From maiden@RedBrick.DCU.IE Sat Mar 15 12:29:30 1997 Received: from tolka.dcu.ie for maiden@RedBrick.DCU.IE by www.ccl.net (8.8.3/950822.1) id MAA27440; Sat, 15 Mar 1997 12:08:40 -0500 (EST) Received: from Nurse.RedBrick.DCU.IE by tolka.dcu.ie; (5.65v3.2/1.1.8.2/14Feb96-0535PM) id AA26873; Sat, 15 Mar 1997 17:08:36 GMT Received: (from maiden@localhost) by RedBrick.DCU.IE (SomeMTA) id RAA17220; Sat, 15 Mar 1997 17:06:52 GMT Organisation: DCU Networking Society X-Disclaimer: The DCUNS is not responsible for the content of this message Date: Sat, 15 Mar 1997 17:06:52 +0000 (GMT) From: michael nolan To: chemistry@www.ccl.net Subject: Z-matrix query...summary Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi Here's a summary of the replies to my Z-matrix query... michael The imp himself has spoken *************************************************************************************** Michael Nolan (nearly BSc.) Dublin City University (quantum chem.) 41 Woodview Chemistry + german Year 4 Lucan Co. Dublin Ireland / Republic of Ireland / ROI / Eire Email: Maiden@redbrick.dcu.ie **************************************************************************************** original question: > > I am working on manganese tricarbonyl comlexes using G94..... > When inputting the z-matrix what is the form of it for the metal to > arene ring distnace {and angles}? Do I use a dummy atom or do I use > the metal to centroid distance.... > > > any answers will be gratefully recieved.. > > ***************************************************************************** Answer 1: From gaussian.com!fox@lorentzian.com Sun Mar 2 16:42:57 1997 Date: Wed, 19 Feb 1997 12:22:39 -0500 (EST) From: Doug Fox To: Mick the imp Subject: Re: CCL:G:query....Z-matrix Mr. Nolan, The simplest option is to input cartesian coordinates and let G94 generate a set of redundant internal coordinates. It will actually use the set of all Metal-ring bonds rather than either the centroid or a dummy atom. If you do want to generate a Z-matrix for some other purpose the use of a dummy atom is the easiest to control. Douglas J. Fox Director of Technical Support help@gaussian.com ****************************************************************************** Answer 2: From chpajt@bath.ac.uk Sun Mar 2 16:50:23 1997 Date: Sat, 15 Feb 1997 13:23:53 +0000 (GMT) From: A J Turner To: "Mick the imp (not my account!)" Subject: Re: CCL:G:query....Z-matrix Hi! Unless you use opt=zmat it does not matter as the algorithum will covert >from z => carts and then to redundant internals. It make no difference to the minimisation if you put in dummies or not - as long as the structure you come up with puts all the chemical atoms in the correct place. I am not so sure about the idea that the redundant internal optimiser is OK for TS searches. I would still go for an EF search in cartesian coordinates. See papers by Jorgensen and/or J Baker. Best wishes Alex **************************************************************************** From maiden@RedBrick.DCU.IE Sat Mar 15 13:15:19 1997 Received: from tolka.dcu.ie for maiden@RedBrick.DCU.IE by www.ccl.net (8.8.3/950822.1) id MAA27415; Sat, 15 Mar 1997 12:01:31 -0500 (EST) Received: from Nurse.RedBrick.DCU.IE by tolka.dcu.ie; (5.65v3.2/1.1.8.2/14Feb96-0535PM) id AA24932; Sat, 15 Mar 1997 17:01:27 GMT Received: (from maiden@localhost) by RedBrick.DCU.IE (SomeMTA) id QAA16887; Sat, 15 Mar 1997 16:59:42 GMT Organisation: DCU Networking Society X-Disclaimer: The DCUNS is not responsible for the content of this message Date: Sat, 15 Mar 1997 16:59:42 +0000 (GMT) From: michael nolan To: chemistry@www.ccl.net Subject: Tansition states....summary Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi CCL... Thanks to the replies for my (amended) query on transition states... the answers are summarised below michael The imp himself has spoken *************************************************************************************** Michael Nolan (nearly BSc.) Dublin City University (quantum chem.) 41 Woodview Chemistry + german Year 4 Lucan Co. Dublin Ireland / Republic of Ireland / ROI / Eire Email: Maiden@redbrick.dcu.ie **************************************************************************************** Original question: >hello CCL >It seems that my question on trnasition states and PES was too vague... >What I want to know is the following: >I am reading a paper from Merrill, Gronert and Kass J. Phys. Chem. A >1997, 101, 208-218. >In this paper they use DFT to study the PES of a base induced elimination. >My question is what would be the input for such a calculation {on >Gaussian92/DFT}? Is the input the same for an