From mercie@med.cornell.edu Sat Dec 4 00:24:09 1993 Received: from cumc.cornell.edu for mercie@med.cornell.edu by www.ccl.net (8.6.4/930601.1506) id XAA08435; Fri, 3 Dec 1993 23:40:24 -0500 Received: from localhost (mercie@localhost) by cumc.cornell.edu (8.6.4/ECH1.13) id XAA02688; Fri, 3 Dec 1993 23:38:46 -0500 Date: Fri, 3 Dec 1993 23:38:43 -0500 (EST) From: Gustavo Mercier Subject: Re: POPULATION ANALYSIS To: Patrick Bultinck cc: CHEMISTRY@ccl.net In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi! To add to the discussion... In the 70's when the electrostatic potential became popular to explain molecular reactivity, it was very common to use MK derived charges to compute the potential from "large" molecules. It was very clear then that in many cases the dipole computed from the charges was poor. Nevertheless, the concept was useful, and in spite of poor dipole moments the TOTAL energy computed with the charges for a CONGENERIC family of molecules could be used to reproduce trends in chemical reactivity. In these cases systematic errors probably were controlled by only comparing SIMILAR molecules. But when is similar "dissimilar"? mercier mercie@cumc.cornell.edu From b_duke@lacebark.ntu.edu.au Sat Dec 4 01:52:55 1993 Received: from lacebark.ntu.edu.au for b_duke@lacebark.ntu.edu.au by www.ccl.net (8.6.4/930601.1506) id BAA08747; Sat, 4 Dec 1993 01:08:17 -0500 From: Received: by lacebark.ntu.edu.au (AIX 3.2/UCB 5.64/4.03) id AA07765; Sat, 4 Dec 1993 13:30:37 -0600 Message-Id: <9312041930.AA07765@lacebark.ntu.edu.au> Subject: Different UMP2 results with G90 and G92. To: CHEMISTRY@ccl.net (chemistry) Date: Sat, 4 Dec 1993 13:30:35 -0600 (CST) X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Length: 1927 Hi, CCL netters, I have found a rather puzzling and disturbing difference between results obtained by Gaussian90 and Gaussian92 for geometry optimisations at UMP2 level. I have been looking at FO2. I tried to duplicate the 6-31G* results of Frank Jensen, Chem Phys Lett 169, 519, 1990. G92 gives his UMP2 result, OF = 1.3832; OO = 1.2497 amd FOO angle = 109.6. This is with UMP2=FULL. G90 however, with UMP2 (FULL is default) gives results very close to the ones described by Jensen as AUMP2 (which were frozen core) - OF = 1.6894 (1.6985); OO = 1.1669 (1.1641); FOO angle = 113.3 (113.5) - Jensen values in (). AUMP2 appears to do the MP2 expansion on the UHF wave function after spin projection, while UMP2 does it before spin projection (i.e. on the straight UHF wave function). The frequencies likewise essentially agree with Jensen's UMP2 and AUMP2 results, as do the average values of S^2 before and after projection and the total energies (although the G90 UMP2 FULL is lower as expected than Jensen's AUMP2 frozen core). The two calcualtions both started with OPT=CALCFC from a value near to the UMP2 G92 result. Starting from the G90 geometry, G92 has FOO falling apart, with the F dissociating from O2 exactly as happens at UHF level. This is how I found this by trying to do the frequency with G92 from the G90 geometry on a differennt machine - it gave nonsense so I reoptimised. I can find no mention in the gaussian literature about this difference. Can anyone throw any light on it? It does seem that G90 is doing what Jensen calls AUMP2, while G92 is doing straight UMP2. I would most welcome any comments on this. Regards to everyone. Brian. -- Associate Professor Brian Salter-Duke (Brian Duke) School of Chemistry and Earth Sciences, Northern Territory University, Box 40146, Casuarina, NT 0811, Australia. Phone 089-466702 e-mail: b_duke@lacebark.ntu.edu.au or b_duke@uncl04.ntu.edu.au From m10!frisch@uunet.UU.NET Sat Dec 4 08:52:57 1993 Received: from relay1.UU.NET for m10!frisch@uunet.UU.NET by www.ccl.net (8.6.4/930601.1506) id IAA09582; Sat, 4 Dec 1993 08:40:07 -0500 Received: from spool.uu.net (via LOCALHOST) by relay1.