, indicating a problem somewhere, but I don't know where. --- end of quote --- This is definitely a problem if your molecules are neutral. First, double-check the original wavefn/energy calculation. Perhaps the total charge was defined incorrectly. Second, check your basis set. It is quite common to use effective core potentials (ECP) with organometallic compounds like ferrocene. If you used an ECP basis set, you have effectively deleted some or all of the core electrons from your wavefn. When you use this wavefn to calculate atomic charges you might get unrealistic results (but I would expect to see large *positive* charges due to the missing electrons, and not large negative charges). Third, see if the wavefn and orbital occupancy info is being transferred correctly to AIMPAC. Perhaps AIMPAC thinks the wrong number of orbitals are populated? Good luck, -Alan ==== Alan Shusterman Department of Chemistry Reed College 3203 S.E. Woodstock Blvd. Portland, OR 97202-8199 503-517-7699 Dear Julia - have you checked the quality of the integrations? This is usually best done with the number reported as "L" (I think the fourth quantity output by proaimv). This is the value of the Laplacian integrated over the atomic basin, which should vanish - values of 10^-3 are borderline OK, lower is obviously better. Also check that the total populations add up to the no. electrons in your system. For a system like ferrocene, with rings (and cages?), you might need to tweak some of the surface-tracing paramaters - see the top of the proaimv.f code. If this isn't the source of your problem, you could try the integration methods in Gaussian (disappointing in my experience) or Popelier's MORPHY, which is generally very robust. Hope this helps - Jamie From chemistry-request@ccl.net Thu May 29 13:34:30 2003 Received: from llar.net.uniovi.es (llar.net.uniovi.es [156.35.11.3]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4THYTgC013273; Thu, 29 May 2003 13:34:30 -0400 Received: from CONVERSION-DAEMON.llar.net.uniovi.es by llar.net.uniovi.es (PMDF V6.1 #39514) id <01KWH4FFGCZ40096OW<>llar.net.uniovi.es>; Thu, 29 May 2003 19:34:10 +0200 (MET) Received: from pinon.ccu.uniovi.es (pinon.ccu.uniovi.es [156.35.31.1]) by llar.net.uniovi.es (PMDF V6.1 #39514) with ESMTP id <01KWH4FDHSU20095M2<>llar.net.uniovi.es>; Thu, 29 May 2003 19:34:07 +0200 (MET) Received: from pinon.ccu.uniovi.es (localhost [127.0.0.1]) by pinon.ccu.uniovi.es (8.12.9/8.12.9) with ESMTP id h4THY3n9027486; Thu, 29 May 2003 19:34:04 +0200 (METDST) Received: (from pueyo@localhost) by pinon.ccu.uniovi.es (8.12.9/8.12.8/Submit) id h4THY1nA027484; Thu, 29 May 2003 19:34:01 +0200 (METDST) Date: Thu, 29 May 2003 19:34:01 +0200 (METDST) From: Victor Lua~na >pinon.ccu.uniovi.es> Subject: Re: CCL:'electron density' vs 'electronic configurations'... To: chemistry<>ccl.net, owner-chemistry<>ccl.net Cc: pueyo<>pinon.ccu.uniovi.es Message-id: <200305291734.h4THY1nA027484<>pinon.ccu.uniovi.es> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT > From: Roy Jensen >uvic.ca> > Date: Mon, 26 May 2003 00:47:58 -0700 > The following came up recently and we could not come to a conclusion. > > One premise behind DFT is that a _single_ electron density completely > describes each electronic state of a system. In ab initio systems, > _all_ electronic configurations contribute to the wavefunction for > each electronic state of a system. Is this an artifact of the > approximations in ab initio calculations? Alternatively, if described > exactly, is an exact ab initio wavefunction still composed of all the > electronic configurations? I find some confussion in this description, so perhaps I am not understanding your question. DFT is founded on the Hohenberg-Kohn (HK) theorem, according to which, for a non-degenerated ground state, the exact energy can be obtained as a functional of the exact electron density. Thus the emphasis on the ground state electron density. However, every different electronic state has a different electron density (notice that the Born-Oppenheimer approximation is assumed here). >From a full CI perspective, the wavefunction of a given electronic state will be given as a linear combination of a large number of Slater determinants, each corresponding to a configuration (i.e. to a electronic state under the approximation of independent electrons). Therefore, the single electron density of an electronic state depends (contains information) on the coupling of all the electron-independent configurations entering the CI. > On a second point, would it be correct to say that DFT _variationally_ > optimizes the electron density to