From CHFW@cc.newcastle.edu.au Sat Aug 29 23:29:56 1992 Date: Sun, 30 Aug 1992 13:29 +1000 From: CHFW@cc.newcastle.edu.au Subject: dipole moment again To: chemistry-request@ccl.net Status: RO Hi, The experts: I still could not be able to handle the CI dipole moment problem. I've tried the calculation on HeLiHe+ molecule with the "density=current" on the command line. However, G88 calculated the dipole moment after it finished SCF cycle but before the CI calculations. This means that the wavefunctions used in the dipole moment calculation are the SCF wavefunctions instead of CI wavefunctions. I am wondering that can G88 do the CI dipole moment calculations? Does anybody have some ideas about this problem? The part of my input file follows. Thanks in advance. ************************************************************************** $ Run [CHVO.G88]G88 #QCISD=FULL/GEN 6D units=(AU,deg) density=current HeLiHe+ 1 1 Li He 1 R1 He 1 R1 2 150.0 R1=3.92 ... Dipole moment(Debye) x =0.0000 y =0.0000 z =1.3907 ************************************************************************** -- | Feng Wang Department of Chemistry, The University of Newcastle | | CallagHan, N.S.W.2308, Australia | | Internet: chfw@cc.newcastle.edu.au Tel: +61 49 215 475 | |_____________________________________________________________________________| From lim@rani.chem.yale.edu Sun Aug 30 16:24:33 1992 From: Dongchul Lim Subject: ampac manual To: chemistry@ccl.net (Computational Chemistry) Date: Sun, 30 Aug 92 16:24:55 EDT Status: RO Was anyone able to unstuff 'ampman.sit' in ampac2.1? I downloaded ampac21.tar.Z from ouchem.chem.oakland.edu using binary mode, and tried to unstuff ampman.sit using 'unsit' on UNIX and 'StuffIt 1.5.1' and 'Stuffit Classic' on Macintosh, but both failed to uncompress it. 'unsit' program on UNIX gave the following diagnosis: Not a StuffIt file Can't read file header I guess this file was generated by an old version of 'StuffIt' or has been corrupted somehow. I'll appreciate any help about this problem. -dongchul lim Department of Chemistry Yale University From lim@rani.chem.yale.edu Sun Aug 30 20:23:06 1992 From: Dongchul Lim Subject: possible bug in newzmat (gaussian program utility) To: chemistry@ccl.net (Computational Chemistry) Date: Sun, 30 Aug 92 20:23:25 EDT Status: RO Using 'newzmat' that comes with Gaussian program, I tried to convert the following zmatrix which has both internal and cartesian zmatrix cards into mopac zmatrix. **** Gaussian Zmatrix **** #N 3-21G MIXED INTERNAL AND CARTESIAN ZMATRIX 0 1 C H 1 R2 H 1 R3 2 A3 6 -1.132208 0.000000 -0.676584 Variables: R2 1.07177300 R3 1.07200800 A3 117.12201900 **** Converted Mopac Zmatrix **** AM1 VECTORS T=10000 MIXED INTERNAL AND CARTESIAN ZMATRIX Title 2 dummied. C 0.000000 0 0.000000 0 0.000000 0 0 0 0 H 1.071773 1 0.000000 0 0.000000 0 1 0 0 H 1.072008 1 117.122019 1 0.000000 0 1 2 0 C -1.132208 0 0.000000 0 -73.255998 0 0 0 0 <=== ??? 00 Shouldn't the fourth zmatrix card be 'C -1.132208 0 0.000000 0 -0.676584 0 0 0 0'? -dongchul lim Department of Chemistry Yale University From ernst@bkfug.kfunigraz.ac.at Mon Aug 31 04:43:39 1992 Date: Mon, 31 Aug 1992 10:43:37 -0100 From: ernst@bkfug.kfunigraz.ac.at To: chemistry@ccl.net Subject: motecc summary Status: RO Here is the promised summary on MOTECC. I was able to contact CRS4, who are distributing the package at the time. Below are the relevant addresses and e-mail contacts for Dr. Clementi and CRS4. Below the information on CRS4 are some other (possibly outdated) contacts, which I did not verify. --------------------------------------------------------------------------- Enrico Clementi CRS4, Centre for Advanced Studies, Research and Development in Sardinia, PO Box 488, 09100 Cagliari, Italy MOTECC is currently distributed by us (CRS4, Italy). If you send me your postal address I can mail you information about MOTECC-91 (the current available version). **************************************************************************** * Dr. Omar G. Stradella - omar@crs4.it * * CRS4 - Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna * * - Centre for Advanced Studies, Research and Development in Sardinia * * -------------------------+----------------------------+----------------- * * Casella Postale 488 | Via Nazario