From qoajnv@usc.es Sun Jan 11 12:14:22 1998 Received: from uscmail.usc.es for qoajnv@usc.es by www.ccl.net (8.8.3/950822.1) id LAA14986; Sun, 11 Jan 1998 11:59:38 -0500 (EST) Received: from localhost (qoajnv@localhost) by uscmail.usc.es (8.8.5/8.8.5) with SMTP id RAA11760 for ; Sun, 11 Jan 1998 17:59:37 +0100 (MET) Date: Sun, 11 Jan 1998 17:59:37 +0100 (MET) From: Armando Juan Navarro Vazquez To: chemistry@www.ccl.net Subject: summary:a question on irc Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII My original question on irc was: "I have tried to perform a irc calculation on a system with a very flat potential surface, using the b3lyp functional in g94, but it could be only completed sucessfully on one side of the potential,however only one point is located in the other side (program considers is as a minimum). How can I calculate further points on this side of the potential?." I have got the following answers, thanks to all these persons: From: Dmitry Khoroshun To: qoajnv@uscmail.usc.es Subject: IRC Dear Armando! Some suggestions: 1. Optimize the transition structure with VeryTight criteria. Use the exact Hessian matrix you are supposed to have now for that optimization. Use the same criteria (verytight) for the subsequent IRC run. and/or, 2. Increase the stepsize for the IRC search. The underlying problem is that your "TS" is a little bit deviated from the ideal TS for that method. Each reaction path point for the IRC method is defined as a minima on a sphrere of some radius. Due to the shallow nature of the surface, small deviation from the TS in terms of forces and displacements obtained using updated (approximate) Hessian satisfy usual criteria. However, the point with the lowest energy on the sphere with the center at your TS still has gradients small enough (even the one corresponding to the gradient along the reaction coordinate) to be considered a stationary point. Hence, eigher yo should tighten the criteria, or you shoould increase the stepsize. First approach is more accurate. In difficult cases, one needs to combine both approaches. Hope that helps. Sincerely, Dmitry Khoroshun ------------------------------------------------------------- Dear Armando, You may try to reduce the step size on your IRC calculation. Including the option StepSize=N would suffice. N is the setp size along the reaction path in the desired direction in units of 0.01 au. Hope this helps. Regards, Albeiro. ########## ( o o ) ( -- ) +-----------------------------oooO------Oooo--------------------------+ |Albeiro Restrepo | | |Department of Chemistry | | |Rice University |e-mail: albeiro@rice.edu | |6100 Main Street |Fax : (713) 285 51 55 | |Houston, Texas, 77005-1892, USA |voice : (713) 527 40 82 | +------------------------------------+--------------------------------+ | | | | (___) (___) ------------------------------------------------------------- Dear Armando, Try bigger step size. This often helps to find points on the reaction coordinate which are indeed lower in energy than the TS; once you're going downhill, further IRC following becomes easier. I've bumbed into this situations myself several times, and this was the solution to at least my problems. Don't make the stepsize TOO big, as this might mean that you lose track of the IRC, and start to wander about on the potential energy surface. Finally, several questions and answers on problems with IRC are available in the CCL archives, which -with the new web search engine- is a really easy to use tool. Best regards, Han _____________________________________________________________________ Han Zuilhof | E-mail: ZUILHOF@SG1.OC.WAU.NL Laboratory of Organic Chemistry | FAX : 31-317-484914 Department of Biomolecular Sciences | phone : 31-317-482367 Wageningen Agricultural University | Dreijenplein 8 | 6703 HB Wageningen | "Excite a photochemist!" The Netherlands | ------------------------------------------------------------- dear Armando I also had this problem o short time ago (thanks to udo for solving the problem). It might be possible that the first point is not far away enough from the transition state so that the convergence criteria seems to be already satisfied. Thus , try to increase the stepsize of your irc run. Doing this you come to a geometry in which the convergence criteria are not satisfied and irc then goes down the hill until they are fullfilled again. Hope this helps Christoph Widauer ----- Christoph Widauer Laboratorium f. Anorganische Chemie ETH-Zentrum CAB C8 Universittsstrasse 6 CH-8092 Zuerich Phone (Zurich): +41 (1) 632 61 19 Fax (Zurich): +41 (1) 632 10 90 e-mail: widauer@inorg.chem.ethz.ch -------------------------------------------------------------- Armando Navarro Departamento de quimica Organica. Facultade de quimica. Universidade de Santiago de Compostela. e-mail:qoajnv@uscmail.usc.es From leclerc@tammy.harvard.edu Sun Jan 11 18:14:25 1998 Received: from tammy.harvard.edu for leclerc@tammy.harvard.edu by www.ccl.net (8.8.3/950822.1) id RAA15800; Sun, 11 Jan 1998 17:40:21 -0500 (EST) Received: from kei.harvard.edu (kei [128.103.92.153]) by tammy.harvard.edu (8.8.2/8.8.2) with ESMTP id RAA29451; Sun, 11 Jan 1998 17:46:22 -0500 (EST) Received: (from leclerc@localhost) by kei.harvard.edu (8.8.2/8.8.2) id RAA20615; Sun, 11 Jan 1998 17:36:55 -0500 (EST) Date: Sun, 11 Jan 1998 17:36:55 -0500 (EST) From: Fabrice Leclerc Message-Id: <199801112236.RAA20615@kei.harvard.edu> To: chemistry@www.ccl.net Subject: 2',3' cyclic phosphate Hi Netters, My current research involves RNA catalysis in particular ribozyme-mediated cleavage. In some ribozymes, the reaction mechanism involves the formation of nucleotides with a 2',3' cyclic phosphate terminus. I'm looking for some structural data on this particular type of cyclic nucleotide (NMR or X-ray). -- \ /\ \/ \______________________________________ /\ A--T Fabrice Leclerc, Ph.D. (----) Departement of Chemistry G--C and Chemical Biology \/ Harvard University /\ 12 Oxford Street - 212 T--A Cambridge, MA 02138 USA. (----) Tel. (617) 495-1768 C--G Fax. (617) 496-3204 \/ leclerc@tammy.harvard.edu /\ _____________________________________ / \ / \/