determine the minimum energy of the > system? There are *many* DFT techniques. Perhaps the most commonly found (at least in the molecular world) go back to the orbital description and solve the Kohn-Sham equations. KS equations are not an outcome of the variational principle. However, they are variational in the sense that, once an exchange and correlation functional has been fixed, improving the flexibility of the basis set will diminish the energy and converge to the best solution that could have been obtained from that basis set. Notice, however, that in most DFT methods neither the energy nor the electron density is obtained by solving a variational equation. Regards, Victor Lua~na -- HomePage %%http://www3.uniovi.es/~quimica.fisica/qcg/vlc/luana.html%% Free codes %%http://www3.uniovi.es/~quimica.fisica/qcg/%% +----------------------------------------------+ +---^---/ / ! Victor Lua~na ! | ~ / Just in case ! Departamento de Quimica Fisica y Analitica ! | | you don't ! Universidad de Oviedo, 33006-Oviedo, Spain ! < / remember ! e-mail: %%victor<>carbono.quimica.uniovi.es%% ! | / where Oviedo ! phone: +34-985-103491 fax: +34-985-103125 ! |____ ___/ is ;-) +----------------------------------------------+ \/ From chemistry-request@ccl.net Thu May 29 13:57:08 2003 Received: from aleph.net.uniovi.es (aleph.net.uniovi.es [156.35.11.1]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4THv8gC013674; Thu, 29 May 2003 13:57:08 -0400 Received: from CONVERSION-DAEMON.aleph.net.uniovi.es by aleph.net.uniovi.es (PMDF V6.1 #39514) id <01KWH57TSV8G00SP69|at|aleph.net.uniovi.es>; Thu, 29 May 2003 19:57:02 +0200 (MET) Received: from pinon.ccu.uniovi.es (pinon.ccu.uniovi.es [156.35.31.1]) by aleph.net.uniovi.es (PMDF V6.1 #39514) with ESMTP id <01KWH57SOF4U00Y6VQ|at|aleph.net.uniovi.es>; Thu, 29 May 2003 19:57:01 +0200 (MET) Received: from pinon.ccu.uniovi.es (localhost [127.0.0.1]) by pinon.ccu.uniovi.es (8.12.9/8.12.9) with ESMTP id h4THuxn9027785; Thu, 29 May 2003 19:56:59 +0200 (METDST) Received: (from pueyo@localhost) by pinon.ccu.uniovi.es (8.12.9/8.12.8/Submit) id h4THuwtP027784; Thu, 29 May 2003 19:56:58 +0200 (METDST) Date: Thu, 29 May 2003 19:56:58 +0200 (METDST) From: Victor Lua~na Subject: Re: CCL:calculating atomic charges with AIMPAC To: chemistry|at|ccl.net, owner-chemistry|at|ccl.net Cc: pueyo|at|pinon.ccu.uniovi.es Message-id: <200305291756.h4THuwtP027784|at|pinon.ccu.uniovi.es> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT > From: "Dr. Julia Metzker" > Date: Wed, 28 May 2003 15:13:03 +0100 > Subject: CCL:calculating atomic charges with AIMPAC > > I currently have an undergraduate student who is attempting to compare > charges of methylated benzene and ferrocene systems using the AIMPAC > software. In the case of hexamethyl benzene and the ferrocene analogues he > is getting very large negative charges on some the ring carbons (-5 to -20) > and small positive charges on iron (0.5) Can anyone tell me if there is a > problem with the program or if the ring system can not be treated with this > method? The charge integrated on the basin of a carbon atom goes from -5 to -20 electrons??? That looks as big mistake. Either, the wavefunction is not converged and it is made mostly of noise or the integration has completely failed in AIMPAC. I suggest you first examining the actual topology of the electron density. EXTREME is pretty fast and you can easily discard any wrong-looking wavefunction. If everythhing looks right you can try the *two* different integration techniques included within AIMPAC. See the documentation of PROAIM. Notice that the laplacian, L(r), should integrate to zero within a basin. An integral of 1e-4 a.u. is reasonable given the finite precision of numerical integrations. A value larger than some 1e-3 a.u. should make you suspect an error. Regards, Victor Lua~na -- HomePage %%http://www3.uniovi.es/~quimica.fisica/qcg/vlc/luana.html%% Free codes %%http://www3.uniovi.es/~quimica.fisica/qcg/%% +----------------------------------------------+ +---^---/ / ! Victor Lua~na ! | ~ / Just in case ! Departamento de Quimica Fisica y Analitica ! | | you don't ! Universidad de Oviedo, 33006-Oviedo, Spain ! < / remember ! e-mail: %%victor|at|carbono.quimica.uniovi.es%% ! | / where Oviedo ! phone: +34-985-103491 fax: +34-985-103125 ! |____ ___/ is ;-) +----------------------------------------------+ \/ From chemistry-request@ccl.net Thu May 29 14:17:31 2003 Received: from llar.net.uniovi.es (llar.net.uniovi.es [156.35.11.3]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TIHVgC014056; Thu, 29 May 2003 14:17:31 -0400 Received: from