Sauro, 10 | 39 13' 19" North * * 09100 Cagliari CA | 09123 Cagliari CA | 9 6' 17" East * * Italy | Italy | * * -------------------------+----------------------------+----------------- * * Phone: +39 70 27 96 236 FAX: +39 70 27 96 220 * **************************************************************************** another e-mail address: gina@crs4.it ----------------------------------------------------------------------------- ***************************************************************************** ----------------------------------------------------------------------------- Here are some other responses, none of which I verified. Some appear to be out of date by now. ------------------------------------------------------------------------------ MOTECC is available from IBM International Business Machines Corporation Kingston New York 12401 USA ------------------------------------------------------------------------------ You obtain it by signing a joint study agreement with IBM and paying something in the order of US$ 200. There are two volumes of documentation, edited by Clementi. Contact your local IBM reps (they probably won't know) and have them contact their nearest scientific support center (IBM people at Basel are informed). ------------------------------------------------------------------------------ Contact address for MOTECC as of 1991: Dr.Daniel Frye Scientific and Engineering Computations Department 48B/428 Neighborhood Road Kingston, NY 12401 ------------------------------------------------------------------------------ MOTECC -- Modern Techniques in Computational Chemistry is an IBM software Club which was organized by Enrico Clementi. You were paying IBM some $500 and got in exchange 20 or something software packages from atomic ab initio to molecular dynamics. With the programs you were getting the set of manuals and a very nice book. The last edition was: Enrico Clementi, MOTECC -- Modern Techniques in Computational Chemistry, Escom Publ., Leiden 1991, Netherlands, ISBN 90-72199-10-3 In 1991 the group of Clementi was dissolved. The only person who stayed in Kingston IBM is Michel Dupuis, michel@kgnvma.vnet.ibm.com. You may ask him for details. ------------------------------------------------------------------------------- ----- End of the message ----- From UDIM018@FRORS31.bitnet Mon Aug 31 04:44:45 1992 Date: 31 Aug 92 09:40:26 EDT From: Subject: ON ATOM CENTERED CHARGES To: chemistry@ccl.net Status: RO E. M. EVLETH Dynamique des Interactions Moleculaires Universite Pierre et Marie Curie 4 Place Jussieu, Tour 22, Paris 75005 (1) 44 27 42 08 UDIM018 at FRORS31 In response to Mark Murcko's request for information on which is the best current method for obtaining fitted charges, the picture is NOT CLEAR. F. Colonna has developed an electric field fitting method (J. Comp. Chem. in press, fall 1992) and we have run some comparisons using those in Gaussian 92. Let me show one simple example of methanol--optimized 6-31+G* geometry--This SCF level ab initio dipole moment is 2.00. The molecule has Cs symmetry with the H on HO staggered to the two H on the methyl. ________________________________________________________________ Method _________________________________________________________________ Atom Charge CHELPG CHELP MK Electric Field Fitted _________________________________________________________________ Carbon 0.39 0.41 0.28 0.22 Oxygen -0.73 -0.70 -0.74 -0.73 H(on O) 0.44 0.40 0.45 0.46 H(on C) 0.01 -0.01 0.05 0.06 H(on C) -0.05 -0.05 -0.02 0.00 H(on C) -0.05 -0.05 -0.02 0.00 Dipole (regen.) 1.96 1.92 1.99 2.00 ab initio = 2.00 Number of Grid points 5900 411 419 1500 ________________________________________________________________ The items of particular note are that the carbon atom charges vary with the method and that the various methods use various numbers of grid points in the fitting procedure. First it must be remembered that all these are fitting procedures, do not attempt (in my view) to make chemical sense out of them. For instance: If one imposes a ZERO charge value on the carbon atom, the other atoms will gives field fitted charges of (O),-0.68; (H),0.46;(H),0.12;(H),0.05; (H),0.05, a charge distribution which gives a dipole moment of 1.99, i.e. nearly as good as fully fitted model. The