From ep7@dent.okayama-u.ac.jp Mon Jun 8 02:46:35 1998 Received: from deews1.dent.okayama-u.ac.jp (deews1.dent.okayama-u.ac.jp [150.46.202.36]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id CAA22378 Mon, 8 Jun 1998 02:46:32 -0400 (EDT) From: ep7@dent.okayama-u.ac.jp Received: from [150.46.140.80] (yobou2.dent.okayama-u.ac.jp [150.46.140.80]) by deews1.dent.okayama-u.ac.jp (8.8.8/3.6W) with SMTP id PAA03578 for ; Mon, 8 Jun 1998 15:47:54 +0900 (JST) Message-Id: <199806080647.PAA03578@deews1.dent.okayama-u.ac.jp> Date: Mon, 8 Jun 1998 16:03:40 +0900 To: CHEMISTRY@www.ccl.net Subject: L MIME-Version: 1.0 Content-Type: text/plain; charset=iso-2022-jp X-Mailer: Eudora-J(1.3.8.5-J13) Dear Sir: I bought GAUSSIAN 94 User's Reference. However, I don't know what is L in optimized structural parameters. Please replay to me. Thank you. Masao Masamura Preventive Dentistry Okayama University Dental School Shikata-cho, 2-5-1 Okayama 700 Japan FAX: 81-86-235ー6714 e-mail: ep7@dent.okayama-u.ac.jp From dario@rs5.csrsrc.mi.cnr.it Mon Jun 8 03:38:30 1998 Received: from rs6.csrsrc.mi.cnr.it (rs6.csrsrc.mi.cnr.it [155.253.3.36]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id DAA22603 Mon, 8 Jun 1998 03:38:16 -0400 (EDT) Received: from rs5.csrsrc.mi.cnr.it (rs5.csrsrc.mi.cnr.it [155.253.3.35]) by rs6.csrsrc.mi.cnr.it (8.8.8/8.8.8) with ESMTP id JAA22463 for ; Mon, 8 Jun 1998 09:32:03 +0200 Received: (from dario@localhost) by rs5.csrsrc.mi.cnr.it (8.8.8/8.8.8) id IAA12006 for CHEMISTRY@www.ccl.net; Mon, 8 Jun 1998 08:31:37 +0100 Date: Mon, 8 Jun 1998 08:31:37 +0100 From: Dario Bressanini Message-Id: <199806080731.IAA12006@rs5.csrsrc.mi.cnr.it> To: CHEMISTRY@www.ccl.net Subject: ONE Electron Molecule. Hi Chem-Netters, I am looking for references on calculations and studies on One electron molecules (starting with H2+ and beyond) Even old "hystoric" papers are welcomed Thanks Dario Bressanini dario@rs5.csrsrc.mi.cnr.it From Lommatzsch@chemie.uni-frankfurt.de Mon Jun 8 04:08:46 1998 Received: from faust27-s.rz.uni-frankfurt.de (pp@faust27-s.rz.uni-frankfurt.de [141.2.149.151]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id EAA22754 Mon, 8 Jun 1998 04:08:43 -0400 (EDT) From: Lommatzsch@chemie.uni-frankfurt.de Message-Id: <199806080808.EAA22754@www.ccl.net> Received: from rz.uni-frankfurt.de by faust27-s.rz.uni-frankfurt.de id <32462-0@faust27-s.rz.uni-frankfurt.de>; Mon, 8 Jun 1998 10:08:06 +0000 X-Mailer: exmh version 2.0.2 2/24/98 To: chemistry@www.ccl.net Subject: Program for solving 1d nuclear motion schroedinger equation Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 08 Jun 98 10:08:05 +0200 X-Host: vulpius.rz.uni-frankfurt.de Sender: Lommatzsch@chemie.uni-frankfurt.de Dear CCLers, I am searching for a program, which solves the one-dimensional Schroedinger equation for the nuclear motion, after fitting (polynomial) the potential curve to some calculated energy points. The program should yield the energies of the vibrational levels and the expectation values for the internuclear distance. I need the program for educational purposes in a quantum chemistry course. (HCl rotation-vibration spectrum) The program should run under WIN NT, preferably. Thanks in advance Uwe Lommatzsch **************************************************************************** Johann Wolfgang Goethe Universitaet Frankfurt Inst. fuer Physikalische und Theoretische Chemie Marie-Curie-Str. 11 60439 Frankfurt / Main Germany Phone: &49 (69) 798- 29425 Fax: &49 (69) 798- 29560 WWW: http://www.rz.uni-frankfurt.de/~lommatz **************************************************************************** From