From owner-chemistry@ccl.net Thu Dec 19 10:00:00 2013 From: "Kai Leonhard kai.leonhard~!~ltt.rwth-aachen.de" To: CCL Subject: CCL: AW: BSSE in transition states Message-Id: <-49460-131219053810-24727-yrRxjbvJtcRwTJeHponNoQ,,server.ccl.net> X-Original-From: Kai Leonhard Content-Language: de-DE Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Thu, 19 Dec 2013 10:37:58 +0000 MIME-Version: 1.0 Sent to CCL by: Kai Leonhard [kai.leonhard-,-ltt.rwth-aachen.de] Dear all, The application of CPC to TSs seems to be quite controversial to me. There are other Papers that claim that applying the CPC to TSs is better than not doing so (or even the only acceptable way). E.g. Kobko and Dannenberg claim that 'Clearly, CP-corrected surfaces lead to noticeable improvements in the geometrically optimized TSs obtained with relatively small basis sets. The geometries and associated activation energies become closer to those obtained with larger basis sets. The CP corrections to these parameters tend to diminish as the basis sets are improved. These observations may be contrasted with the earlier suggestion by Lendvay and Mayer that CP correction is inappropriate for the optimizations of TSs.[18] They quite correctly noted that certain choices for the fragments used in the CP correction calculation would be inappropriate. In particular, they criticized two applications of CP to TSs.[21] For example, had we used two fragments (CH4 and CH3*) for the methane/methyl H-transfer reaction, we would have obtained an inappropriate, unsymmetrical TS. However, a flexible choice of the fragments will usually allow reasonable CP calculations of the TS, at least for cases where all fragments are neutral. Nevertheless, the seemingly arbitrary nature of the fragment choices bear some detailed discussion.' My impression is that maybe Lendvay and Mayer wrote their paper mainly to motivate their CHA method because they state that CPC is not perfect and that a better method is desirable. This is true, but applying CPC seems better to me than completely ignoring BSSE. More recently, it was discussed by Richard et al. that intramolecular BSSE is even relevant for stable molecules. They suggest a 3:1 weighted average of CPC and non-cpc results as best energy estimates (for cluster interactions, though). My question is now whether the debate can be considered resolved now or it is still an ongoing discussion. Any thoughts or experiences on CPC for TSs are highly appreciated. Best regards, Kai Prof. Dr. Kai Leonhard Juniorprofessor for Model-Based Fuel Design Room 215 Schinkelstraße 8 D-52056 Aachen Phone: +49 (0)241 80 - 98174 Fax: +49 (0)241 80 - 92255 E-Mail: kai.leonhard|-|ltt.rwth-aachen.de Kobko, N. & Dannenberg, J. J. Effect of basis set superposition error (BSSE) upon ab initio calculations of organic transition states J. Phys. Chem. A, {2001}, {105}, {1944-1950} Richard, R. M.; Lao, K. U. & Herbert, J. M. Approaching the Complete-Basis Limit with a Truncated Many-Body Expansion J. Chem. Phys., 2013, 139, 224102 -----Ursprüngliche Nachricht----- Von: owner-chemistry+leonhard==ltt.rwth-aachen.de|-|ccl.net [mailto:owner-chemistry+leonhard==ltt.rwth-aachen.de|-|ccl.net] Im Auftrag von Florent LOUIS florent.louis(-)univ-lille1.fr Gesendet: Mittwoch, 11. Dezember 2013 18:11 An: Kai Leonhard Betreff: CCL: BSSE in transition states Sent to CCL by: "Florent LOUIS" [florent.louis~~univ-lille1.fr] Dear John, The BSSE correction with the counterpoise method is applicable to weakly interacting molecular complexes but not to TS, where it can lead to discontinuous potential surfaces (see paper from Lendvay and Mayer, Some Difficulties in Computing BSSE Corrected Potential Surfaces of Chemical Reactions. Chem. Phys. Lett.1998, 297 (56), 365373) Therefore, the CP correction can be just viewed as an estimate of the error bars for reaction barriers. Best regards, Florent > "John Simmie john.simmie]_[nuigalway.ie" wrote: > > Sent to CCL by: "John Simmie" [john.simmie .. nuigalway.ie] I have > recently seen quite large BSSE corrections of ca 20 kJ/mol claimed for > a TS (C, H, O only) for coupled cluster calculations with basis set > aug-cc-pVDZ for zero-point corrected electronic energy barriers of ca > 90 kJ/mol The correction decreases to ca 10 kJ/mol with aug-cc-pVTZ > > is this well-known and/or documented somewhere please? > > John M. Simmie//NUI Galway//Irelandhttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt