From owner-chemistry@ccl.net Fri Jul 21 00:26:20 1995 Received: for jkl@ccl.net by www.ccl.net (8.6.10/930601.1506) id AAA17813; Fri, 21 Jul 1995 00:25:12 -0400 Date: Fri, 21 Jul 1995 00:25:12 -0400 From: Jan Labanowski Message-Id: <199507210425.AAA17813@www.ccl.net> To: chemistry@ccl.net Subject: Coordinator message to all of you Cc: jkl@ccl.net > I am trying to find out the email address for Prof. Janos Hajdu > (a protein crystallographer) at the Laboratory of Molecular Biophysics, > University of Oxford. I have tried the usual channels but can't seem to > find anything. Can anyone help me? This message prompted me to remind all of you that I stile live (for what it is worth) and I watch you... If someone brakes the rules, his/her e-mail address ends up in my reject file. And the next message from this person will not appear on the list automatically. Before the message is posted, it needs to be reviewed by me. And since I am busy (or pretending to be), your message may sit there and wait for me for a long time... Address inquires are not appropriate for this list. This is against the rules of the list. Rules ("help file") can be retrieved by sending a message consisting of a single line: help chemistry to MAILSERV@ccl.net If you want to find an e-mail address, read files in the directory /pub/chemistry/info/e-mail-addresses in CCL archives (how to access CCL archives is written in the "help file"), but basically archives can be reached as: via Web: http://www.ccl.net/chemistry.html via gopher gopher.ccl.net 73 via ftp ftp.ccl.net via e-mail by sending requests to MAILSERV@ccl.net (described in "help file"). This info is in the footnote of each file you receive from CCL. Now... For larger organizations it is easy to find e-mail addresses via their gopher servers which provide phone/e-mail/address searchable directories. You start your gopher and click. In this case: All gopher servers of the world Europe Oxford University Top Level (gopher.ox.ac.uk port 70) Phone and e-mail Search Oxford Email Directory (entered: hajdu) Hajdu Janos CHRI janos.hajdu@christ-church.oxford.ac.uk Hajdu Janos MOLBIO bioplaue@vax.ox.ac.uk In this case you can also use Web (usually main entry points for Universities or larger organizations have a link to a directory of the personel). In our case: netscape http://www.ox.ac.uk/ (or whatever Web browser you use:) Phone and e-mail Email and phone details for individuals Who do you want: Hajdu Janos Hajdu janos.hajdu@christ-church.oxford.ac.uk (CHRI) However, when you loose hope, you can always contact me, and I will try to help if I can. BUT PLEASE, try to keep this list a high quality discussion forum, rather than a noisy, unfocused, junk-mail feed. So please, no questions or messages like: What is Mr. Novak address... Where can I find the density of water at 4 deg C? What is ab initio, and why people use it? What happened to the Barbie which did not like math? Do you have a good job for a smart guy like me? End of the world is coming... Are you prepared? No nukes... Vote for Citizen Kane... etc... If you have reasons to believe that nobody but you cares about the answer to your questions, please, do not ask the question here. Ask me directly, maybe I can help. PLEASE remember about the option to search archives. Send the message: help search to MAILSERV@ccl.net to find out how to search CCL archives. Many questions were answered before, and even if the update is needed, it is always good to know, what people said about it before. Rules are in the help file, and let us stick to them. Some rules (like allowing for short commercial adds) are a compromise, but this is what it is all about, isn't it? And someone has to take a blame... So flame me, or the abuser, but please, do not send non-CompChem messages to the list. Please... And one more thing... Please do not start the discussion about what the CCL list should be. But as always, I will gladly listen to your suggestions (actually, I am really thankful for them, since it helps me in keeping this forum alive -- many of you contributed lots of good ideas to the new proposal, which I am now resubmitting). Jan Labanowski -- still going