From chemistry-request@ccl.net Fri Jul 8 00:13:51 2005 Received: from chem.umn.edu (chemsun.chem.umn.edu [128.101.157.2]) by server.ccl.net (8.13.1/8.13.1) with SMTP id j684Dko5001958 for ; Fri, 8 Jul 2005 00:13:47 -0400 Received: (qmail 2101 invoked from network); 8 Jul 2005 03:13:44 -0000 Received: from unknown (HELO ?10.0.1.2?) (134.84.5.63) by 0 with SMTP; 8 Jul 2005 03:13:44 -0000 Mime-Version: 1.0 (Apple Message framework v622) Content-Type: multipart/alternative; boundary=Apple-Mail-1-751306894 Message-Id: <452bfca9f0024f1b3c7e931d0d8a8047 ## chem.umn.edu> Cc: CHEMISTRY ## ccl.net From: Christopher Cramer Subject: Fwd: CCL: dissociation problem in PCM model ? Date: Thu, 7 Jul 2005 22:13:47 -0500 To: Elmar Gerwalin X-Mailer: Apple Mail (2.622) X-Spam-Status: No, score=0.0 required=5.0 tests=none autolearn=failed version=3.0.4 X-Spam-Checker-Version: SpamAssassin 3.0.4 (2005-06-05) on server.ccl.net --Apple-Mail-1-751306894 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Elmar, My textbook, Essentials of Computational Chemistry, actually has a specific section (11.4.4 Potentials of Mean Force and Solvent Structure) that mentions your problem. The issue is that continuum models like the PCM variants involve the definition of a solute cavity, typically defined as roughly the union of a set of van der Waals spheres on all of the atoms. If you pull your solute apart, breaking a bond, at some point you will exceed van der Waals contact and the poor program has to decide what to do. It is quite unpleasant to solve the Poisson equation with two cavities in a dielectric continuum... If memory serves, PCM leaves some sort of neck between the two for awhile, so that there is still a single cavity, but this clearly soon becomes a bad approximation, since there is really solvent there. At some point, the program probably just delivers an error message (and a cryptic one, too, no doubt). But, if you compute the separated species individually, then there is no problem at all. Each has a well defined cavity and is infinitely far from anything but solvent. The two calculations will not agree, since the cavities were defined very differently. The only widely used continuum model that does NOT suffer from this discontinuity in a potential of mean force owing to cavity separation is the Generalized Born model, which takes atomic radii as inputs but does not use them to construct a formal cavity and is smooth and continuous over all distances. If you would like to try a GB model, you are welcome to employ one of our SMx solvation models, freely available in a variety of codes, including SMxGAUSS, which can either read a Gaussian output file as input and run as a stand-alone code, or can be driven by G03 using the "external" keyword. See comp.chem.umn.edu/mccdir/software.htm for full details. Best regards, Chris Begin forwarded message: > From: Elmar Gerwalin > Date: July 7, 2005 3:29:35 PM CDT > To: CHEMISTRY !! ccl.net > Subject: CCL: dissociation problem in PCM model ? > Reply-To: chemistry !! ccl.net > > Hi, > > I'm trying to describe dissociation processes in some small organic > molecules. > Therefore, I'm using B3LYP+SCRF(C-PCM, solvent=methanol) single point > jobs to get the > solvation energy at each step. Finally, a potential energy curve E(R) > is my goal. > R is the bond distance I'm interested in; e.g. a C-O, C-Cl bond or , > for testing purposes > Na-Cl. > > But, unfortunately, wrong asymptotic behaviour (like a "size > inconsistency problem") is > observed in all cases: the "supermolecule's" solvation energy at large > distances (10 or 20 A) > is differing about more than 20 kcal/mole from the sum of the > solvation energies of the two ions. > > My question is: is the problem known (references ?)? How can be it be > accounted for ? > Do I have to use specific energy terms given in gaussian's output > and/or male a correction to them ? > > Any help is appreciated. > > Bye, > > Yours, > Elmar > > > -- > ======================================================== > Elmar Gerwalin , University of Kaiserslautern,Germany > Dept. of Theoretical Chemistry > elg !! chemie.uni-kl.de > ======================================================== > > > -= This is automatically added to each message by the 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 is mail bouncing from ccl.net domain due to spam filters, > please > use the Web based form from CCL Home Page > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- > +-+ > > > > > > -- Christopher J. Cramer University of Minnesota Department of Chemistry 207 Pleasant St. SE Minneapolis, MN 55455-0431 -------------------------- Phone: (612) 624-0859 || FAX: (612) 626-2006 Mobile: (612) 597-5275 cramer !! pollux.chem.umn.edu http://pollux.chem.umn.edu/~cramer (website includes information about the textbook "Essentials of Computational Chemistry: Theories and Models, 2nd Edition") --Apple-Mail-1-751306894 Content-Transfer-Encoding: 7bit Content-Type: text/enriched; charset=US-ASCII Elmar, My textbook, Essentials of Computational Chemistry, actually has a specific section (11.4.4 Potentials of Mean Force and Solvent Structure) that mentions your problem. The issue is that continuum models like the PCM variants involve the definition of a solute cavity, typically defined as roughly the union of a set of van der Waals spheres on all of the atoms. If you pull your solute apart, breaking a bond, at some point you will exceed van der Waals contact and the poor program has to decide what to do. It is quite unpleasant to solve the Poisson equation with two cavities in a dielectric continuum... If memory serves, PCM leaves some sort of neck between the two for awhile, so that there is still a single cavity, but this clearly soon becomes a bad approximation, since there is really solvent there. At some point, the program probably just delivers an error message (and a cryptic one, too, no doubt). But, if you compute the separated species individually, then there is no problem at all. Each has a well defined cavity and is infinitely far from anything but solvent. The two calculations will not agree, since the cavities were defined very differently. The only widely used continuum model that does NOT suffer from this discontinuity in a potential of mean force owing to cavity separation is the Generalized Born model, which takes atomic radii as inputs but does not use them to construct a formal cavity and is smooth and continuous over all distances. If you would like to try a GB model, you are welcome to employ one of our SMx solvation models, freely available in a variety of codes, including SMxGAUSS, which can either read a Gaussian output file as input and run as a stand-alone code, or can be driven by G03 using the "external" keyword. See comp.chem.umn.edu/mccdir/software.htm for full details. Best regards, Chris Begin forwarded message: 0000,0000,0000From: Elmar Gerwalin < 0000,0000,0000Date: July 7, 2005 3:29:35 PM CDT 0000,0000,0000To: CHEMISTRY !! ccl.net 0000,0000,0000Subject: CCL: dissociation problem in PCM model ? 0000,0000,0000Reply-To: chemistry !! ccl.net Hi, I'm trying to describe dissociation processes in some small organic molecules. Therefore, I'm using B3LYP+SCRF(C-PCM, solvent=methanol) single point jobs to get the solvation energy at each step. Finally, a potential energy curve E(R) is my goal. R is the bond distance I'm interested in; e.g. a C-O, C-Cl bond or , for testing purposes Na-Cl. But, unfortunately, wrong asymptotic behaviour (like a "size inconsistency problem") is observed in all cases: the "supermolecule's" solvation energy at large distances (10 or 20 A) is differing about more than 20 kcal/mole from the sum of the solvation energies of the two ions. My question is: is the problem known (references ?)? How can be it be accounted for ? Do I have to use specific energy terms given in gaussian's output and/or male a correction to them ? Any help is appreciated. Bye, Yours, Elmar -- ======================================================== Elmar Gerwalin , University of Kaiserslautern,Germany Dept. of Theoretical Chemistry elg !! chemie.uni-kl.de ======================================================== -= This is automatically added to each message by the 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 is mail bouncing from ccl.net domain due to spam filters, please use the Web based form from CCL Home Page -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ -- Christopher J. Cramer University of Minnesota Department of Chemistry 207 Pleasant St. SE Minneapolis, MN 55455-0431 -------------------------- Phone: (612) 624-0859 || FAX: (612) 626-2006 Mobile: (612) 597-5275 cramer !! pollux.chem.umn.edu http://pollux.chem.umn.edu/~cramer (website includes information about the textbook "Essentials of Computational Chemistry: Theories and Models, 2nd Edition") --Apple-Mail-1-751306894--