From chemistry-request@ccl.net Sat Jun 14 10:02:22 2003 Received: from dinamica.localdomain ([200.183.85.218]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h5EE2K5u023122 for >ccl.net>; Sat, 14 Jun 2003 10:02:21 -0400 Received: from garantes by dinamica.localdomain with local (Exim 3.35 #1 (Debian)) id 19RBbn-0000At-00 for >ccl.net>; Sat, 14 Jun 2003 11:02:19 -0300 Date: Sat, 14 Jun 2003 11:02:18 -0300 From: Guilherme Menegon >webcable.com.br> To: chemistry<>ccl.net Subject: CCL: Re: NDDO semiempirical parameters Message-ID: <20030614140205.GA648<>webcable.com.br> Reply-To: dinamica<>webcable.com.br References: <20030613111408.M52463<>webcable.com.br> <4.2.0.58.20030613124252.00ae43a0<>postoffice.worldnet.att.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4.2.0.58.20030613124252.00ae43a0<>postoffice.worldnet.att.net> User-Agent: Mutt/1.3.28i Thanks James, Mark and jmmckel@attglobal for the replies. Let me post all together: On Fri, Jun 13, 2003 at 12:49:10PM -0600, James J. P. Stewart wrote: > Guilherme Menegon asked, > > >1) Do you have any routine or the iterative formulas to obtain the derived > >parameters, rho1 and rho2, from the optimized ones (namely 1 center 2 > >electron integrals, g_ij, and zeta_i) ?? > > The recursive, iterative procedures for obtaining these parameters were > first described in M. J. S. Dewar and W. Thiel. Ground States of Molecules, > 38. The MNDO Method. Approximations and Parameters. J. Am. Chem. Soc., > 99:4899-4907, 1977. > > For a brief discussion, please see: > http://www.cachesoftware.com/mopac/Mopac2002manual/node447.html Thanks for this url because it has the correct expresions for a_n. The ones given in Stewart's paper in J. Computer-Aided Molecular Design (Vol. 4, 1990, 1-105) seem incorrect. There is also some difference between the expression for the initial estimate of rho_2 given by Stewart in the J. Computer-Aided paper and that given by Dewar and Thiel in TCA 46, 1977, 89-104. But this seems to be unimportant for the final iterated rho_2 value. > > 2) Can you suggest any references which explicitly comment on the > > individual influence of the NDDO parameters upon calculated > > molecular properties? > > E.g.: If one varies the resonance integrals (beta_i), what is > > the effect upon geometries? Or if one varies the one center one > > electron integrals (U_ii), what is the effect upon charge > > distribution? > > ... > u can experiment with some of these param modifications, by hand ... Yes, in fact that is what I have been doing for a while, but it is quite puzzling because there is some interplay between the parameters and most of it does not seem physically clear (to my simple mind). Many of Dewar's papers say that geometries and heats of formation are quite insensitive upon variations in the one center one electron (U_ii) and one center two electron integrals (g_ij). In my experience, this is true for the g_ij terms but not for the U_ii, specially concerning the energies. On the other hand variation in U_ii results in quite unreasonable charge polarization (what is justifiable since one can correlate U_ii to the atomic electronegativity - e.g., M.J. Field, Mol. Phys. 91, 1997, 835-745). It would be great to hear more details from James and Mark, who should have a lot of experience to share about this art of semi-empirical parameter optimization. On Fri, Jun 13, 2003 at 05:02:20PM -0400, jmmckel<>attglobal.net wrote: > > ... > > See routine CALPAR Are you talking about MOPAC routines? Any special version/release? I am not using this software. Thanks again for the replies, Guilherme Menegon IQUSP, Brasil From chemistry-request@ccl.net Sat Jun 14 06:19:19 2003 Received: from gannet.scg.man.ac.uk (gannet.scg.man.ac.uk [130.88.94.110]) by server.ccl.net (8.12.8/8.12.8) with ESMTP id h5EAJI5u017225 for ; Sat, 14 Jun 2003 06:19:18 -0400 Received: from caleb.pa.man.ac.uk ([130.88.11.79] helo=man.ac.uk) by gannet.scg.man.ac.uk with esmtp (Exim 2.05 #6) id 19R87s-000Iln-00; Sat, 14 Jun 2003 11:19:12 +0100 Sender: mbpsspb1/at/ccl.net Message-ID: <3EEAF0E4.5D6010C4/at/man.ac.uk> Date: Sat, 14 Jun 2003 09:54:44 +0000 From: Pascal Bonnet X-Mailer: Mozilla 4.76C-SGI [en] (X11; I; IRIX64 6.5 IP30) X-Accept-Language: en MIME-Version: 1.0 To: zaldini/at/ufpe.br CC: chemistry/at/ccl.net Subject: Re: CCL:FORTRAN & C communication References: <1272.150.161.174.74.1055506373.squirrel/at/www.webmail.ufpe.br> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Hi, You will find some information in this web page. http://owen.sj.ca.us/rkowen/howto/FandC/ Hope it helps. pascal zaldini/at/ufpe.br wrote: > > Dear CCLers, > > I would like to find some good reference ou instruction about the > communication between program codes written in FORTRAN and C. > Thanks in advance. > > Prof. Marcelo Z. Hernandes > UFPE - Brazil > > -= 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) > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ -- *************************************************** Dr. Pascal Bonnet School of Pharmacy and Pharmaceutical Sciences, University of Manchester, Oxford Road, Manchester, M13 9PL, U.K Tel:(+44) (0)161 275 2431 http://www.man.ac.uk/~mbpsspb3