From owner-chemistry@ccl.net Tue Sep 6 07:31:26 2005 From: "CCL" To: CCL Subject: CCL: GB+QM with mulliken charges ! Message-Id: <-29071-050906071417-11987-2NbaQmdzjXHCm+8GjNpz2w_+_server.ccl.net> X-Original-From: Marcel Swart Content-Type: multipart/alternative; boundary=Apple-Mail-3--482907637 Date: Tue, 6 Sep 2005 11:57:44 +0200 Mime-Version: 1.0 (Apple Message framework v622) Sent to CCL by: Marcel Swart [m.swart_+_few.vu.nl] --Apple-Mail-3--482907637 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=WINDOWS-1252; format=flowed Dear John, Gaussian-Type Orbitals (GTO's) have shorter tails than Slater-Type=20 Orbitals (STO's), and GTO's don't show the cusp at the origin, which are two strong=20 arguments for using STO's. Regarding the atomic charges, it might be better to use more recent and=20= accurate methods, like the Voronoi Deformation Density or Multipole Derived=20 Charge analysis. Both are available within the ADF program. For further information: C. Fonseca Guerra, J.-W. Handgraaf, E.J. Baerends, F.M. Bickelhaupt "Voronoi Deformation Density (VDD) charges. Assessment of the Mulliken, Bader, Hirshfeld, Weinhold and VDD methods=20= for Charge Analysis" J. Comput. Chem. 2004, 25, p. 189-210 M. Swart, P.Th. van Duijnen, J.G. Snijders "A charge analysis derived from an atomic multipole expansion" J. Comput. Chem. 2001, 22, p. 79-88 On Sep 2, 2005, at 6:13 PM, CCL wrote: > > Sent to CCL by: John McKelvey [jmmckel]*[attglobal.net] > Chris, > > Just curious... PLease remind me.. do gaussian basis sets have=20 > longer tails than Slaters? If so, would using Slaters, as in ADF,=20 > make a significant difference in Mulliken and Loewdin populations? > > Best regards, > John McKelvey > > CCL wrote: > >> Pradipta, >> >> Mulliken charges are extraordinarily basis set dependent and tend to=20= >> become highly nonphysical as one moves beyond a minimal basis=20 >> representation. Since the solvation free energy is non-linear in the=20= >> GB charge representation, this can lead to severe problems. That is=20= >> why the SMx GB models rely on Loewdin charges (these, too, show some=20= >> basis set dependence, which is one reason the CMx charge models are=20= >> used to "preprocess" the Loewdin charges when we compute SMx=20 >> solvation free energies). >> >> You might find the following reference (and references therein)=20 >> germane: >> >> Li, J.; Hawkins, G. D.; Cramer, C. J.; Truhlar, D. G. "Universal=20 >> Reaction Field Model Based on ab initio Hartree-Fock Theory" / Chem.=20= >> Phys. Lett./ *1998*, /288/, 293. >> >> Best regards, >> >> Christopher J. Cramer >> University of Minnesota =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96= =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96 dr. Marcel Swart Theoretische Chemie Vrije Universiteit Amsterdam Faculteit der Exacte Wetenschappen De Boelelaan 1083 1081 HV Amsterdam The Netherlands Tel +31-(0)20-5987619 Fax +31-(0)20-5987629 E-mail m.swart_+_few.vu.nl Web http://www.few.vu.nl/~swart =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96= =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96 --Apple-Mail-3--482907637 Content-Transfer-Encoding: quoted-printable Content-Type: text/enriched; charset=WINDOWS-1252 Dear John, Gaussian-Type Orbitals (GTO's) have shorter tails than Slater-Type Orbitals (STO's), and GTO's don't show the cusp at the origin, which are two strong arguments for using STO's. Regarding the atomic charges, it might be better to use more recent and accurate methods, like the Voronoi Deformation Density or Multipole Derived Charge analysis. Both are available within the ADF program. For further information: C. Fonseca Guerra, J.-W. Handgraaf, E.J. Baerends, F.M. Bickelhaupt "Voronoi Deformation Density (VDD) charges.