tajkhors@mbp-sgi7.inet.dkfz-heidelberg.de Mon Jun 8 04:50:44 1998 Received: from mbp-sgi7.inet.dkfz-heidelberg.de (mbp-sgi7.inet.dkfz-heidelberg.de [193.174.59.125]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id EAA22997 Mon, 8 Jun 1998 04:50:12 -0400 (EDT) Received: by mbp-sgi7.inet.dkfz-heidelberg.de (951211.SGI.8.6.12.PATCH1502/940406.SGI.AUTO) for chemistry@www.ccl.net id RAA06588; Mon, 8 Jun 1998 17:50:23 GMT From: "Emadeddin Tajkhorshid" Message-Id: <9806081050.ZM6584@mbp-sgi7.inet.dkfz-heidelberg.de> Date: Mon, 8 Jun 1998 10:50:10 -0700 Reply-To: E.Tajkhorshid@DKFZ-Heidelberg.de X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: chemistry@www.ccl.net (Computational Chemistry List) Subject: Summary: Frequencies calculated at different levels of theory Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="PART-BOUNDARY=.19806081050.ZM6584.inet.dkfz-heidelberg.de" --PART-BOUNDARY=.19806081050.ZM6584.inet.dkfz-heidelberg.de Content-Description: Text Content-Type: text/plain ; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable X-Zm-Decoding-Hint: mimencode -q -u Dear all Last week, I posted a question about the use of frequencies to judge abou= t the accuracy of geometries. Thanx to anybody who responded to my question. Sp= eciall thanx to Prof. Shusterman who, to my eyes, gave the exact answer to my question. I also beleive that we may not generally use the closeness of the frequen= cies to experimental values as a measure of the accuracy of geometries. Howeve= r, there are different opinions among the answers I received. Here are the summary of the answers I got from the community. *************************************************************************= ******* Question: Can we decide on the basis of the closeness of the frequencies obtaiened = at one level to judge about the accuracy of the geometrical parameters calculate= d in that model? *************************************************************************= ******* The answer to this question is no. The most widely understood example of this is the performance of the Hartree-Fock model which can give very accurate geometries for a wide ran= ge of molecules. The accuracy of these predictions is somewhat basis set depend= ent (and better basis sets are not always more accurate). The interesting thing is that the Hartree-Fock model always gives poor vibrational frequencies regardless of the basis set, although the errors = depend on the type of vibration (larger errors are seen for stretching vibration= s). The explanation for this paradoxical behavior, good geometries + poor frequencies, is that the latter depends on information derived from sever= al geometries and correlation errors are geometry dependent. Therefore, mini= ma on the Hartree-Fock energy surface may occupy nearly the same positions as t= he minima on the true surface, but the curvature around the minima on these surfaces can be substantially different. Alan ----------- Alan Shusterman Department of Chemistry Reed College Portland, OR www.reed.edu/~alan *************************************************************************= ****** There is an excellent review by Scott and Radom in either J. Phys. Chem o= r J. Chem. Phys. from a year or so ago that compares levels of theory on vibrational frequencies. Also, Hehre (I think) was an author of a book a= few years ago comparing Ab Initio with experimental properties. Hope that helps. Darren Andrews. PostGraduate Student, School of Chemistry, University of Leeds, Leeds. LS2 9JT. England. Darrena@chem.leeds.ac.uk Tel: 0113 233 6594. Fax: 0113 233 