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA11250; Sat, 4 Dec 93 08:40:06 -0500 Message-Id: <9312041340.AA11250@relay1.UU.NET> Received: from m10.UUCP by uucp6.uu.net with UUCP/RMAIL (queueing-rmail) id 083840.17748; Sat, 4 Dec 1993 08:38:40 EST Received: by m10.gaussian.com; Sat, 4 Dec 93 08:39:31 EST Date: Sat, 4 Dec 93 08:39:31 EST From: m10!frisch@uunet.UU.NET (Michael Frisch) Subject: Re: Different UMP2 results with G90 and G92. To: chemistry@ccl.net In-Reply-To: , Sat, 4 Dec 1993 13:30:35 -0600 (CST) Brian Salter-Duke asks: Hi, CCL netters, I have found a rather puzzling and disturbing difference between results obtained by Gaussian90 and Gaussian92 for geometry optimisations at UMP2 level. I have been looking at FO2. I tried to duplicate the 6-31G* results of Frank Jensen, Chem Phys Lett 169, 519, 1990. G92 gives his UMP2 result, OF = 1.3832; OO = 1.2497 amd FOO angle = 109.6. This is with UMP2=FULL. G90 however, with UMP2 (FULL is default) gives results very close to the ones described by Jensen as AUMP2 (which were frozen core) - OF = 1.6894 (1.6985); OO = 1.1669 (1.1641); FOO angle = 113.3 (113.5) - Jensen values in (). AUMP2 appears to do the MP2 expansion on the UHF wave function after spin projection, while UMP2 does it before spin projection (i.e. on the straight UHF wave function). The frequencies likewise essentially agree with Jensen's UMP2 and AUMP2 results, as do the average values of S^2 before and after projection and the total energies (although the G90 UMP2 FULL is lower as expected than Jensen's AUMP2 frozen core). The two calcualtions both started with OPT=CALCFC from a value near to the UMP2 G92 result. Starting from the G90 geometry, G92 has FOO falling apart, with the F dissociating from O2 exactly as happens at UHF level. This is how I found this by trying to do the frequency with G92 from the G90 geometry on a differennt machine - it gave nonsense so I reoptimised. I can find no mention in the gaussian literature about this difference. Can anyone throw any light on it? It does seem that G90 is doing what Jensen calls AUMP2, while G92 is doing straight UMP2. I would most welcome any comments on this. Both G92 and G90 do regular UMP2, period. The only diffence between them other than peformance is that G92 can do frozen-core MP2 gradients and, as Brian notes, therefore does frozen-core optimizations by default. The projected (PMP) energies printed by Gaussian are not used during geometry optimization. Note, however, that CALCFC means different things in G90 and G92. G90 could only do SCF analytic second derivatives, and CALCFC meant "do the SCF Hessian at the beginning of the optimization" regardless of whether HF, MP2, QCISD, or whatever was being optimized. Since G92 can the MP2 Hessian as well, there are now two versions of this options: CALCFC means what it says: calculate the force constants at the level of theory for the optimization (i.e., HF for an HF optimization, MP2 for an MP2 optimization). CALCHFFC means do the SCF force constants regardless of the level used for the optimization. I would speculate that the two optimizations took different paths because of different initial