=20 Assessment of the Mulliken, Bader, Hirshfeld, Weinhold and VDD methods for Charge Analysis" J. Comput. Chem. 2004, 25, p. 189-210 M. Swart, P.Th. van Duijnen, J.G. Snijders "A charge analysis derived from an atomic multipole expansion" J. Comput. Chem. 2001, 22, p. 79-88 On Sep 2, 2005, at 6:13 PM, CCL wrote: Sent to CCL by: John McKelvey [jmmckel]*[attglobal.net] Chris, Just curious... PLease remind me.. do gaussian basis sets have longer tails than Slaters? If so, would using Slaters, as in ADF, make a significant difference in Mulliken and Loewdin populations? Best regards, John McKelvey CCL wrote: Pradipta, Mulliken charges are extraordinarily basis set dependent and tend to become highly nonphysical as one moves beyond a minimal basis representation. Since the solvation free energy is non-linear in the GB charge representation, this can lead to severe problems. That is why the SMx GB models rely on Loewdin charges (these, too, show some basis set dependence, which is one reason the CMx charge models are used to "preprocess" the Loewdin charges when we compute SMx solvation free energies). You might find the following reference (and references therein) germane: Li, J.; Hawkins, G. D.; Cramer, C. J.; Truhlar, D. G. "Universal Reaction Field Model Based on ab initio Hartree-Fock Theory" / Chem. Phys. Lett./ *1998*, /288/, 293. Best regards, Christopher J. Cramer University of Minnesota = Helvetica=96=96=96=96=96=96=96=96= =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96= =96=96=96=96=96=96=96=96=96=96=96 = Papyrusdr. Marcel Swart = Papyrus = OsakaTheoretische Chemie Vrije Universiteit Amsterdam Faculteit der Exacte Wetenschappen De Boelelaan 1083 1081 HV Amsterdam The Netherlands Tel +31-(0)20-5987619 Fax +31-(0)20-5987629 E-mail m.swart_+_few.vu.nl Web http://www.few.vu.nl/~swart = Helvetica=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96= =96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96=96 --Apple-Mail-3--482907637-- From owner-chemistry@ccl.net Tue Sep 6 09:32:46 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29072-050906093207-5394-2NbaQmdzjXHCm+8GjNpz2w(!)server.ccl.net> X-Original-From: "Torsten Bruhn" Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de] Hello, can anyone tell me, why calculations with Gaussiantype-Orbitals are faster than calculations with STO? Is there literature where this is described in detail? From owner-chemistry@ccl.net Tue Sep 6 10:27:10 2005 From: "CCL" To: CCL Subject: CCL: W:Keyword External in Gaussian 03 Message-Id: <-29073-050906053408-6947-2NbaQmdzjXHCm+8GjNpz2w]~[server.ccl.net> X-Original-From: "JuanManuel OrtizSnchez" Sent to CCL by: "JuanManuel OrtizSnchez" [juanma]~[klingon.uab.es] Greetings, I am trying to run a Gaussian calculation to locate a Transition State with the Time-Dependent DFT method. I want to connect the Gaussian with the Turbomole program (with a proper script) to compute the gradients. As I know, I have to type the keyword External in the Gaussian input file, but all that I get is the error message Unrecognized post-SCF IPrc10 in PutPrc. Is anybody trying this kind of calculations with more succed, or has anybody found the same problem? Many thanks in advance ---------------------------------- Juan Manuel Ortiz Snchez Unitat de Quimica Fisica Departament de Quimica Edifici C Universitat Autonoma de Barcelona 08193 Bellaterra, Spain Phone: +34 93 5812814 Fax: +34 93 5812920 E-mail: juanma]~[klingon.uab.es ---------------------------------- From owner-chemistry@ccl.net Tue Sep 6 11:08:49 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29074-050906110632-1720-2NbaQmdzjXHCm+8GjNpz2w]*[server.ccl.net> X-Original-From: Laurence Cuffe Content-disposition: inline Content-language: en Content-transfer-encoding: 7BIT Content-type: text/plain; charset=us-ascii Date: Tue, 06 Sep 2005 16:06:24 +0100 MIME-version: 1.0 Sent to CCL by: Laurence Cuffe [Laurence.Cuffe]*[ucd.ie] ----- Original Message ----- > From: CCL Date: Tuesday, September 6, 2005 2:33 pm Subject: CCL: W:GTO and STO > > Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni- > oldenburg.de]Hello, > > > can anyone tell me, why calculations