6565. *************************************************************************= ***** I believe that yes, you can draw this conclusion. At least for DFT, it ha= s been shown that the reference geometry is more important for accurate frequencies and force constants than the actual flavor of DFT used. Cf., e.g., A. B=E9rces ("Berces" if your mailing program can't read the "=E9" character) and T. Ziegler, in: Topics in Current Chemistry, Vol. 182, R. = F. Nalewajski (ed.), Springer-Verlag, Berlin, 1996; (same authors), Topics i= n Curr. Chem. 1996, 182, 41. Besides, it is also possible to calculate vibrational frequencies at any given reference geometry, even if it is no= t the stationary point at that particular level of theory. This is, however= , not possible to do with the Gaussian program. Cf., e.g., the same references, but also others by A. B=E9rces as well as: P. Pulay et al. JA= CS 1979, 101, 2550; G. Fogarasi et al. JACS 1992, 114, 8191. I hope this helps a little ... In any case, I am interested in your summa= ry. Best regards, Georg -- =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D Dr. Georg Schreckenbach Tel: (USA)-505-667 7605 Theoretical Chemistry T-12 FAX: (USA)-505-665 3909 M.S. B268, Los Alamos National E-mail: schrecke@t12.lanl.gov Laboratory, Los Alamos, New Mexico, 87545, USA Internet: http://www.t12.lanl.gov/~schrecke/ =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D *************************************************************************= ****** Well, this is a big problem. Let me start this way. The consistency of the results (bond lengths, frequencies etc) from a particular method is not preserved. Some cases even at HF/4-31G produce very accurate distance= (eg Li-H) - but in general this method is not satisfactory. DFT generally= slightly overestimate bond lengths, again it is not true for all type of molecules. I do not have any ready ref. on comparison of dft and other methods but sure there are more than 10-15 papers. Now in your case it shows DFT (again which coorelation and exchange functional are used, B3LYP is generally good but not good for transition metal) yields frequencies close to expt. So for this particular system distances are accurate. And you can say for the similar kind of molecule it will be acurate. Please check= the scale factor of DFT method. Tapas Kar, Ph. D Asst. Scientist Department of Chemistry Southern Illinois University at Carbondale Illinois 62901-4409 Fax: (618) 453 6408 Tel: (618) 453 6433 *************************************************************************= ***** -- = Emad ********************************************************************* E. Tajkhorshid = German Cancer Research Center; DKFZ Tel: +49 6221 42 2340 Dept. Molecular Biophysics (H0200) FAX: +49 6221 42 2333 P.O.Box 101949 = 69009 Heidelberg, Germany Email: E.Tajkhorshid@DKFZ-Heidelberg.de = ********************************************************************* * "Never express yourself more clearly than you think." -Niels Bohr * ********************************************************************* --PART-BOUNDARY=.19806081050.ZM6584.inet.dkfz-heidelberg.de-- From tajkhors@mbp-sgi7.inet.dkfz-heidelberg.de Mon Jun 8 08:12:04 1998 Received: from mbp-sgi7.inet.dkfz-heidelberg.de (mbp-sgi7.inet.dkfz-heidelberg.de [193.174.59.125]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id IAA23822 Mon, 8 Jun 1998 08:12:03 -0400 (EDT) Received: by mbp-sgi7.inet.dkfz-heidelberg.de (951211.SGI.8.6.12.PATCH1502/940406.SGI.AUTO) for chemistry@www.ccl.net id VAA09945; Mon, 8 Jun 1998 21:12:28 GMT From: "Emadeddin Tajkhorshid" Message-Id: <9806081412.ZM9943@mbp-sgi7.inet.dkfz-heidelberg.de> Date: Mon, 8 Jun 1998 14:12:16 -0700 Reply-To: E.Tajkhorshid@DKFZ-Heidelberg.de X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) To: chemistry@www.ccl.net (Computational Chemistry List) Subject: Summary2 : Frequencies calculated Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Hi again I have got two more respnses to my question which I will enclose here. Thanx for your consideration. BTW, any further coments are still welcome:-) **************************************************************************** Regarding your query about geom vs freq accuracy. Since the calculation of the freq. depend on the second derivative of the energy and how the basis set describes the anharmonicity of a bond stretch ( or bend ), there is not so simple as to compare with the exp. spectra and then say that the geom. is accurately described. More often there are problem with getting some calc. freqs to correlate well with spectra and others worse, which ones can change with the basis set and theory used, which again leaves you with the problem of which level of calc. to choose for the "correct" geometry... But, and I guess that this is more important, I believe that in _most_ cases ( I have no refs or reasons ) the geom. is OK, when the calc. freqs. are...if you get other replies saying differently I'm willing to reconsider best wishes Patrik -- Dr Patrik Johansson Inorg. Chem. Angstromlaboratory Box 538 SE-751 21 Uppsala Uppsala University SWEDEN patrikj@moore.kemi.uu.se ********************************************************************** If you have used harmonic approximation to calculate frequensies from force fiels and space structure in this case in first level of approach the bond deformation (bending) are not depended from the interatomic distances. This fact is due the kinematic coefficient for this type of vibration [k=(1/m1+1/m2), where m1 and m2 are atomic masses] is not depended from the interatomic distances. For other type of vibration the kinematic coefficients are depended from the interatomic distances but these dependences are so complicated. To rest on our twenty years experience in vibration calculation the force field quality makes the main influence on frequencies. Dr. V.Khavryutchenko Head of computation spectroscopy group Kiev, UKRAINE e-mail: vkhavr@compchem.kiev.ua ************************************************************************** -- Emad ********************************************************************* E. Tajkhorshid German Cancer Research Center; DKFZ Tel: +49 6221 42 2340 Dept. Molecular Biophysics (H0200) FAX: +49 6221 42 2333 P.O.Box 101949 69009 Heidelberg, Germany Email: E.Tajkhorshid@DKFZ-Heidelberg.de ********************************************************************* * "Never express yourself more clearly than you think." -Niels Bohr * ********************************************************************* From dapprich@tchibm3.chemie.uni-karlsruhe.de Mon Jun 8 15:14:50 1998 Received: from nz15.rz.uni-karlsruhe.de (nz15.rz.uni-karlsruhe.de [129.13.64.15]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id PAA27303 Mon, 8 Jun 1998 15:14:47 -0400 (EDT) Received: from tchopin (tchopin.chemie.uni-karlsruhe.de [129.13.107.204]) by nz15.rz.uni-karlsruhe.de with smtp id 0yj7Mt-0006CZ-00; Mon, 8 Jun 1998 21:14:07 +0200 Message-ID: <000f01bd9311$1cc5e3e0$cc6b0d81@tchopin.chemie.uni-karlsruhe.de> From: "Stefan Dapprich" To: Subject: JAVA frontend for TURBOMOLE? Date: Mon, 8 Jun 1998 21:10:27 +0200 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 4.72.2106.4 X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4 Dear CCLers: The quantum