Hessians. If you started with the bond lengths on the long side of equilibrium, then whether the optimization moved inward or dissociated could change with whether the optimizer thought it was in a region of positive or negative curvature, which might well be different for the SCF and MP2 Hessians. Mike Frisch ------- From eloranta@tukki.jyu.fi Sat Dec 4 13:53:00 1993 Received: from tukki.jyu.fi for eloranta@tukki.jyu.fi by www.ccl.net (8.6.4/930601.1506) id NAA10454; Sat, 4 Dec 1993 13:11:46 -0500 Received: by tukki.jyu.fi id AA18794 (5.67a/IDA-1.4.4 for chemistry@ccl.net); Sat, 4 Dec 1993 20:08:25 +0200 From: Jussi Eloranta Message-Id: <199312041808.AA18794@tukki.jyu.fi> Subject: amsterdam dft code... (questions) To: chemistry@ccl.net Date: Sat, 4 Dec 1993 20:08:23 +0200 (EET) X-Mailer: ELM [version 2.4 PL21] Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 8bit Content-Length: 197 Can someone give more details on this code (Amsterdam DFT)? More specifically is it single point or optimization code? And is it commercial or free (for academic) users? Regards, Jussi Eloranta From 100012.1163@CompuServe.COM Sat Dec 4 14:07:28 1993 Received: from arl-img-1.compuserve.com for 100012.1163@CompuServe.COM by www.ccl.net (8.6.4/930601.1506) id NAA10521; Sat, 4 Dec 1993 13:45:27 -0500 Received: from localhost by arl-img-1.compuserve.com (8.6.4/5.930129sam) id NAA05893; Sat, 4 Dec 1993 13:44:55 -0500 Date: 04 Dec 93 13:42:11 EST From: "100012.1163@compuserve.com Rainer Stumpe" <100012.1163@CompuServe.COM> To: , "To:" , "To:" , "To:" , "To:" Subject: Springer-Verlag's Server updated Message-ID: <931204184210_100012.1163_BHB22-1@CompuServe.COM> To: >internet:CHEMISTRY@ccl.net; To: >internet:CHMINF-L@IUBVM.UCS.INDIANA.EDU; To: >internet:orgchem@extreme.chem.rpi.edu; To: >internet:ORGANOMET@MAILSERV.ZDV.UNI-TUEBINGEN.DE; To: >internet:POLYMERP@HEARN.BITNET Dear Fellow Netters, I have recently updated several files in the CHEMISTRY directory of Springer-Verlag's file server. A list of available files is attached. You can get these and other files through e-mail and ftp. 1. E-Mail: Send a message GET /CHEMISTRY/ to: SVSERV@VAX.NTP.SPRINGER.DE NOTE: Binaries (extension *.exe and *.zip) will be sent UUdeCoded. That requires you have the UUCODE software. A MSDOS version is available from the directory /UTILITY. If you order UUCODE.ZIP you will also need the file BINSTART.TXT. For a help file with all commands available explained, send the message HELP; for a complete list of files available send the command: DIRALL to the server's internet address. 2. FTP: The address of the anonymous ftp-server is: TRICK.NTP.SPRINGER.DE (or 192.129.24.12) User ID is ANONYMOUS Password is Your_e-mail_address All files will be updated at irregular intervals. If you have any suggestions for additional information, please drop me a message. Springer-Verlag also offers a Journals_Preview_Service which makes tables of contents for some biosciences journals available a few weeks before release of the issues. *** ALL SERVICES MENTIONED ARE FREE *** Sincerely yours Rainer Stumpe Sciences Editorial Springer-Verlag Tiergartenstr. 