with Gaussiantype-Orbitals are > faster than calculations with STO? Is there literature where this > is described in detail? The short answer is that calculating the overlap between two Gaussiantype -Orbitals can be done in closed form. That is given two GTO's A and B you can write an relatively simple algebraic expression for the size of the overlap between them. This is not possible with Slater type orbitals. While this is generally covered in most textbooks on computational chemistry, a google search on "GTO STO calculating overlap integrals" should get you some useful information. e.g http://www.mse.kth.se/~delin/sams_files/chapter7+8.pdf Jan Labanowski also has a good description of basis sets and how they're arrived at up on the web. All the best Dr Laurence Cuffe > > > > -= 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> > 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> -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- > +-+-+ > > > > From owner-chemistry@ccl.net Tue Sep 6 11:52:03 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29075-050906113241-18992-2NbaQmdzjXHCm+8GjNpz2w]~[server.ccl.net> X-Original-From: Joop van Lenthe Content-Type: multipart/alternative; boundary=Apple-Mail-2--466290237 Date: Tue, 6 Sep 2005 16:34:41 +0200 Mime-Version: 1.0 (Apple Message framework v622) Sent to CCL by: Joop van Lenthe [joop]~[chem.uu.nl] --Apple-Mail-2--466290237 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed A product of two GAUSSIANS is another GAUSSIAN somewhere in between, resulting in at most 2 center integrals. This is not the case for Slaters. Also I would imgine, that the fact that gaussians fall off more quickly makes integration easier. Joop There must be papers by e.g. V.R.Saunders in proceedinmgs of summerschools. On Sep 6, 2005, at 15:37, CCL wrote: > > Sent to CCL by: "Torsten Bruhn" > [torsten.bruhn(!)mail.uni-oldenburg.de] > Hello, > > > can anyone tell me, why calculations with Gaussiantype-Orbitals are > faster than calculations with STO? Is there literature where this is > described in detail?> To send e-mail to subscribers of CCL put the string CCL: on your > Subject: line> > Send your subscription/unsubscription requests to: > CHEMISTRY-REQUEST]~[ccl.net> > If your is mail bouncing from ccl.net domain due to spam filters, > please> -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- > +-+ > > > ========================================= Joop van Lenthe Theoretical Chemistry Group Debye Institute, Utrecht University Padualaan 8 3584 CH Utrecht -31-30-2532733 joop]~[chem.uu.nl ========================================= --Apple-Mail-2--466290237 Content-Transfer-Encoding: 7bit Content-Type: text/enriched; charset=US-ASCII A product of two GAUSSIANS is another GAUSSIAN somewhere in between, resulting in at most 2 center integrals. This is not the case for Slaters. Also I would imgine, that the fact that gaussians fall off more quickly makes integration easier. Joop There must be papers by e.g. V.R.Saunders in proceedinmgs of summerschools. On Sep 6, 2005, at 15:37, CCL wrote: Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de] Hello, can anyone tell me, why calculations with Gaussiantype-Orbitals are faster than calculations with STO? Is there literature where this is described in detail?========================================= Joop van Lenthe Theoretical Chemistry Group Debye Institute, Utrecht University Padualaan 8 3584 CH Utrecht -31-30-2532733 joop]~[chem.uu.nl ========================================= --Apple-Mail-2--466290237-- From owner-chemistry@ccl.net Tue Sep 6 12:21:49 2005 From: "CCL" To: CCL Subject: CCL: W:New release of MayaChemTools package... Message-Id: <-29076-050906122009-740-2NbaQmdzjXHCm+8GjNpz2w[a]server.ccl.net> X-Original-From: "Manish Sud" Sent to CCL by: "Manish Sud" [msud[a]san.rr.com] A new release of MayaChemTools, a growing collection of Perl scripts to support computational discovery needs, with initial focus