absence of carbon in the fitting procedure merely forces the other atoms to adjust, the dipole doesn't change much although the error in the regeneration of the electric field gets slightly worse (not shown). One can anticipate that as the molecule gets larger, large fluctuations could be obtained for particular charges on particular atoms even using the same method if atoms were removed from or fictive atoms added to the fitting procedure. Secondly, these all being grid methods the charges will depend on the number of grid points and the position of the grid vis-a-vis the nearest atoms. Thirdly, different methods use different statistical methods of evaluated the precision of the fit and one can not know which is the best until these are all compared on the same basis, mere accuracy in the regeneration of the dipole moment is not sufficient. Finally, atom fitted charges modeling can not intrinsically, accurately, and generally reproduce the potential or electric field to same precision as one obtains adding multipoles to the fitting procedure. I will also add that using a MP2 density matrix may change the fitted charges by 0.1 unit or more. In stating that the situation is not clear, I mean to say that until the above (and William's among others) methods are statistically compared by the same methods, the 'best' method can not be stated. One basis for claiming that CHELPG charges are superior to CHELP has been established in the literature. A major point is that CHELPG uses a large number of grid points. However, all the ESP method are subject to penetration effects if one makes an error in setting the grid too close, this not being a problem with Colonna's method. In any case, there is more work to be done to determine the best method. E. M. Evleth From omar@malena.crs4.it Mon Aug 31 04:49:14 1992 To: chemistry@ccl.net Subject: MOTECC information Date: Mon, 31 Aug 92 10:49:07 +0100 From: Omar G. Stradella Status: RO Due to the large number of requests of information about MOTECC that I received in the last few days, I decided to set up a anonymous ftp to retrieve the documents. The machine is: malena.crs4.it (156.148.7.12), look in the directory: pub/motecc. Currently the files are: ---------------------------------------------------------------------------- motecc.info.txt Information about MOTECC-91 in plain ascii format. ---------------------------------------------------------------------------- motecc.info.troff Information about MOTECC-91 in troff format. motecc.form.troff MOTECC-91 order form in troff format. motecc.license.troff MOTECC-91 license agreement in troff format. ---------------------------------------------------------------------------- motecc.info.ps Information about MOTECC-91 in PostScript format. motecc.form.ps MOTECC-91 order form in PostScript format. motecc.license.ps MOTECC-91 license agreement in PostScript format. ---------------------------------------------------------------------------- -- **************************************************************************** * Dr. Omar G. Stradella - omar@crs4.it * * CRS4 - Centro di Ricerca, Sviluppo e Studi Superiori in Sardegna * * - Centre for Advanced Studies, Research and Development in Sardinia * * -------------------------+----------------------------+----------------- * * Casella Postale 488 | Via Nazario Sauro, 10 | 39 13' 19" North * * 09100 Cagliari CA | 09123 Cagliari CA | 9 6' 17" East * * Italy | Italy | * * -------------------------+----------------------------+----------------- * * Phone: +39 70 27 96 236 FAX: +39 70 27 96 220 * **************************************************************************** From rwm@stan.xx.swin.OZ.AU Mon Aug 31 05:36:19 1992 Date: Mon, 31 Aug 92 19:37:35 +1000 From: rwm@stan.xx.swin.OZ.AU (MAIR R) To: chemistry@ccl.net Subject: More NMR software Status: RO Netters, I was going through some old news printouts last night, and came across a reference to a package called IRMA - relaxation matrix software, by Kaptein in Holland. The printout did not have a sender on it, so I was wondering: Has anybody heard of it? Do you know what it does? Can it be obtained easily, and preferably without money? Thanks in advance, Ross Mair Swinburne University of Technology Hawthorn, Victoria, Australia. From sender@chemdc1.tau.ac.il Mon Aug 31 10:43:19 1992 Date: Mon, 31 Aug 92 17:40:37 -0100 From: sender@chemdc1.tau.ac.il (Senderowitz Hanoch) To: CHEMISTRY@ccl.net Status: RO Hello everybody Does anyone know what's happening with the MGS Autumn meeting on 'Entropy and Solvation' held at Oxford on September 17 and 18? We have been trying to contact Dr. Forster and Dr. Broughton via e-mail, Fax and telephone for the last 3 weeks but got no reply and time is running real short... Thanks, Hanoch Senderowiz (sender@chemdc1.tau.ac.il) From jdurant@snll-arpagw.llnl.gov Mon Aug 31 11:24:39 1992 From: jdurant@snll-arpagw.llnl.gov (durant joseph) Subject: IRC in G92 To: chemistry@oscsunb.ccl.net (Computational Chemistry) Date: Mon, 31 Aug 92 8:24:36 PDT Status: R Hello all, I am doing an IRC calculation with G92, and was wondering what G92 does if the reaction coordinate bifurcates. Does it print out a message, crash, or chose one of the paths to follow? If it is this last option, is there a way I can test each point for two negative eigenvalues? A bit of background: I am looking at the HNNO potential energy surface using G90 and now G92. I have found the transition state for H-atom loss (there appears to be only one, for either cis or trans HNNO). I want to follow the reaction coordinate back and see if it bifurcates (presumably one fork going to cisHNNO, one to transHNNO) or if it only goes back to cisHNNO (presumably meaning that the transHNNO isomerizes before dissociating). Thanks for your help on this. Joe Durant jdurant@snll-arpagw.llnl.gov Combustion Research Facility Sandia National Labs Livermore, CA From lrellick@magnus.acs.ohio-state.edu Mon Aug 31 14:08:01 1992 Date: Mon, 31 Aug 92 14:07:57 EDT From: Lorraine M Rellick To: chemistry@ccl.net Subject: "best" set of Van der Waal's radii Status: RO Dear Netters, I need to use Van der Waal's radii for an application program I am writing. I would like to know what the current thinking is on the "best" set to use. Many Thank-yous, Lori Rellick, Biochemistry Dept. OSU . From brianh@fai.fai.com Mon Aug 31 17:13:30 1992 Date: Mon, 31 Aug 92 14:16:47-1795 From: Brian Hammond To: chemistry@ccl.net Subject: Response to Meiko MPP product question Status: RO In regard to the question about Meiko's new MPP product and Fujitsu, my understanding is that Meiko has decided to go with SPARC cpu chips and the new Fujitsu vector processor chips, which were announced at the hot chips conference. Yours, Brian Hammond Computational Research Division Fujitsu America, Inc. 3055 Orchard Drive, San Jose 95134 Tel: (408) 456-7322 Email: brianh@fai.com From doherty@msc.edu Mon Aug 31 17:36:01 1992 From: doherty@msc.edu (David C. Doherty) Subject: Re: "best" set of Van der Waal's radii To: lrellick@magnus.acs.ohio-state.edu (Lorraine M Rellick) Date: Mon, 31 Aug 92 16:35:46 CDT Status: RO > >Dear Netters, I need to use Van der Waal's radii for an application program >I am writing. I would like to know what the current thinking is on the >"best" set to use. Many Thank-yous, Lori Rellick, Biochemistry Dept. OSU > a good place to start is: Bondi, A., _J. Phys. Chem._, 68 (3), 441 (1964). -- David C. Doherty Minnesota Supercomputer Center doherty@msc.edu From frederik@pollux.acs.uci.edu Mon Aug 31 20:05:05 1992 To: Lorraine M Rellick Subject: Re: "best" set of Van der Waal's radii Date: Mon, 31 Aug 92 17:04:55 -0700 From: "Donald M. Frederick" Status: R > Dear Netters, I need to use Van der Waal's radii for an application program > I am writing. I would like to know what the current thinking is on the > "best" set to use. Many Thank-yous, Lori Rellick, Biochemistry Dept. OSU > Dear Netters, I need to use Van der Waal's radii for an application program > I am writing. I would like to know what the current thinking is on the > "best" set to use. Many Thank-yous, Lori Rellick, Biochemistry Dept. OSU Would you please summarize the responses to the Net for those of us who might be interested? Thanks. ------------------------------------------------------------------------ Donald Frederick | Office of Academic Computing frederik@uci.edu | University of California, Irvine frederik@UCI.BITNET | Irvine, CA 92717 (714) 725-3200 | FAX (714) 725-2069 -----------------------------------------------------------------------