CONVERSION-DAEMON.llar.net.uniovi.es by llar.net.uniovi.es (PMDF V6.1 #39514) id <01KWH5X3BUVK0094K7|at|llar.net.uniovi.es>; Thu, 29 May 2003 20:17:25 +0200 (MET) Received: from pinon.ccu.uniovi.es (pinon.ccu.uniovi.es [156.35.31.1]) by llar.net.uniovi.es (PMDF V6.1 #39514) with ESMTP id <01KWH5X0R7TU009BLP|at|llar.net.uniovi.es>; Thu, 29 May 2003 20:17:21 +0200 (MET) Received: from pinon.ccu.uniovi.es (localhost [127.0.0.1]) by pinon.ccu.uniovi.es (8.12.9/8.12.9) with ESMTP id h4TIHJn9028035; Thu, 29 May 2003 20:17:20 +0200 (METDST) Received: (from pueyo@localhost) by pinon.ccu.uniovi.es (8.12.9/8.12.8/Submit) id h4TIHIZK028034; Thu, 29 May 2003 20:17:18 +0200 (METDST) Date: Thu, 29 May 2003 20:17:18 +0200 (METDST) From: Victor Lua~na Subject: Re: CCL:calculating atomic charges with AIMPAC To: chemistry|at|ccl.net, owner-chemistry|at|ccl.net Cc: pueyo|at|pinon.ccu.uniovi.es Message-id: <200305291817.h4TIHIZK028034|at|pinon.ccu.uniovi.es> MIME-version: 1.0 Content-type: text/plain; charset=us-ascii Content-transfer-encoding: 7BIT > From owner-chemistry|at|ccl.net Thu May 29 08:35:48 2003 > From: Wai-To Chan > Date: Thu, 29 May 2003 02:09:19 -0400 (EDT) > Subject: CCL:calculating atomic charges with AIMPAC > <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< > I currently have an undergraduate student who is attempting to compare > charges of methylated benzene and ferrocene systems using the AIMPAC > software. In the case of hexamethyl benzene and the ferrocene analogues he > is getting very large negative charges on some the ring carbons (-5 to -20) > and small positive charges on iron (0.5) Can anyone tell me if there is a > problem with the program or if the ring system can not be treated with this > method? > >>>>>>>>>>>>>>>>>>>>>>>>>>>>>>> > I remember reading about the tendency of AIM changes to be larger in magnitude > than expected on the basis of chemical intuition. There is a paper by > Kenneth Wiberg from the early 90s in J. comp. chem. about the comparison of > several atomic-charge partitioning scheme including Bader's you may find useful. > You may also consider using the Natural atomic charges method if you want > faster and less 'surprising' results. What "chemical intuition"? The one that comes from looking for many years to Mulliken populations in gaussian outputs? That is not science, it is prejudgment. Whatever interpretation we want to give to them, Bader charges are unique and well defined properties of the electron density of the molecule or solid under study. Both experimental and good theoretical electron densities agree on the calculated Bader charges of hundreths of compounds well characterized so far. On the other hand, Bader charges are not enough to analyze the electron densities. The whole multipolar series (charges are nothing but the zero order poles) should be considered instead. Several recent works by Popelier, for instance, would made a good reading in this respect. In any case, the charges quoted in the original question look like if the wavefunction is wrong by some reason. Perhaps a calculation not converged. Regards, Victor Lua~na -- HomePage %%http://www3.uniovi.es/~quimica.fisica/qcg/vlc/luana.html%% Free codes %%http://www3.uniovi.es/~quimica.fisica/qcg/%% +----------------------------------------------+ +---^---/ / ! Victor Lua~na ! | ~ / Just in case ! Departamento de Quimica Fisica y Analitica ! | | you don't ! Universidad de Oviedo, 33006-Oviedo, Spain ! < / remember ! e-mail: %%victor|at|carbono.quimica.uniovi.es%% ! | / where Oviedo ! phone: +34-985-103491 fax: +34-985-103125 ! |____ ___/ is ;-) +----------------------------------------------+ \/ From chemistry-request@ccl.net Thu May 29 13:40:17 2003 Received: from hqmailgwi1.aspentech.com (nathqmailgwi1.aspentech.com [208.201.88.4]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4THeGgC013412 for ; Thu, 29 May 2003 13:40:16 -0400 Received: by hqmailgwi1.aspentech.com with Internet Mail Service (5.5.2653.19) id ; Thu, 29 May 2003 13:36:28 -0400 Message-ID: <63799630D1424545BB6E6B0E98FE2B560223B28A|at|hqmail1.corp.aspentech.com> From: Paul.Mathias|at|aspentech.com To: dshobe|at|sud-chemieinc.com, wu_guosheng2002|at|yahoo.com, renxiao|at|med.umich.edu, chemistry|at|ccl.net Subject: RE: A question on heat of formation Date: Thu, 29 May 2003 13:40:13 -0400 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: multipart/alternative; boundary="----_=_NextPart_001_01C32609.5C07C4C0" This message is in MIME format. Since your mail reader does not understand this format, some or all of this message may not be legible. ------_=_NextPart_001_01C32609.5C07C4C0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable The NIST-JANAF tables are primarily based upon experimental data. Computational predictions are sratting to be used, but the contribution = is -- yet -- very small. =20 Paul Mathias -----Original Message----- From: Shobe, Dave [mailto:dshobe|at|sud-chemieinc.com] Sent: Thursday, May 29, 2003 12:02 PM To: 'Guosheng Wu'; 'Renxiao Wang'; 'CCL' Subject: CCL:A question on heat of formation ---Guosheng Wu wrote:=20 > It seems to me, dH has not much to do with propabilities of all conformations,=20 > which is essentially entropy contribution.