strong... jkl@ccl.net -- Dr. Jan K. Labanowski, Senior Research/Supercomputer Scientist/Specialist, etc. Ohio Supercomputer Center, 1224 Kinnear Rd, Columbus, OH 43212-1163 ph:(614)-292-9279, FAX:(614)-292-7168, E-mail: jkl@ccl.net JKL@OHSTPY.BITNET From owner-chemistry@ccl.net Fri Jul 21 04:26:26 1995 Received: from rs1.rrz.Uni-Koeln.DE for acp37@rs1.rrz.Uni-Koeln.DE by www.ccl.net (8.6.10/930601.1506) id EAA20583; Fri, 21 Jul 1995 04:11:45 -0400 Received: by rs1.rrz.Uni-Koeln.DE id AA64412 (5.67b/IDA-1.5 for CCL ); Fri, 21 Jul 1995 10:11:26 +0200 Date: Fri, 21 Jul 1995 10:11:26 +0200 (MST) From: Thorsten Koch To: CCL Subject: LST in G94 Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Dear Netters! I have a problem using the LST-keyword in G94. What I did is the following: I optimized a proper structure for reactand and product. For further use in a LST-calculation I made sure that corresponding atoms appear in the same order. I used these structures as an input for the LST-calculation. What I got was: 1. Gaussian was, all of a sudden, not able to determin orbital symmetries and electronic state of reactand and product. It says something like "partially filled degenerate orbitals" (In the previous geometry optimisation it had no difficulties) 2. Gaussian didn't assign all atoms of the reactand to the right corrosponding atoms of the product, it exchanged two atoms - which results in an absolutely silly guess of the transition structure. 3. Finally, after the program attempts to scan the LST path, it aborts with an fortran error message: floating invalid. So my questions are: Are there any reasons for the difficulties handling orbital symmetry and electronic state? Why doesn't Gaussian map the atoms of reactands and products onto another in the order given in the z-matrix? Ok, I checked it - the atoms are really in the order I want them to have. Why does the program abort when I want it to scan the LST-path? Any hints would be appreciated. Many thanks, Thorsten Koch /-----------------------------------------------------------------\ | Thorsten Koch | | Institut fuer physikalische Chemie II der Universitaet zu Koeln | | acp37@rrz.uni-koeln.de | | Tel. +49 [0]221 470 4816 | \-----------------------------------------------------------------/ From owner-chemistry@ccl.net Fri Jul 21 04:41:28 1995 Received: from rzusuntk.unizh.ch for toukie@zui.unizh.ch by www.ccl.net (8.6.10/930601.1506) id EAA20872; Fri, 21 Jul 1995 04:33:52 -0400 Received: by rzusuntk.unizh.ch (4.1/SMI-4.1.9) id AA20437; Fri, 21 Jul 95 10:33:44 +0200 X-Nupop-Charset: Swiss Date: Fri, 21 Jul 1995 10:33:43 +0100 (MET) From: "Hr Dr. S. Shapiro" Sender: toukie@zui.unizh.ch Reply-To: toukie@zui.unizh.ch Message-Id: <38023.toukie@zui.unizh.ch> To: chemistry@ccl.net, toukie@zui.unizh.ch Subject: Question about charged micellar spheres Dear Colleagues; A question has arisen for which I am seeking a definitive answer. Suppose there are two sets of positively-charged spherical micelles one "big" and one "small", and these micelles are capable of binding to a surface. It stands to reason that when two such positively-charged micelles bind to a surface they will not come right up against one another because they will be repulsed by their mutual positively-charged surfaces. Now, if the _density_ of charges on the surface of the spheres is the same (i.e., both the big micelles and the small micelles have the same number of positive charges per unit surface area), will two small micelles be able to sit closer to one another than two large micelles, i.e., will there be greater repulsive forces between two large micelles resulting in their sitting further away >from one another than between the two smaller micelles? In addition to a direct answer, any references bearing on this problem would be appreciated. If you could explain the answer to this problem in a straightforward mathematical way, I would be most grateful. Thanks in advance to all responders. Sincerely, S. Shapiro ZH toukie@zui.unizh.ch From owner-chemistry@ccl.net Fri Jul 21 05:11:28 1995 Received: from voimax.voima.jkl.fi for eloranta@voimax.voima.jkl.fi by www.ccl.net (8.6.10/930601.1506) id EAA21098; Fri, 21 