ab initio calculation? >michael ****************************************************************************** Answer 1: From rbw@msc.eduFri Mar 14 08:58:31 1997 Date: Wed, 5 Mar 1997 07:55:34 -0600 (CST) From: Richard Walsh To: maiden@RedBrick.DCU.IE Subject: Re: CCL:G:Transition states... Michael, One set of Gaussian 94 keywords that would perform a DFT transition search would be: %mem=8000000 %chk=dft_tstate.chk #p b3lyp/cc-pVDZ opt=(ef,ts,z-matrix) scf=direct This assumes that you have a starting geometry which has a PES with the right curvature and your input file is in z-matrix format. The calculation is performed by recomputing the integrals; the eigenvector following algorithm for the PES surface search is used. Memory requested is 8M words. Checkpoint data goes to the file dft_tstate.chk. the '#p' forces printing of link related entry and exit information. Richard Walsh **************************************************************************** answer 2: From kass@chemsun.chem.umn.edu Fri Mar 14 08:58:51 1997 Date: Wed, 5 Mar 1997 12:26:54 -0600 (CST) From: Steven Kass To: maiden@RedBrick.DCU.IE Subject: CCL:G:Transition states... (fwd) Michael, With regard to your question below, the simple answer is yes! To run a DFT calculation with G92/DFT one simply specifies the method (BVWN, BLYP, B3LYP, etc.), a basis set (6-31+G(d,p), aug-cc-pVDZ, etc.), and of course a starting geometry (in the case of the base induced eliminations some of these are transition states so one will need to specify opt=ts instead of just opt, but this is equally true for ab initio methods). Hope this helps. Steven **************************************************************************** Answer 3: >From mlebret@igr.fr Wed Mar 5 05:33 EST 1997 Received: from fw.igr.fr for mlebret@igr.fr by www.ccl.net (8.8.3/950822.1) id FAA03700; Wed, 5 Mar 1997 05:33:45 -05 00 (EST) Received: by fw.igr.fr; (5.65v3.2/1.3/10May95) id AA01447; Wed, 5 Mar 1997 12:34:45 +0100 Date: Wed, 05 Mar 1997 11:25:04 +0100 From: mlebret@igr.fr (Marc Le Bret) Subject: Re: CCL:G:Transition states... yes, just substitute MP2 by B3LYP for instance. ****************************************************************** * Marc LE BRET *Laboratoire de Physicochimie et de Pharmacologie* * *des macromolecules biologiques CNRS URA147 * * *rue Camille Desmoulins F-94805 Villejuif Cedex * * mlebret@igr.fr* Tel (33 1) 42 11 49 79 FAX (33 1) 42 11 52 76 * ****************************************************************** ****************************************************************************** Answer 4: From gaussian.com!fox@lorentzian.comFri Mar 14 08:59:26 1997 Date: Fri, 7 Mar 1997 09:51:48 -0500 (EST) From: Doug Fox To: Mick the imp Subject: Re: CCL:G:Transition states... Michael, With G92/DFT the input for finding transition states(TS) starts with an input structure which approximates the TS and then use OPT=TS or in most cases OPT=(CALCFC,TS) to optimize the structure forcing a maximum in one direction. The only difference between DFT and ab initio methods for this would be the keyword selecting the method, i.e. HF vs BLYP. Douglas J. Fox Director of Technical Support help@gaussian.com ***************************************************************************** From felllm@mcmail.CIS.McMaster.CA Sat Mar 15 13:29:32 1997 Received: from mcmail.CIS.McMaster.CA for felllm@mcmail.CIS.McMaster.CA by www.ccl.net (8.8.3/950822.1) id NAA27667; Sat, 15 Mar 1997 13:13:15 -0500 (EST) Received: (from felllm@localhost) by mcmail.CIS.McMaster.CA (8.7.1/8.7.1) id NAA22070; Sat, 15 Mar 1997 13:13:13 -0500 (EST) Date: Sat, 15 Mar 1997 13:13:13 -0500 (EST) From: "Lorne M. Fell" To: chemistry@www.ccl.net Subject: Parallel G94/GAMESS Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Members of CCL: Our department is very interested in seeing benchmarks of parallel code of either G94 or GAMESS using Linux (or BSD etc...). If anyone has data on this subject, please email the list or contact me directly. A summary will be posted. Thank you ============================================================================ Lorne M. Fell "Argue your limitations and sure enough Department of Chemistry they are yours." McMaster University R. Bach Hamilton, Ont., Canada phone (905) 525-9140 x.23146 email: felllm@mcmail.cis.mcmaster.ca ============================================================================ From maiden@RedBrick.DCU.IE Sat Mar 15 13:36:50 1997 Received: from tolka.dcu.ie for maiden@RedBrick.DCU.IE by www.ccl.net (8.8.3/950822.1) id MAA27421; Sat, 15 Mar 1997 12:02:19 -0500 (EST) Received: from Nurse.RedBrick.DCU.IE by tolka.dcu.ie; (5.65v3.2/1.1.8.2/14Feb96-0535PM) id