=20 There is a small effect on enthalpy, for temperatures above zero = Kelvin. Most people ignore it, since there are more important sources of error. = But I think the NIST/JANAF tables do take this effect into account. It may affect dH/dT more than it affects dH itself. --David Shobe=20 S=FCd-Chemie Inc.=20 phone (502) 634-7409=20 fax (502) 634-7724=20 email dshobe|at|sud-chemieinc.com=20 Don't bother flaming me: I'm behind a firewall.=20 -----Original Message-----=20 From: Guosheng Wu [ mailto:wu_guosheng2002|at|yahoo.com ]=20 Sent: Thursday, May 29, 2003 11:20 AM=20 To: Shobe, Dave; 'Renxiao Wang'=20 Subject: Re: CCL:A question on heat of formation=20 Were you talking about free energy(dG), instead of heat of formation = (dH)?=20 It seems to me, dH has not much to do with propabilities of all conformations,=20 which is essentially entropy contribution.=20 BTW, John Pople and others had a few papers on Gn theory to get very = high quality=20 heat of formation for a small molecule, by ab initio plus scaling strategy... but=20 obviously not interested to most of the Med. Chem. community.=20 Guosheng=20 --- "Shobe, Dave" wrote:=20 > Most people do the following. They find the lowest-energy = conformations of=20 > the molecules involved, and use the energy at those geometries to calculate=20 > delta-H. The computational chemist should of course use any prudent, = > reasonable means to ensure that these are indeed the *global* conformational=20 > minima.=20 >=20 > However...=20 >=20 > I think to really "do it right" one has to calculate all the = low-lying=20 > conformations and determine the relative statistical-mechanical populations=20 > of each conformation, and from that the average enthalpy per molecule = (or=20 > per mole). =20 >=20 > Usually there are enough other sources of error that the correct = procedure > has little advantage over the first approach, at least for enthalpy. = But=20 > the corresponding "correct" procedure is often done for entropy, = where=20 > population of low-lying conformational isomers can make a significant = > difference in the outcome.=20 >=20 > --David Shobe=20 > S=FCd-Chemie Inc.=20 > phone (502) 634-7409=20 > fax (502) 634-7724=20 > email dshobe|at|sud-chemieinc.com=20 >=20 > Don't bother flaming me: I'm behind a firewall.=20 >=20 >=20 >=20 >=20 > -----Original Message-----=20 > From: Renxiao Wang [ mailto:renxiao|at|med.umich.edu ]=20 > Sent: Wednesday, May 28, 2003 8:19 PM=20 > To: chemistry|at|ccl.net=20 > Subject: CCL:A question on heat of formation=20 >=20 >=20 > Hi, CCLers!=20 >=20 > Should the heat of formations of two enantiomers of a chiral compound = be=20 > identical? I think the answer is yes. Please point out some = references=20 > for this.=20 >=20 > A disturbing fact is that, no matter you use QM or MM to calculate = the=20 > heat of formation for a given molecule, the number is=20 > conformation-depentdent. I guess this is because of the steric energy = > part is conformation-dependent. So, if I really want to compare the=20 > stability of two molecules using heat of formation, how may I = calculate=20 > it right?=20 >=20 > Best regards,=20 >=20 > Renxiao Wang=20 > renxiao|at|med.umich.edu=20 >=20 >=20 >=20 > -=3D This is automatically added to each message by mailing script = =3D-=20 > To send e-mail to subscribers of CCL put the string CCL: on your = Subject:=20 > line=20 >=20 > Send your subscription/unsubscription requests to: CHEMISTRY-REQUEST|at|ccl.net=20 >=20 > HOME Page: http://www.ccl.net | Jobs Page: http://www.ccl.net/jobs =20 >=20 >=20 >=20 >=20 >=20 >=20 >=20 __________________________________=20 Do you Yahoo!?=20 Yahoo! Calendar - Free online calendar with sync to Outlook(TM).=20 http://calendar.yahoo.com =20 ------_=_NextPart_001_01C32609.5C07C4C0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable RE: CCL:A question on heat of formation
The=20 NIST-JANAF tables are primarily based upon experimental data. =20 Computational predictions are sratting to be used, but the contribution = is --=20 yet -- very small.