Jul 1995 04:59:48 -0400 Received: (from eloranta@localhost) by voimax.voima.jkl.fi (8.6.10/8.6.9) id LAA03806 for chemistry@ccl.net; Fri, 21 Jul 1995 11:59:43 +0300 From: Jussi Eloranta Message-Id: <199507210859.LAA03806@voimax.voima.jkl.fi> Subject: Saddle point optimization To: chemistry@ccl.net Date: Fri, 21 Jul 1995 11:59:42 +0300 (EET DST) X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 502 Hi, Is there a way to optimize to the geometry which is a minimum with respect to all variables except two for which it should be a maximum (ie. saddle point of order of 2). These two variables are some known Z-matrix variables. I've looked at saddle, ts, and ef params for opt but they don't seem to help with this. BTW would this be as simple as choosing "-Grad" direction for all variables but these two for which it would be "+Grad" ? Regards, Jussi Eloranta PS. I'm using Gaussian 92/DFT From owner-chemistry@ccl.net Fri Jul 21 07:58:00 1995 Received: from vax3.sara.nl for mirko@SARA.NL by www.ccl.net (8.6.10/930601.1506) id HAA22942; Fri, 21 Jul 1995 07:50:27 -0400 Received: from horus.sara.nl by SARA.NL (PMDF V4.2-15 #2498) id <01HT4UGNQNM8AM3KNO@SARA.NL>; Fri, 21 Jul 1995 13:53:52 +0200 (MET-DST) Received: from hk-pc1.chem.uva.nl by horus.sara.nl (AIX 3.2/UCB 5.64/4.03) id AA46922; Fri, 21 Jul 1995 13:49:21 +0200 Date: Fri, 21 Jul 1995 13:49:21 +0200 From: mirko@SARA.NL (Mirko Kranenburg) Subject: Pi-stacking in MOPAC To: chemistry@ccl.net Message-id: X-Envelope-to: chemistry@ccl.net MIME-version: 1.0 Content-type: text/plain; charset="us-ascii" Content-transfer-encoding: 7BIT X-Sender: a412knsm@horus.sara.nl Dear Netters, Several people asked me to summarize the answers to the question: >Dear Netters, > >Currently I am performing geometry optimizations on compounds in which >pi-stacking has an important contribution to the overall geometry. >The results from MOPAC 6 calculations (AM1, PM3 and MNDO) show no pi- >stacking whatsoever, even when a X-ray structure was used as input. > >My question: does MOPAC take pi-stacking into account, and are there any >relevant references concerning this problem. In addition I found one other relevant reference: Kurita, Y.; Takayama, C.; Tanaka, S. J. Comput. Chem. 1994, 15, 1013-1018. I would like to thank everyone who replied! Here are the answers: From: lrbu00@xd88.kodak.com Subject: Re: CCL:pi-stacking in MOPAC? In-Reply-To: mirko@sara.nl (Mirko Kranenburg) "CCL:pi-stacking in MOPAC?" (Jul 13, 12:25pm) To: mirko@SARA.NL (Mirko Kranenburg) Message-Id: <9507130903.ZM14423@xd88.kodak.com> X-Envelope-To: a412knsm@horus.sara.nl Mime-Version: 1.0 X-Mailer: Z-Mail (3.0.0 15dec93) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7BIT References: I know of no references, but I can assure you thet no VdW systems were involved in the parameterization of MNDO,AM1, and PM3.... Regards, John -- John M. McKelvey email: mckelvey@Kodak.COM Computational Science Laboratory phone: (716) 477-3335 2nd Floor, Bldg 83, RL Eastman Kodak Company Rochester, NY 14650-2216 -- From: newhoir@duc.auburn.edu (Irene Newhouse) Subject: Re: CCL:pi-stacking in MOPAC? To: mirko@SARA.NL Message-Id: <9507131308.AA26317@mallard.duc.auburn.edu> X-Envelope-To: a412knsm@horus.sara.nl Content-Transfer-Encoding: 7BIT No, MOPAC doesn't handle pi-stacking at all. The problem is, that in these interactions, the long-r *tails* of the wave functions are involved, which is rather different from the situation for which the parameters are derived! I've done some work in this area, & have a folder of references somewhere- I'd have to dig them out. S. Fraga, in Canada, wrote a program that uses a force- field to compute the intermolecular interactions between molecules of fixed geometry that I've found works rather well. This work has been recently ex- tended by R. E. Cachau. A. Warshell claims that his program, available for purchase or as a QCPE version, will do intermolecular interactions, but I've found it to be rather flaky, to be honest. If you don't have access to an on-line catalog that can help you find these refs., let me know & I'll try to find my file... Irene Newhouse From: bowlus@palres.dnet.sandoz.com (Steve