17 D-69121 Heidelberg Phone: +49-(0)6221-487 310 Fax: +49-(0)6221-487 366 Internet: Stumpe@spint.compuserve.com ----------------------------------------------------------- Directory /CHEMISTRY Enzyme Handbook by Schomburg etal (GBF): enz_hb.zip Enzyme Handbook indexes Vols. 1-6 Advances in Polymer Science (Review series): aps1_108.exe Biblio. Ref. "Adv. Polym. Sci." Vols 1-108 Molecular Modelling Software MOBY: mobynl01.txt MOBY Newsletter, Issue 1 (1993) mobydemo.exe MOBY 1.5 Demo Version in German/English manualps.exe English Manual in PS format for MOBY1.5 Theoretica Chimica Acta: tca_v82.txt Theoretica Chimica Acta, ToC Vol. 82 tca_v83.txt Theoretica Chimica Acta, ToC Vol. 83 tca_V84.exe Theoretica Chimica Acta, ToCs of Vol. 84 tca_V85.exe Theoretica Chimica Acta, ToCs of Vol. 85 tca_V86.exe Theoretica Chimica Acta, ToC Iss. 86/1-6 tca87_12.txt Theoretica Chimica Acta, ToC Iss. 87/1,2 tcasubs.txt Theoretica Chimica Acta, Subscri. Info. Topics in Current Chemistry (Review series): tcc_cont.exe Topics in Cur. Chem. Ind. Vols. 26--168 tcc_v163.txt TCC Volume TOC Vol. 163 tcc_v164.txt TCC Volume TOC Vol. 164 tcc_v165.txt TCC Volume TOC Vol. 165 tcc_v166.txt TCC Volume TOC Vol. 166 tcc_v167.txt TCC Volume TOC Vol. 167 tcc_v168.txt TCC Volume TOC Vol. 168 tcc_read.me Comment to TCC_CONT.EXE Lecture Notes in Chemistry (series): lnche-60.exe Lect. Notes in Chem. Dir Vol 1 to 60 lnchem61.txt Lect. Notes in Chem. Vol.61 Structure and Bonding (Review series): sb_cont.exe Struc. Bonding Index Vols 1-81 sb_v80.txt Struc. Bonding Index Vol. 80 sb_v81.txt Struc. Bonding Index Vols. 81 sb_read.me Comment to SB_CONT.EXE Directory /UTILITY Binstart.txt Start for newcomer (MSDOS) Help-utl.txt Help to utilities Pkzip.zip PKZIP (PKWARE Inc.) (MSDOS) Pkunzip.exe PKUNZIP (PKWARE Inc.) (MSDOS) Uucode.zip UuCode for MSDOS by R. Marks Unzip-42.hqx UNZIP for MAC UUlite-14.hqx UUDECODE for MAC by Jeff Strobel Unzip.tar_z UNZIP for UNIX systems ----------------------End of message--------------------- From lith!kestas@ucdavis.edu Sat Dec 4 21:53:05 1993 Received: from ucdavis.ucdavis.edu for lith!kestas@ucdavis.edu by www.ccl.net (8.6.4/930601.1506) id UAA11646; Sat, 4 Dec 1993 20:54:56 -0500 Received: from lith.UUCP by ucdavis.ucdavis.edu (4.1/UCD2.05) id AA12181; Sat, 4 Dec 93 17:43:44 PST Received: by lith.UUCP (smail2.5.3-coh) id AA11236; 4 Dec 93 17:10:20 PST (Sat) Subject: Automatic Analogue Proposal (Summary) To: chemistry@ccl.net From: Kestutis Sliupas X-Mailer: ELM [version 2.3 PL11] Message-Id: <9312041710.AA11234@lith.UUCP> Date: 4 Dec 93 17:10:20 PST (Sat) Hi: I posted the following message, and received two fax's from vendors, Tripos and Bio-Sym. They are sending literature; however, I am really surprised that no users dropped a line. Perhaps it was the US holidays? When the lit. arrives, I will filter and post a summary. As an aside, does anyone know how to contact Dr. Marsili, the author of ANALOG, or obtain a copy for evaluation? Forwarded message: > Hi: > What programs are being used to suggest analogues of > lead compounds for biological evaluation? I personally am aware > of 'ANALOG' by Dr. M. Marsili, as described in a recent issue > of 'The Journal of Chemical Information and Computer Sciences', > but I am sure there must be several more. > What is their availability (ie. public domain, commercial), > strengths and weaknesses, ease of use, and requirements for input/output > formats? What is their general utility? Do they save time and money on > synthesis? Do they do an adequate job on proposing sufficiently diverse > analogues needed for a first-pass to explore the chemical and physical > parameters needed for bio. activity, or are they an interesting excercise? > Please reply via e-mail. I will summarise in a couple of > weeks. -- Regards, Kestas Kestutis Sliupas x (510) 231-1540 (work voice) kestas%lith.UUCP@ucdavis.edu x (510) 231-1285 (work FAX) USA CA