 
Paul=20 Mathias
-----Original Message-----
From: Shobe, Dave=20 [mailto:dshobe|at|sud-chemieinc.com]
Sent: Thursday, May 29, = 2003 12:02=20 PM
To: 'Guosheng Wu'; 'Renxiao Wang'; = 'CCL'
Subject: CCL:A=20 question on heat of formation


---Guosheng Wu wrote:
> It seems to=20 me, dH has not much to do with propabilities of all = conformations,=20
> which is essentially entropy contribution. =

There is a small effect on enthalpy, for = temperatures above=20 zero Kelvin.  Most people ignore it, since there are more = important=20 sources of error.  But I think the NIST/JANAF tables do take = this effect=20 into account.  It may affect dH/dT more than it affects dH=20 itself.

--David Shobe
S=FCd-Chemie Inc.=20
phone (502) 634-7409
fax     (502) 634-7724
email  dshobe|at|sud-chemieinc.com

Don't bother flaming me: I'm behind a = firewall.=20



-----Original Message-----
From:=20 Guosheng Wu [mailto:wu_guosheng2002@yahoo.c= om]=20
Sent: Thursday, May 29, 2003 11:20 AM =
To: Shobe, Dave; 'Renxiao Wang'
Subject: Re:=20 CCL:A question on heat of formation


Were you talking about free energy(dG), instead of = heat of=20 formation (dH)?

It seems to me, dH has not much to do with = propabilities of=20 all conformations,
which is essentially = entropy=20 contribution.

BTW, John Pople and others had a few papers on Gn = theory to=20 get very high quality
heat of formation for = a small=20 molecule, by ab initio plus scaling strategy... but
obviously not interested to most of the Med. Chem. = community.=20

Guosheng

--- "Shobe, Dave" <dshobe|at|sud-chemieinc.com>=20 wrote:
> Most people do the = following.  They=20 find the lowest-energy conformations of
> the=20 molecules involved, and use the energy at those geometries to = calculate=20
> delta-H.  The computational chemist = should of=20 course use any prudent,
> reasonable = means to=20 ensure that these are indeed the *global* conformational =
> minima.
>
>=20 However...
>
&g= t; I think=20 to really "do it right" one has to calculate all the low-lying =
> conformations and determine the relative=20 statistical-mechanical populations
> of = each=20 conformation, and from that the average enthalpy per molecule = (or=20
> per mole). 
>=20
> Usually there are enough other sources = of error=20 that the correct procedure
> has little = advantage=20 over the first approach, at least for enthalpy.  But =
> the corresponding "correct" procedure is often done for = entropy,=20 where
> population of low-lying = conformational=20 isomers can make a significant
> = difference in the=20 outcome.
>
> --David=20 Shobe
> S=FCd-Chemie Inc. =
> phone (502) 634-7409
>=20 fax     (502) 634-7724
>=20 email  dshobe|at|sud-chemieinc.com
>=20
> Don't bother flaming me: I'm behind a=20 firewall.
>
>=20
>
> =
> -----Original Message-----
> From:=20 Renxiao Wang [mailto:renxiao|at|med.umich.edu]<= /FONT>=20
> Sent: Wednesday, May 28, 2003 8:19 PM =
> To: chemistry|at|ccl.net
> = Subject: CCL:A=20 question on heat of formation
> =
>
> Hi, CCLers! =
>
> Should the heat of = formations of two=20 enantiomers of a chiral compound be
> = identical? I=20 think the answer is yes. Please point out some references =
> for this.
> =
> A disturbing fact is that, no matter you use QM or MM = to calculate=20 the
> heat of formation for a given = molecule, the=20 number is
> conformation-depentdent. I = guess this=20 is because of the steric energy
> part = is=20 conformation-dependent. So, if I really want to compare the =
> stability of two molecules using heat of formation, how = may I=20 calculate
> it right?
>=20
> Best regards,
>=20
> Renxiao Wang
>=20 renxiao|at|med.umich.edu
>
>
>
> -=3D This=20 is automatically added to each message by mailing script =3D- =
> To send e-mail to subscribers of CCL put the string = CCL: on your=20 Subject:
> line
>=20
> Send your subscription/unsubscription = requests=20 to: CHEMISTRY-REQUEST|at|ccl.net
> =
> HOME Page: http://www.ccl.net   | Jobs = Page: http://www.ccl.net/jobs=20
>
> =
>
>
>=20
>
> =


__________________________________
Do=20 you Yahoo!?