chemistry group at the University of Karlsruhe is currently evaluating the various possibilities for developing a graphical frontend for the popular ab initio package TURBOMOLE. Because such a GUI should be platform independent and easy to maintain we got the impression that JAVA might be an appropriate programming language for this task. However, first we would like to get some general advice and feedback from potential users: - What do people think about the necessity of a GUI for TURBOMOLE? - Is JAVA (AWT/SWING) the best basis for its realization? - What about speed issues? - What features should be implemented? Any comments, hints, and suggestions are appreciated. Stefan -- Dr. Stefan Dapprich Institut fuer Physikalische Chemie Universitaet Karlsruhe (TH) Engesserstrasse 7 D-76128 Karlsruhe Phone: ++49-721-608-7229 FAX: ++49-721-608-4856 E-Mail: dapprich@tchibm3.chemie.uni-karlsruhe.de From pc1@itr2.p.lodz.pl Mon Jun 8 15:36:49 1998 Received: from itr2.p.lodz.pl (pc1@itr2.p.lodz.pl [193.59.47.60]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with SMTP id PAA27424 Mon, 8 Jun 1998 15:36:46 -0400 (EDT) Received: (from pc1@localhost) by itr2.p.lodz.pl (8.6.12/8.6.9) id VAA05255; Mon, 8 Jun 1998 21:42:50 +0200 Date: Mon, 8 Jun 1998 21:42:49 +0200 (MET DST) From: Przemek Czyryca To: chemistry@www.ccl.net Subject: Re: CCL:PC GAMESS or Linux? In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII On Thu, 4 Jun 1998, Alistair Nelson wrote: > My question is about whether it is possible to remotely access a pc using > PC GAMESS, if it is how do you do it. It's possible. There is a free VNC (Virtual Network Computing) package, acting approximately like Unix's X-server and client (over TCP/IP network). However, It's very slow and a 10Mbit Ethernet is a minimum requirement. Working using a modem is probably impossible. More about it: http://www.orl.co.uk/vnc/ > If it isn't then could someone give > me some idea of how easy it is to install and, get running, Linux. The installation is really simple. You may encounter some (but not really serious) problems while compiling GAMESS (for example - the Fortran compiler is invoked by 'fort77' instead of 'f77' used in GAMESS'S scripts). Regards, Przemek G. Czyryca pczyryca@itr2.p.lodz.pl From slawek@alchmist.scs.uiuc.edu Mon Jun 8 17:35:41 1998 Received: from labrador.scs.uiuc.edu (labrador.scs.uiuc.edu [130.126.227.79]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id RAA27968 Mon, 8 Jun 1998 17:35:41 -0400 (EDT) Received: from piglet (piglet.scs.uiuc.edu [130.126.227.130]) by labrador.scs.uiuc.edu (AIX4.2/UCB 8.7/8.7) with SMTP id QAA15974 for ; Mon, 8 Jun 1998 16:19:20 -0500 (CDT) From: "Slawek Janicki" To: Subject: FW: PC GAMESS or Linux? Date: Mon, 8 Jun 1998 16:35:45 -0500 Message-ID: <005401bd9325$64f961a0$82e37e82@piglet.scs.uiuc.edu> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook 8.5, Build 4.71.2173.0 Importance: Normal X-MimeOLE: Produced By Microsoft MimeOLE V4.72.2106.4 Windows NT Resource Kit (both Workstation and Server) has a utility called REMOTE.EXE that opens a command prompt session on a remote NT machine. This is all you need to run PC GAMESS remotely. If you need full graphics consider Microsoft Windows Terminal Server (code name HYDRA). For a modem connection, REMOTE.EXE will do well of you dial in to the NT box. Otherwise pcAnywhere is a viable option. Slawomir Janicki slawek@alchmist.scs.uiuc.edu -----Original Message----- From: Computational Chemistry List [mailto:chemistry-request@www.ccl.net] On Behalf Of Przemek Czyryca Sent: Monday, June 08, 1998 2:43 PM To: chemistry@www.ccl.net Subject: CCL:PC GAMESS or Linux? On Thu, 4 Jun 1998, Alistair Nelson wrote: > My question is about whether it is possible to remotely access a pc using > PC GAMESS, if it is how do you do it. It's possible. There is a free VNC (Virtual Network Computing) package, acting approximately like Unix's X-server and client (over TCP/IP network). However, It's very slow and a 10Mbit Ethernet is a minimum requirement. Working using a modem is probably impossible. More about it: http://www.orl.co.uk/vnc/ > If it isn't then could someone give > me some idea of how easy it is to install and, get running, Linux. The installation is really simple. You may encounter some (but not really serious) problems while compiling GAMESS (for example - the Fortran compiler is invoked by 'fort77' instead of 'f77' used in GAMESS'S scripts). Regards, Przemek G. Czyryca pczyryca@itr2.p.lodz.pl -------This is added Automatically by the Software-------- -- Original Sender Envelope Address: pc1@itr2.p.lodz.pl -- Original Sender From: Address: pc1@itr2.p.lodz.pl CHEMISTRY@www.ccl.net: Everybody | CHEMISTRY-REQUEST@www.ccl.net: Coordinator MAILSERV@www.ccl.net: HELP CHEMISTRY or HELP SEARCH | Gopher: www.ccl.net 73 Anon. ftp: www.ccl.net | CHEMISTRY-SEARCH@www.ccl.net -- archive search Web: http://www.ccl.net/chemistry.html From mn1@helix.nih.gov Mon Jun 8 18:52:20 1998 Received: from helix.nih.gov (helix.nih.gov [128.231.2.3]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id SAA28302 Mon, 8 Jun 1998 18:52:19 -0400 (EDT) Received: (from mn1@localhost) by helix.nih.gov (8.8.5/8.8.5) id SAA20376; Mon, 8 Jun 1998 18:52:20 -0400 (EDT) Date: Mon, 8 Jun 1998 18:52:20 -0400 (EDT) From: "M. Nicklaus" Message-Id: <199806082252.SAA20376@helix.nih.gov> To: CHEMISTRY@www.ccl.net Subject: Enhanced NCI Database Browser back on-line Cc: mn1@helix.nih.gov To all beta testers of the Enhanced NCI Database Browser: The U.S. server for the Enhanced NCI Database Browser is back on-line. Wolf-Dietrich Ihlenfeldt Marc C. Nicklaus Computer Chemistry Center, Lab. of Medicinal Chemistry, University of Erlangen-Nuernberg National Cancer Institute, NIH D-91052 Erlangen (Germany) Bethesda, MD 20892 (USA) ------------------------------------------------------------------------ Marc C. Nicklaus National Institutes of Health E-mail: mn1@helix.nih.gov Bldg 37, Rm 5B29 Phone: (301) 402-3111 37 Convent Dr, MSC 4255 Fax: (301) 496-5839 BETHESDA, MD 20892-4255 USA http://rex.nci.nih.gov/RESEARCH/basic/medchem/mcnbio.htm Laboratory of Medicinal Chemistry, National Cancer Institute, & Center for Molecular Modeling, Ctr. for Information Technology, NIH ------------------------------------------------------------------------ From ep7@dent.okayama-u.ac.jp Mon Jun 8 20:57:14 1998 Received: from deews1.dent.okayama-u.ac.jp (deews1.dent.okayama-u.ac.jp [150.46.202.36]) by www.ccl.net (8.8.3/8.8.6/OSC/CCL 1.0) with ESMTP id UAA29164 Mon, 8 Jun 1998 20:57:11 -0400 (EDT) From: ep7@dent.okayama-u.ac.jp Received: from [150.46.140.80] (yobou2.dent.okayama-u.ac.jp [150.46.140.80]) by deews1.dent.okayama-u.ac.jp (8.8.8/3.6W) with SMTP id JAA08500 for ; Tue, 9 Jun 1998 09:58:29 +0900 (JST) Message-Id: <199806090058.JAA08500@deews1.dent.okayama-u.ac.jp> Date: Tue, 9 Jun 1998 10:14:17 +0900 To: CHEMISTRY@www.ccl.net Subject: Summary: ESP MIME-Version: 1.0 Content-Type: text/plain; charset=iso-2022-jp X-Mailer: Eudora-J(1.3.8.5-J13) Dear Sir: I would like to ask about determination of atomic charges from ab initio electrostatic potentials: For example, in CHELP, CHELPG, and Merz and Kollman, various points were normally putted at the van der Waals contact region surrounding