Bowlus) Subject: pi-stacking in MOPAC To: "mirko@SARA.NL"@SNDZEH.dnet.sandoz.com Cc: BOWLUS@sandoz.com Message-Id: <199507131453.AA10374@tigger.jvnc.net> X-Envelope-To: a412knsm@horus.sara.nl Content-Transfer-Encoding: 7BIT I would be interested in any useful replies. I'm also interested in formation of aromatic-aromatic complexes, in which pi-stacking is one possible form of interaction. I've not had experimental results to compare, but the structures I've dealt with seemed reasonable, given the normal amount of jiggling about to relieve steric interactions ( these are not the simplest of aromatics!) sb =========================================================================== Stephen B. Bowlus, Ph.D. Computer-Aided Molecular Design Research Division e-mail: bowlus@sandoz.com Sandoz Agro, Inc. Phone: + 1 415 354 3904 975 California Ave. Fax: + 1 415 857 1125 Palo Alto, CA 94304 =========================================================================== From: dan@sage.syntex.com (Dr. Daniel L. Severance) Subject: Re: CCL:pi-stacking in MOPAC? In-Reply-To: mirko@SARA.NL (Mirko Kranenburg) "CCL:pi-stacking in MOPAC?" (Jul 13, 12:25pm) To: mirko@SARA.NL (Mirko Kranenburg) Message-Id: <9507130845.ZM3630@sage.syntex.com> X-Envelope-To: a412knsm@horus.sara.nl Mime-Version: 1.0 X-Mailer: Z-Mail (3.2.0 26oct94 MediaMail) Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7BIT References: Hi, You need to take into account the other interactions going on in the crystal structure. It's likely that pi-stacking isn't a minimum, but rather, the residual required to allow the other interactions to take place... Dan -- Dr. Daniel L. Severance dan@sage.syntex.com Staff Researcher Work phone: (415) 354-7509 Syntex Discovery Research Home phone: (415) 969-5818 R6W-002 Fax (Work): (415) 354-7363 3401 Hillview Ave Palo Alto, CA 94303 From: CHMORA@LSUVAX.SNCC.LSU.EDU Subject: Re: CCL:pi-stacking in MOPAC? To: mirko@SARA.NL Message-Id: <01HSTKAHYE1EA2HA17@LSUVAX.SNCC.LSU.EDU> X-Envelope-To: a412knsm@horus.sara.nl X-Vms-To: IN%"mirko@SARA.NL" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Dear Mirko: Interesting question. Could you please send me a summary and or the answers you get? Thanks a lot in advance. Sincerely, Guillermo A. Morales internet: chmora@sn01.sncc.lsu.edu From: Ernest Chamot Subject: Re: CCL:pi-stacking in MOPAC? To: Mirko Kranenburg Message-Id: X-Envelope-To: a412knsm@horus.sara.nl X-Mailer: VersaTerm Link v1.1.6 Content-Transfer-Encoding: 7BIT Dear Mirko, I am not surprised that: >Currently I am performing geometry optimizations on compounds in which >pi-stacking has an important contribution to the overall geometry. >The results from MOPAC 6 calculations (AM1, PM3 and MNDO) show no pi- >stacking whatsoever, even when a X-ray structure was used as input. > At one point I was trying to model charge transfer complexes, which should have a pi stacking interaction. I assume this would be similar to the system you are trying to study. I had a GREAT deal of difficulty getting Mopac to find any stable complex, even when I (as you) started with two molecules in their relative positions as determined by X-ray diffraction. I did finally find one ever so slightly stable local minimum. At least in my experience (and evidently in yours) the packing forces may be too weak relative to other contributions to energy for Mopac to optimize effectively. It may just be that they are in fact weak, but I would have thought them to be significant for a charge transfer complex. I suspect that the wavefunctions in AM1 and PM3 formalisms do not have enough diffuse character to model the longer range interactions correctly. This is only my own observation, so if you get replies confirming this as a weakness, or identifying some other cause of the problem, I would like to hear them. Thanks. EC --- Ernest Chamot Consultant in Computational Chemistry Applications Chamot Labs, Inc. 530 E. Hillside Rd. Naperville, Illinois 60540 echamot@xnet.com Phone/Fax: (708) 637-1559 From: MARTINN@UNCWIL.EDU Subject: Re: CCL:pi-stacking in MOPAC? To: mirko@SARA.NL Message-Id: <01HSTNBZQ3WI8WVZNH@UNCWIL.EDU> X-Envelope-To: a412knsm@horus.sara.nl X-Vms-To: IN%"mirko@SARA.NL" Mime-Version: 1.0 Content-Type: TEXT/PLAIN; CHARSET=US-ASCII Content-Transfer-Encoding: 7BIT Mirko, We,too, have experienced what you state: no evidence of pi stacking in any of the semi-empirical methods. We have found an attraction using either HF or MP2 calculations, however (not yet published). Ned Martin ********************************************************************** Ned H. Martin, Chair Department of Chemistry University of North Carolina at Wilmington Voice: 910-395-3453 601 S. College Rd Fax: 910-395-3013 Wilmington, NC 28403-3297 martinn@vxc.uncwil.edu ********************************************************************** From: uk10000@cus.cam.ac.uk (Dr U. Koch) Subject: Re: CCL:pi-stacking in MOPAC? In-Reply-To: from "Mirko Kranenburg" at Jul 13, 95 12:25:58 pm To: mirko@SARA.NL (Mirko Kranenburg) Message-Id: X-Envelope-To: a412knsm@horus.sara.nl X-Mailer: ELM [version 2.4 PL21] Content-Type: text Content-Transfer-Encoding: 7BIT Content-Length: 833 Dear Mirko, which system are you looking at? Generally, 2 contributions are normally important for the geometry of pi-pi interactions: Dispersion, which favours maximum overlap between the pi systems and electrostatics. Since the dispersion energy is included in the quantumchemical correlation energy, you can't expect to reproduce it by calculations on the SCF level. The electrostatic part is often determined by a signifcant anisotropy of the atomic charge distribution (pi-densities). To reproduce these you need polarizable basis functions. I think for pi-pi interactions force field methods with a more sophisticated description of the electrostatics (see: S.L. Price , A.J. Stone , JCP 86 (1987), 2859 or C.A. Hunter and J.K.M. Sanders, JACS 112 (1990), 5525) are often a good alternative. Best wishes Uwe Koch From: " (Thomas Bally)" Subject: Re: CCL:pi-stacking in MOPAC? To: " (Mirko Kranenburg)" Message-Id: X-Envelope-To: a412knsm@horus.sara.nl Content-Identifier: Re: CCL:pi-st... Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7BIT X400-Content-Type: P2-1984 (2) X400-Mts-Identifier: [/PRMD=switch/ADMD=arcom/C=CH/;Currently I am performing geometry optimizations on compounds in which >pi-stacking has an important contribution to the overall geometry. >The results from MOPAC 6 calculations (AM1, PM3 and MNDO) show no pi- >stacking whatsoever, even when a X-ray structure was used as input. > >My question: does MOPAC take pi-stacking into account, and are there any >relevant references concerning this problem. I have encountered the same problem in a different context. The reason for the absence of pi-stacking is, that in all of Dewar's semiempirical methods (MINDO / MNDO / AM1 / PM3) interaction between pi-systems is vastly underestimated at all distances and is nearly zero at distances such as they prevail typically in pi-stacked systems. Somehow this appears to be a well kept secret as I know of no reference where this is treated in any systematic way (we intend to publish a paper on this on occasion). I brought up the point at a Faraday Discussion in response to a paper presented there by Tim Clark (see: J. Chem. Soc. Farad. Trans. 90, p. 1808 (1994), in particular Fig 6 and the corresponding text on this page). In a later contribution on the same topic, Josef Michl mentioned a paper by a group from Du Pont (J.Am. Chem. Soc. 98, 8401 (1976)) who have apparently attempted to remedy the situation, but with no success. I don't know where the problem comes from because the Slater AO's which are used to calculate overlaps in all-valence semiempirical methods ought to be well suited for that kind of problem. So it is the parametri- zation which is to be blamed and this is difficult to fix. I am in correspondence with Mike Zerner, the author of ZINDO, who thinks that he knows how one would need to go about doing this properly, but I have not heard from him for a while. I think if you want to calculate pi-stacked systems you need to go to some ab-initio method. DFT is probably also not so good as it often has problems with van der Waals type interactions. I am sorry that I cannot help you more at this point! Best wishes thomas *-------------------------------------------------------------------------* | Prof. Thomas Bally | E-mail: Thomas.Bally@unifr.ch | | Institute for Physical Chemistry | WWW page: | | University of Fribourg | http://sgich1.unifr.ch/pc.html | | Perolles | | | CH-1700 FRIBOURG | Tel: 011-41-37 29 8705 | | Switzerland | FAX: 011-41-37 29 9737 | *-------------------------------------------------------------------------* From: kcousins@wiley.csusb.edu (Kimberley Cousins) Subject: pi stacking To: mirko@SARA.NL Message-Id: <199507180538.AA07605@wiley.csusb.edu> X-Envelope-To: a412knsm@horus.sara.nl X-Mailer: ELM [version 2.4 PL22] Content-Type: text Content-Transfer-Encoding: 7BIT Content-Length: 764 I'm hoping this will help. I do not have the original reference here at home, and am about to leave the country, so my memory will have to serve. There is a paper on the 3-D structure of hexaphenylbenzene published about ten years ago in a less common (Indian?) journal. Its the only paper that I know of with calculations of hexaphenylbenzene, and as I recall has some discussion of pi stacking (or lack there of) in the calculated vs. crystal structure of this compound. If you would like more information, and can wait til after August 1, please email me back and let me know. Kimberley Cousins Department of Chemistry California State University, San Bernardino 5500 University Parkway San Bernardino, CA 92407 (909)880-5391 kcousins@wiley.csusb.edu Mirko Kranenburg Universiteit van Amsterdam J.H. van 't Hoff Research Institute Dept. of Inorganic Chemistry and Homogeneous Catalysis Nieuwe Achtergracht 166 1018 WV Amsterdam The Netherlands tel. +31-20-5256417 fax. +31-20-5256456 e-mail: mirko@sara.nl From owner-chemistry@ccl.net Fri Jul 21 08:41:27 1995 Received: from slater.cem.msu.edu for young@slater.cem.msu.edu by www.ccl.net (8.6.10/930601.1506) id IAA23560; Fri, 21 Jul 1995 08:39:45 -0400 Received: by slater.cem.msu.edu (950215.SGI.8.6.10/940406.SGI.AUTO) for chemistry@ccl.net id IAA13714; Fri, 21 Jul 1995 08:41:51 -0400 Date: Fri, 21 Jul 1995 08:41:51 -0400 From: young@slater.cem.msu.edu (Dave Young) Message-Id: <199507211241.IAA13714@slater.cem.msu.edu> To: chemistry@ccl.net Subject: Biograf address Hello, The Biograf software package does not run under IRIX 5.3 Molecular Simulations, which did market biograf is no longer supporting biograf. I have heard rumors that the rights to biograf have been sold to another company. Does anyone have the address of the company currently holding the rights to sell biograf? Thank you in advance. Dave Young young@slater.cem.msu.edu -------------------------------------------------------------------------- Q: How do you avoid making the errors of the HF central field approximation in the first place in stead of spending all of your time trying to correct for them. A: Quantum Monte Carlo -------------------------------------------------------------------------- From owner-chemistry@ccl.net Fri Jul 21 09:26:30 1995 Received: from violet.csv.warwick.ac.uk for mailer-daemon@csv.warwick.ac.uk by www.ccl.net (8.6.10/930601.1506) id JAA24804; Fri, 21 Jul 1995 09:23:39 -0400 Message-Id: <25305.199507211323@violet.csv.warwick.ac.uk> Received: from chem.warwick.ac.uk by violet.csv.warwick.ac.uk with SMTP id OAA25305; Fri, 21 Jul 1995 14:23:08 +0100 Received: by chemhp.chem.warwick.ac.uk (1.37.109.4/16.2) id AA11413; Fri, 21 Jul 95 14:38:42 +0100 From: Craig Wilson Subject: Re: CCL:Saddle point optimization To: eloranta@voimax.voima.jkl.fi Date: Fri, 21 Jul 95 14:38:40 BST Cc: CHEMISTRY@ccl.net Reply-To: msrge@csv.warwick.ac.uk In-Reply-To: <199507210859.LAA03806@voimax.voima.jkl.fi>; from "Jussi Eloranta" at Jul 21, 95 11:59 am Reply-To: msrge@csv.warwick.ac.uk Organization: University of Warwick, COVENTRY, CV4 7AL, England, UK. Telephone: 0203-522187 (International +44 203-522187) Mailer: Elm [revision: 70.85] > Hi, > > > Is there a way to optimize to the geometry which is a minimum with respect > to all variables except two for which it should be a maximum > (ie. saddle point of order of 2). These two variables are some known Z-matrix > variables. > > I've looked at saddle, ts, and ef params for opt but they don't seem to > help with this. > > BTW would this be as simple as choosing "-Grad" direction for all variables > but these two for which it would be "+Grad" ? > > Regards, > > Jussi Eloranta > > > PS. I'm using Gaussian 92/DFT You should be able to optimize to a second order saddle point by specifying Opt(Saddle=2) in the route section Hope this helps, Craig -- ******************************************************************************* * Craig Wilson e-mail: msrge@csv.warwick.ac.uk * * Dept. of Chemistry * * University of Warwick * * COVENTRY Phone: 01203-523523 ext. 2541 (UK) * * CV4 7AL * * England, UK. "This isn't a soap opera - it's real life!" * ******************************************************************************* From owner-chemistry@ccl.net Fri Jul 21 11:26:34 1995 Received: from Arizona.EDU for jaimeco@pecos.rc.arizona.edu by www.ccl.net (8.6.10/930601.1506) id LAA27723; Fri, 21 Jul 1995 11:17:27 -0400 Received: from pecos.rc.arizona.edu by Arizona.EDU (PMDF V5.0-3 #2381) id <01HT4IPIYXLCBETFR7@Arizona.EDU> for chemistry@ccl.net; Fri, 21 Jul 1995 08:17:25 -0700 (MST) Received: by pecos.rc.arizona.edu (5.x/SMI-SVR4) id AA16620; Fri, 21 Jul 1995 08:16:35 -0700 Date: Fri, 21 Jul 1995 08:16:35 -0700 From: jaimeco@pecos.rc.arizona.edu (Jaime E. Combariza) Subject: Problem with Gaussian job To: chemistry@ccl.net Cc: jaimeco@pecos.rc.arizona.edu Message-id: <9507211516.AA16620@pecos.rc.arizona.edu> MIME-version: 1.0 Content-type: TEXT/PLAIN; CHARSET=US-ASCII Content-transfer-encoding: 7BIT Hi Netters: I have the following problem and will appreciate your input! I am performing a rather large CI calculation with the option density=curr.I notice that the CI run appears to be converged, with a .rwf file about 1Gbyte in size. Then it looks like the 2-electron density matrix is being calculated, the .rwf file soars to 2Gbytes in size and the job crashes due to lack of disk space. Why is the 2-electron density matrix being calculated and is there anyway to avoid the creation of this large file? The end of the output file is appended: Thanks. ---------- ITERATION NR. 16 ********************** DD1Dir will call FoFDir 8 times, MxPair= 80 NAB= 300 NAA= 0 NBB= 0. Singles A-vector converged to 1.77D-08 The Euclidean norm of the A-vectors is .6768732D-06 RLE energy= -.7277190981 DE(CI)= -.72771910D+00 E(CI)= -.19304241802D+04 NORM(A)= .11005301D+01 SIZE-CONSISTENCY CORRECTION: S.C.C.= -.18211846D+00 E(CI,SIZE)= -.19306062986D+04 *************************************************************** Selected functions: Compute CISD 2PDM. IO1PDM= 0 IOW= 0 IO2PDM= 0. NGot= 2000000. Frozen-core window: NFC= 11 NFV= 0. File extend in NtrExt1 failed. File extend in NtrExt1 failed: Invalid argument bsh: 15869 Abort: A memory image file is created as "core". From owner-chemistry@ccl.net Fri Jul 21 12:11:43 1995 Received: from gatekeeper.tripos.com for matt@metis.tripos.com by www.ccl.net (8.6.10/930601.1506) id MAA29210; Fri, 21 Jul 1995 12:05:40 -0400 Received: (from news@localhost) by gatekeeper.tripos.com (8.6.12/8.6.12) id LAA09931; Fri, 21 Jul 1995 11:04:36 -0500 Received: from tripos.tripos.com(192.160.145.1) by gatekeeper.tripos.com via smap (V1.3) id sma009929; Fri Jul 21 11:04:23 1995 Received: from metis.UUCP (metis.tripos.com [192.160.145.41]) by tripos.tripos.com (8.6.12/8.6.12) with SMTP id LAA17363; Fri, 21 Jul 1995 11:02:38 -0500 Received: by metis.UUCP (4.0/4.7) id AA10901; Fri, 21 Jul 95 11:02:32 CDT From: matt@metis.tripos.com (Matt Clark) Message-Id: <9507211602.AA10901@metis.UUCP> To: Charles Letner Cc: Computational Chemistry List , matt@tripos.tripos.com Subject: Re: CCL:Benchmarks In-Reply-To: Your message of Thu, 20 Jul 95 18:31:47 -0400. Date: Fri, 21 Jul 95 11:02:32 EDT Regarding what measurement to use in a benchmark, I feel you should measure what you are really interested in. Namely "how long does it take to run", or wall clock time. If you use CPU time you would add up the time for each individual CPU, and that isn't a very useful measure of the performance increase of a parallel computer. Wall clock time takes into account any paging/swapping/io that the program causes, which are important practical issues. From