Yahoo! Calendar - Free online = calendar=20 with sync to Outlook(TM).
http://calendar.yahoo.com =

------_=_NextPart_001_01C32609.5C07C4C0-- From chemistry-request@ccl.net Thu May 29 13:21:27 2003 Received: from web20609.mail.yahoo.com (web20609.mail.yahoo.com [216.136.226.167]) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4THLRgC013020 for ; Thu, 29 May 2003 13:21:27 -0400 Message-ID: <20030529172126.5031.qmail|at|web20609.mail.yahoo.com> Received: from [206.111.112.115] by web20609.mail.yahoo.com via HTTP; Thu, 29 May 2003 10:21:26 PDT Date: Thu, 29 May 2003 10:21:26 -0700 (PDT) From: Sengen Sun Subject: Re: CCL:Orbitals To: chemistry|at|ccl.net MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii E. Lewars wrote: If MOs have no physical reality for multielectron species, why (a) is Koopmans' theorem useful, why (b) do photoelectron spectra match the predictions of MO energy-level diagrams, and why (c) does the Hueckel 4n+2 rule, which is based on MO diagrams, work? These are extremely interesting questions to me. They are very difficult to be answered, as implied by two very prominent theoretical physicists in the last century Gell-Mann and Feynman: http://www.prometheus.demon.co.uk/02/02kumar.htm I agree with Scerri, Spanget-Larsen, and many others that MOs have no physical reality. But I don t think that we, as suggested by Feynman, should stop asking why MOs are so powerful for predictions in chemistry. If we stop thinking, we would accept the control of the universe by something of no physical reality; And we would not be able to achieve a real understanding of chemistry and the universe; And we would have conceptual or logical conflicts... __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com From chemistry-request@ccl.net Thu May 29 14:42:00 2003 Received: from web20511.mail.yahoo.com (web20511.mail.yahoo.com [216.136.175.150]) by server.ccl.net (8.12.8/8.12.8) with SMTP id h4TIg0gC014577 for ; Thu, 29 May 2003 14:42:00 -0400 Message-ID: <20030529184200.75383.qmail|at|web20511.mail.yahoo.com> Received: from [62.104.205.83] by web20511.mail.yahoo.com via HTTP; Thu, 29 May 2003 11:42:00 PDT Date: Thu, 29 May 2003 11:42:00 -0700 (PDT) From: Noriaki Ikuta Subject: RE: ic50 values in dtp aids dataset To: Paul Hawkins Cc: chemistry|at|ccl.net MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Paul Hawkins thank you for your respond. But it seems to be imposible the categorise the compounds based on the provided IC50 values. The ranges do significantly overlapp. categy mean min max #cmpds (several outliers are left out) CA -4.5 -7.0 -2.5 705 CM -4.0 -7.4 -2.7 1360 CI -4.0 -10.0 -1.9 36289 NA -4.4 -7.3 -3.0 294 Note on reliability: The expression "Log10HiConc - Log10IC50 > 1.0" is only met in roughly 18% of the cases. How useful are those IC50 values at all? Noriaki --- Paul Hawkins wrote: > Noriaki, > > I suspect the statement > > "concentration of compound that inhibits the growth of uninfecte > cells by 50%" > > is a typo. IC50 is the concentration of a compound required to reduce > the > readout of the assay by 50%, whether you are measuring enzyme activity, > receptor activation or cell growth/division. > The classification into the different activity classes is somewhat > arbitrary. Generally speaking a compound having an IC50 greater than > 10^-5 > is not considered active as the doses required to achieve those > concentrations in an animal (let alone a human) would be huge. For > comparison most anti-HIV drugs on the market now have IC50's in this > sort of > assay around 10^-6 and have IC50's against their target receptor (the > protease or reverse transcriptase mostly) in the 10^-9 range. > There may also be a known correlation between IC50's in this assay and > doses > required to cause remission of infection in an animal model of HIV in > humans > and again the maximum acceptable dose may translate to an IC50 in this > assay > of 10^-5. > I am unaware of any papers on this dataset in particular though a good > many > studies have used this dataset for various purposes. > > One word of warning Log10HiConc should always be around 10x greater than > the > IC50 so as the Log10HiConc approaches the reported IC50 the IC50 result > becomes less and less reliable. > > Paul Hawkins > Applications Scientist > Tripos Inc > 1699 South Hanley Road > St. Louis MO 63144 > > Ph: 617-899-4151 > E-mail: phawkins|at|tripos.com > > www.tripos.com > __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com From chemistry-request@ccl.net Thu May 29 15:16:56 2003 Received: from ns0.scripps.edu (ns0.scripps.edu [192.42.82.58]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TJGtgC015979 for ; Thu, 29 May 2003 15:16:56 -0400 Received: from antigen.scripps.edu (antigen.scripps.edu [137.131.200.100]) by ns0.scripps.edu (8.11.7/TSRI-4.2rx) with SMTP id h4TJGto10626 for ; Thu, 29 May 2003 12:16:55 -0700 (PDT) Received: from relay2.scripps.edu(137.131.200.30) by antigen.scripps.edu via csmap id 18814; Thu, 29 May 2003 12:13:36 -0700 (PDT) Received: from degas.scripps.edu (degas [137.131.252.44]) by relay2.scripps.edu (8.11.7/TSRI-4.4.1rAVB) with ESMTP id h4TJDvi11015; Thu, 29 May 2003 12:13:57 -0700 (PDT) Received: (from lou@localhost) by degas.scripps.edu (8.11.6+Sun/TSRI-4.1.0) id h4TJDtO13881; Thu, 29 May 2003 12:13:55 -0700 (PDT) Date: Thu, 29 May 2003 12:13:55 -0700 (PDT) From: Lou Noodleman Message-Id: <200305291913.h4TJDtO13881.