AA25208; Sat, 15 Mar 1997 17:02:16 GMT Received: (from maiden@localhost) by RedBrick.DCU.IE (SomeMTA) id RAA16905; Sat, 15 Mar 1997 17:00:32 GMT Organisation: DCU Networking Society X-Disclaimer: The DCUNS is not responsible for the content of this message Date: Sat, 15 Mar 1997 17:00:32 +0000 (GMT) From: michael nolan To: chemistry@www.ccl.net Subject: Tansition states....summary Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hi CCL... Thanks to the replies for my (amended) query on transition states... the answers are summarised below michael The imp himself has spoken *************************************************************************************** Michael Nolan (nearly BSc.) Dublin City University (quantum chem.) 41 Woodview Chemistry + german Year 4 Lucan Co. Dublin Ireland / Republic of Ireland / ROI / Eire Email: Maiden@redbrick.dcu.ie **************************************************************************************** Original question: >hello CCL >It seems that my question on trnasition states and PES was too vague... >What I want to know is the following: >I am reading a paper from Merrill, Gronert and Kass J. Phys. Chem. A >1997, 101, 208-218. >In this paper they use DFT to study the PES of a base induced elimination. >My question is what would be the input for such a calculation {on >Gaussian92/DFT}? Is the input the same for an ab initio calculation? >michael ****************************************************************************** Answer 1: From rbw@msc.eduFri Mar 14 08:58:31 1997 Date: Wed, 5 Mar 1997 07:55:34 -0600 (CST) From: Richard Walsh To: maiden@RedBrick.DCU.IE Subject: Re: CCL:G:Transition states... Michael, One set of Gaussian 94 keywords that would perform a DFT transition search would be: %mem=8000000 %chk=dft_tstate.chk #p b3lyp/cc-pVDZ opt=(ef,ts,z-matrix) scf=direct This assumes that you have a starting geometry which has a PES with the right curvature and your input file is in z-matrix format. The calculation is performed by recomputing the integrals; the eigenvector following algorithm for the PES surface search is used. Memory requested is 8M words. Checkpoint data goes to the file dft_tstate.chk. the '#p' forces printing of link related entry and exit information. Richard Walsh **************************************************************************** answer 2: >From kass@chemsun.chem.umn.edu Fri Mar 14 08:58:51 1997 Date: Wed, 5 Mar 1997 12:26:54 -0600 (CST) From: Steven Kass To: maiden@RedBrick.DCU.IE Subject: CCL:G:Transition states... (fwd) Michael, With regard to your question below, the simple answer is yes! To run a DFT calculation with G92/DFT one simply specifies the method (BVWN, BLYP, B3LYP, etc.), a basis set (6-31+G(d,p), aug-cc-pVDZ, etc.), and of course a starting geometry (in the case of the base induced eliminations some of these are transition states so one will need to specify opt=ts instead of just opt, but this is equally true for ab initio methods). Hope this helps. Steven **************************************************************************** Answer 3: >From mlebret@igr.fr Wed Mar 5 05:33 EST 1997 Received: from fw.igr.fr for mlebret@igr.fr by www.ccl.net (8.8.3/950822.1) id FAA03700; Wed, 5 Mar 1997 05:33:45 -05 00 (EST) Received: by fw.igr.fr; (5.65v3.2/1.3/10May95) id AA01447; Wed, 5 Mar 1997 12:34:45 +0100 Date: Wed, 05 Mar 1997 11:25:04 +0100 From: mlebret@igr.fr (Marc Le Bret) Subject: Re: CCL:G:Transition states... yes, just substitute MP2 by B3LYP for instance. ****************************************************************** * Marc LE BRET *Laboratoire de Physicochimie et de Pharmacologie* * *des macromolecules biologiques CNRS URA147 * * *rue Camille Desmoulins F-94805 Villejuif Cedex * * mlebret@igr.fr* Tel (33 1) 42 11 49 79 FAX (33 1) 42 11 52 76 * ****************************************************************** ****************************************************************************** Answer 4: From gaussian.com!fox@lorentzian.comFri Mar 14 08:59:26 1997 Date: Fri, 7 Mar 1997 09:51:48 -0500 (EST) From: Doug Fox To: Mick the imp Subject: Re: CCL:G:Transition states... Michael, With G92/DFT the input for finding transition states(TS) starts with an input structure which approximates the TS and then use OPT=TS or in most cases OPT=(CALCFC,TS) to optimize the structure forcing a maximum in one direction. The only difference between DFT and ab initio methods for this would be the keyword selecting the method, i.e. HF vs BLYP. Douglas J. Fox Director of Technical Support help@gaussian.com *****************************************************************************