the molecule ? Thank you. Masao Masamura Preventive Dentistry Okayama University Dental School Shikata-cho, 2-5-1 Okayama 700 Japan FAX: 81-86-235-6714 e-mail: ep7@dent.okayama-u.ac.jp Date: Tue, 26 May 1998 13:22:46 +0200 From: ulf@hugin.teokem.lu.se (Ulf Ryde) To: ep7@dent.okayama-u.ac.jp Subject: Re: CCL:ESP X-UIDL: 76f2788761e07a72a5d44d1577ae3865 Dear Prof. Masamura, we discuss related questions in our article: E. Sigfridsson & U. Ryde (1998) A comparison of methods for deriving atomic charges from the electrostatic potential and moments. J. Comp. Chem. 19, 377-395. It is also described in an electronic poster: http://garm.teokem.lu.se/~teoema/chargeposter.html Best Regards, Ulf ------------------------------------------------------------------- Dr. Ulf Ryde Ulf.Ryde@teokem.lu.se Assistant professor http://signe.teokem.lu.se/~ulf Department of Theoretical Chemistry University of Lund Chemical centre, P. O. Box 124 Phone: +46-46-2224502 S-221 00, Lund, SWEDEN Fax: +46-46-2224543 ------------------------------------------------------------------- Dear Sir: In CHELMO and CHELP-BOW, points can be putted within van der Waars radii ? Thank you. Date: Wed, 3 Jun 1998 07:24:23 +0200 From: ulf@hugin.teokem.lu.se (Ulf Ryde) To: ep7@dent.okayama-u.ac.jp Subject: Re: X-UIDL: 0b23c747e0840b455770337f714df0d5 Dear Prof. Masamura, in CHELMO, the charges are determined by a fit directly to the electrostatic moments of the molecule. Therefore, no potential is used and we do not have to decide what points to use. In the CHELP-BOW method, the charges are determined by a fit to the electrostatic potential as usual, but all points are Boltzmann-weighted according to the distance between the potential point and the atoms in the molecule using the potential function of the program in which the charges will be used. Thus, points arbitrary close to any atoms in the molecule are used (i.e. also within the van der Waals radius of the atoms), but they will have a very small weight. Best regards, Ulf Date: Sat, 30 May 1998 15:59:41 +0200 From: Konrad Hinsen To: ep7@dent.okayama-u.ac.jp Subject: Re: CCL:ESP X-UIDL: b1a7d35d0fd869af7c95c5ffb21b6b64 > I would like to ask about determination of atomic charges from ab > initio electrostatic potentials: For example, in CHELP, CHELPG, and Merz > and Kollman, various points were normally putted at the van der Waals > contact region surrounding the molecule ? Right, but the procedures differ in detail. And there are also other methods. The two essential points in which charge fitting methods differ are: 1) The choice of evaluation points. Grid-based points have many problems, mostly due to the dependence of the results on the grid axes. Points on shells or simple random points in a shell are better choices. 2) The fitting algorithm. CHELP and CHELPG use an algorithm that is known to be numerically unstable, leading to unrealistically high charges. The Merz/Kollman approach tries to get around this problem by adding charge constraints, but doesn't change the fit algorithm itself. An SVD-based least-squares fit is safer, and in fact this has been the standard least-squares algorithm for 20 years in all field except computational chemistry. -- ------------------------------------------------------------------------------- Konrad Hinsen | E-Mail: hinsen@ibs.ibs.fr Laboratoire de Dynamique Moleculaire | Tel.: +33-4.76.88.99.28 Institut de Biologie Structurale | Fax: +33-4.76.88.54.94 41, av. des Martyrs | Deutsch/Esperanto/English/ 38027 Grenoble Cedex 1, France | Nederlands/Francais -------------------------------------------------------------------------------