owner-chemistry@ccl.net Fri Jul 21 12:56:33 1995 Received: from MIZZOU1.missouri.edu for PHYSPLMP@MIZZOU1.missouri.edu by www.ccl.net (8.6.10/930601.1506) id MAA00542; Fri, 21 Jul 1995 12:47:48 -0400 From: Message-Id: <199507211647.MAA00542@www.ccl.net> Received: from MIZZOU1 by MIZZOU1.missouri.edu (IBM VM SMTP V2R3) with BSMTP id 3026; Fri, 21 Jul 95 11:30:58 CDT Received: by MIZZOU1 (Mailer R2.10 ptf000) id 7161; Fri, 21 Jul 95 11:30:58 CDT Date: Fri, 21 Jul 95 11:24:26 CDT To: chemistry@ccl.net Subject: Problems with Mopac7 I have downloaded Mopac7 from osc. After some problems with the Makefile, etc I have compiled it and have an executable for an SGI Indigo Irix 5.3. Many of the testdata appear to work just fine but attemps to use the keywords THERMO (with correct ROT=n)and/or CHARGE=1(or -1) kill the run and return Bus Error ( coredump). I've run other versions of MOPAC successfully on other platforms and am stumped. Any and all suggestions will be appreciated. I'll summarize to the list if there is interest. Many Thanks, Pat Plummer, Univ. of Missouri, physplmp@mizzou1.missouri.edu From owner-chemistry@ccl.net Fri Jul 21 13:11:33 1995 Received: from remcure.bmb.wright.edu for cletner@remcure.bmb.wright.edu by www.ccl.net (8.6.10/930601.1506) id MAA00510; Fri, 21 Jul 1995 12:46:23 -0400 Received: by remcure.bmb.wright.edu (931110.SGI/921111.SGI.AUTO) for CHEMISTRY@ccl.net id AA11466; Fri, 21 Jul 95 15:40:13 -0400 Date: Fri, 21 Jul 1995 15:33:20 -0400 (EDT) From: Charles Letner Subject: RE:Benchmarks inital summary To: Computational Chemistry List Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Hello, Yesterday I posted a question about benchmarking software. I have gotten a number of inital responses. Most indicate that "wall clock" time is probably one of the better benchmarks. The most common reason: it takes into account issues such as multipule processors, memory swapping, etc.. To eliminate factors such as other jobs and other users one indiviual mentioned that they make the machine unavailable to other users while the benchmark job is running. I will post a full summary next week that includes the response I have already recieved as well as any I recieve over the weekend. Thanks to everyone who responded. Chuck Charles Letner Wright State University Department of Biochemistry Dayton, OH 45435 e-mail: cletner@remcure.bmb.wright.edu From owner-chemistry@ccl.net Fri Jul 21 13:56:33 1995 Received: from msi.com for Don_Gregory@msi.com by www.ccl.net (8.6.10/930601.1506) id NAA02123; Fri, 21 Jul 1995 13:56:23 -0400 Received: from qmail-gw.msi.com by msi.com (4.1/SMI-4.1) id AA25387; Fri, 21 Jul 95 13:52:37 EDT Message-Id: Date: 21 Jul 1995 13:59:10 -0400 From: "Don Gregory" Subject: BioGraf and MSI To: "Comp. Chem. Listserver" X-Mailer: Mail*Link SMTP-QM 3.0.2 Subject: Time:12:37 PM OFFICE MEMO BioGraf and MSI Date:7/21/95 Hello! Dave Young asked about the status of BioGraf. [Suggest deleting this next sentence and replacing it with what follows in brackets.] Molecular Simulations, Inc., retains the rights to BioGraf (PolyGraf, NMR-Graf, etc.), however we no longer actively market or sell these products. Rather, virtually all of the scientific functionality (and more!) of BioGraf is now available through MSI either via Quanta (for macromolecular science), or Cerius2 (for polymers, combinatorial/drug-design, and materials science). [Suggest deleting the next two sentences and replacing them with what follows in brackets.] At the time of this functionality-transition, existing users of BioGraf were allowed an across-the-board trade-out of the BioGraf functionality in these new products, and nearly eveyone opted to do so. But, and this is really important, MSI continues to answer scientific-support questions for those users still using these products. The CCL really isn't the appropriate forum for sales/marketing blurbs, so f [IF] anyone is interested in more information about these products and/or MSI in general, check-out our web page (www.msi.com). There are also all kinds of phone numbers, e-mail addresses etc. if anyone would like, .e.g. 1-617-229-9800 Don Gregory Protein Engineering Product Manager MSI