-at-.degas.scripps.edu> To: chemistry.-at-.ccl.net Subject: Orbitals and Reality Cc: lou.-at-.scripps.edu, tlovell.-at-.scripps.edu, wengehan.-at-.scripps.edu, ithink.-at-.san.rr.com, sharples.-at-.scripps.edu, himo.-at-.TheoChem.kth.se X-Sun-Charset: US-ASCII Dear Dr. Fournier and Group at CCL, At the risk of getting into deep metaphysics, I want to make a few simple observations about this issue of the reality of orbitals. I have found some of the discussion surprising. "Are orbitals real?" No, orbitals are a model (equivalently, a representation of reality). Only "reality", "the things out there that we deal with" are real. Orbitals are " a map" of that reality, or more precisely, they are part of a family of theories of a certain level of accuracy. Are "many electron wavefunctions real"? No, when properly used, they may provide a more accurate representation of the real world, than wavefunctions based on a single set of orbitals. However, the information and more importantly, connections and generalizations to other systems may be more difficult to obtain. Is the solution to the many electron Schrodinger equation real? No, for example, what if the Born-Oppenheimer separation fails separating the nuclear wavefunction from the geometry dependent electronic wavefunction? What about the problems of dealing with the quantum mechanics of open systems? What about relativistic effects and their many subtleties? I do give higher priority at least as a touchstone to QM observables like electron density, which are measurable for example, > from high precision X-ray structures of well defined systems. But these are neither exact nor precise even in principle, given coupling to nuclear motion which is quantum mechanical, and in some cases not separable. Also, the world is an open system, with all the problems that gives. Since the world and all of its interconnected parts have an infinite number of facets, we will never know reality, but it is fun to understand something about some of the pieces and their linkages. Concerning orbitals, beyond their other uses, I recall from earlier work of Chris Brion that in e-2e inelastic scattering experiments, something very much like orbitals can be mapped out. In photoelectron spectroscopy, the multiple peaks in final electron kinetic energy for a given photon energy can usually be correlated with orbitals (but sometimes this collapses (see L. Cederbaum's work)) and for optical spectra, orbital analysis can be done in Time Dependent DFT, as shown by Rich Martin at Los Alamos. Concepts have value even when they are not perfect, and no concept is perfect (see Ralph Waldo Emerson). In general, I think of orbitals as somewhat real, less than electron densities, or atoms, but more than Lewis bond structures (at least, most of the time). > From a philosophical viewpoint, nothing that I have said is very new. "The gods did not reveal from the beginning All things to us; but in the course of time, Through seeking, men find that which is better. But as for certain truth, no man has known it. Nor will he know it." Xenophanes (6th Century B.C.) (quoted from Marion, "Classical Electromagnetic Radiation",1965). Also, Newton's quotation, better known among scientists, is as valid today as it was for him over 300 years ago. We forget the past at our peril. Lou Noodleman Department of Molecular Biology, TPC15 The Scripps Research Institute La Jolla, CA . From chemistry-request@ccl.net Thu May 29 16:11:43 2003 Received: from medusa.bioc.aecom.yu.edu (medusa.bioc.aecom.yu.edu [129.98.80.187]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TKBhgC017550 for ; Thu, 29 May 2003 16:11:43 -0400 Received: from medusa.bioc.aecom.yu.edu (localhost.bioc.aecom.yu.edu [127.0.0.1]) by medusa.bioc.aecom.yu.edu (8.12.8/8.12.5) with ESMTP id h4TKBeJx009396 for ; Thu, 29 May 2003 16:11:41 -0400 Received: from localhost (alewand@localhost) by medusa.bioc.aecom.yu.edu (8.12.8/8.12.8/Submit) with ESMTP id h4TKBetD009392 for ; Thu, 29 May 2003 16:11:40 -0400 Date: Thu, 29 May 2003 16:11:40 -0400 (EDT) From: alewand$at$medusa.bioc.aecom.yu.edu To: chemistry$at$ccl.net Subject: TS optimization with restraints in G98 Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII I 'm making optimization of transition state structure with restraints imposed especially along reaction coordinate. I got some converged structures on semiepmirical level with only one negative frequency (probably on DFT it will be more difficult to converge) but I still consider if approach is correct. Thank you for every regards concerning this calculation. Andrzej From chemistry-request@ccl.net Thu May 29 16:00:05 2003 Received: from roche.bath.ac.uk (roche.bath.ac.uk [138.38.32.21]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TK04gC017252 for ; Thu, 29 May 2003 16:00:04 -0400 Received: from midge.bath.ac.uk ([138.38.32.34] ident=mmdf) by roche.bath.ac.uk with smtp id 19LTZE-0008Uw-83 for chemistry$at$ccl.net; Thu, 29 May 2003 21:00:04 +0100 Received: from prpcn144 ( prpc-n144.bath.ac.uk [138.38.128.144] ) by bath.ac.uk id aa24625 for ; 29 May 2003 21:00 +0100 From: James Robinson To: chemistry$at$ccl.net Subject: RE: Orbitals and Reality Date: Thu, 29 May 2003 21:00:20 +0100 Message-ID: MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0) X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2600.0000 Importance: Normal In-Reply-To: <200305291913.h4TJDtO13881$at$degas.scripps.edu> X-Scanner: *19LTZE-0008Uw-83*/iEjJ0RUofk* Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by server.ccl.net id h4TK05gC017253 If we do not find anything very pleasant, at least we shall find something new.. Volatire. Is the basis of aromaticity and Woodward-Hoffman rules to be placed in a bin? I remember that some scientist captured an image of the letters IBM some years ago, written by aligning individual atoms (Switzerland I think) I wondered what the image was, what it meant. MOs themselves may be nothing more than the result of applied mathematics, however, I have difficulty in understanding a plausible alternative to describe the wave-nature of electrons. James, Bath. From chemistry-request@ccl.net Thu May 29 18:16:03 2003 Received: from ultra3000.if.sc.usp.br (ultra3000.if.sc.usp.br [143.107.228.1]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4TMG2gC020094 for ; Thu, 29 May 2003 18:16:02 -0400 Received: from joselap (zappa.if.sc.usp.br [143.107.228.35]) by ultra3000.if.sc.usp.br (8.11.6/8.11.0) with ESMTP id h4TMCHH18361 for ; Thu, 29 May 2003 19:12:17 -0300 Message-ID: <000e01c32631$07e30f80$0201a8c0@joselap> From: "JOSE" To: References: Subject: Re: CCL:Orbitals and Reality Date: Thu, 29 May 2003 19:24:12 -0300 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 6.00.2800.1158 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1165 I'm enjoying this discussion about MO. I should say that Molecular Orbitals are not Real, but Complex. It is a mathematical tool that represents the electronic structure of the matter. Please correct me if this is wrong. From chemistry-request@ccl.net Thu May 29 22:28:52 2003 Received: from pacific-carrier-annex.mit.edu (PACIFIC-CARRIER-ANNEX.MIT.EDU [18.7.21.83]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h4U2SpgC025933 for ; Thu, 29 May 2003 22:28:51 -0400 Received: from grand-central-station.mit.edu (GRAND-CENTRAL-STATION.MIT.EDU [18.7.21.82]) by pacific-carrier-annex.mit.edu (8.12.4/8.9.2) with ESMTP id h4U2SpZ7016932 for ; Thu, 29 May 2003 22:28:51 -0400 (EDT) Received: from manawatu-mail-centre.mit.edu (MANAWATU-MAIL-CENTRE.MIT.EDU [18.7.7.71]) by grand-central-station.mit.edu (8.12.4/8.9.2) with ESMTP id h4U2Sok9000362 for ; Thu, 29 May 2003 22:28:50 -0400 (EDT) Received: from cathedral-seven.mit.edu (CATHEDRAL-SEVEN.MIT.EDU [18.7.16.65]) ) by manawatu-mail-centre.mit.edu (8.12.4/8.12.4) with ESMTP id h4U2SoFJ022095 for ; Thu, 29 May 2003 22:28:50 -0400 (EDT) Received: from localhost (damians@localhost) by cathedral-seven.mit.edu (8.9.3p2) with ESMTP id WAA28137; Thu, 29 May 2003 22:28:49 -0400 (EDT) Date: Thu, 29 May 2003 22:28:49 -0400 (EDT) From: Damian A Scherlis Perel To: Subject: CCL: CASSCF and basis set superposition Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear all, I am performing a series of multireference calculations on dimers. In order to apply the basis set superposition correction I must do the calculations also on the monomer, which has as many basis functions but one half the number of electrons. On the other hand, while the dimer is a closed shell system, the monomer is a radical. My concern is the relation between the active spaces to be picked for the dimer and the monomer so that the energy difference may be obtained. For instance, let's say the active space is 8-electrons 6-orbitals (8,6) in the closed shell dimer, what should it be in the open shell monomer? Thanks so much, damian ------------------------ Damian Scherlis Department of Materials Science & Engineering Massachusetts Institute of Technology damians-.at.-mit.edu 617 253 6026