on Cheminformatics, is available as free software; you can redistribute it and/or modify it under the terms of the GNU LGPL. New scripts: o AnalyzeSDFilesData.pl o AnalyzeTextFilesData.pl o InfoTextFiles.pl o ModifySDFilesDataFields.pl o SortSDFiles.pl And a variety of enhancement to existing scripts. Those of you interested in viewing SD files using JMol or other supported structure viewers, check out the new features in SDFilesToHTML.pl. To read more about the scripts and download the package, please visit www.mayachemtools.org. Your feedback is welcome. Happy scripting, Manish Sud msud[a]san.rr.com From owner-chemistry@ccl.net Tue Sep 6 13:00:40 2005 From: "CCL" To: CCL Subject: CCL: W:Spin Contaminationated DFT vs ROHF for force field parameters Message-Id: <-29077-050906125951-21714-2NbaQmdzjXHCm+8GjNpz2w[A]server.ccl.net> X-Original-From: "Gustavo Alberto Mercier" Sent to CCL by: "Gustavo Alberto Mercier" [gamercier[A]yahoo.com] Hi! I need some guidance on how to proceed on the following problem: I am developing parameters for MD/MM using the Amber force field. The system of interest is a Manganese porphyrin. The complex is hi spin Mn(III). In the past I've done unrestricted DFT computations using ADF and gradient corrected XC functionals with success. However, I am now running into problems with spin contamination. To generate the parameters I've constructed two models -- an effort to use the actual complex and do a global fit to the missing parameters was unwieldy and led to non-physical results. The models are the following: Model 1: pyrrolate + Mn+3 charge: 2; spin: 4 excess alpha electrons Geometry: pyrrolate optimized; Mn+3 placed at about 2A from the N H H \ / C ------C / \ HC (-) CH \ / \ / N | Mn(+3) Model 2: 2,5 diethenylpyrrolate + Mn+3 charge: 2; spin: 4 excess alpha electrons Geometry: Adapted from a Mn(III) Porphyrin structure; Mn-N about 2A H H \ / H C ------C H \ / \ / C-C (-) C-C // \ / \\ H--C \ / C--H \ N / H | H Mn(+3) For both models: Basis set: TZ2P XC: OPBE (but tried Becke Perdew or just LDA with little difference) Symmetry: C2v Single Point calculations, no optimizations, yet. ADF generates an estimate of S**2 (not the true S**2 due to complexities in computing S**2). The estimate is 7 while the expected value is 6 ( (4/2)*((4/2)+1) ). Odd enough if I move to HF methods (ie. ROHF -> UHF-> UHF/MP2; Mn basis SBKJC ECP, others 6-31g++(d,p) as implemented in GAMESS-US) the spin contamination for Model 1 is trivial (6.07 vs 6) but more for Model 2, 6.5-6.7 vs 6. My questions are: 1) Is there a way to minimize spin contamination in the ADF runs? I rather do DFT computations with ADF (faster and uses STO) 2) What is the experience with ROHF for the purposes of geometry optimization, and generation of force field parameters, if any. A quick look for reviews found nothing useful on this topic. It seems that people have done UHF optimization and then single point ROHF, but this seems odd. Because I am not pleased with the spin contamination of Model 2 (for organic compounds 10% contamination is significant), I may be forced to do ROHF/MP2 computations to generate the parameters -- bond, angle, and dihedrals in the Amber force field for Mn+3. ROHF/MP2 geometry optimizations are not possible using GAMESS-US, so I may have to use ROHF optimized structures. These may be inadequate for such complexes. Or am I wrong? I don't trust the UHF or UHF/MP2 runs due to the spin contamination. Thanks! GMercier gamercier[A]yahoo.com gustavom[A]baylorhealth.edu From owner-chemistry@ccl.net Tue Sep 6 14:21:13 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29078-050906140903-14953-2NbaQmdzjXHCm+8GjNpz2w:+:server.ccl.net> X-Original-From: Serguei Patchkovskii Content-Type: TEXT/PLAIN; charset=US-ASCII Date: Tue, 6 Sep 2005 13:31:29 -0400 (EDT) MIME-Version: 1.0 Sent to CCL by: Serguei Patchkovskii [ps:+:ned.sims.nrc.ca] > Sent to CCL by: Laurence Cuffe [Laurence.Cuffe]*[ucd.ie] > The short answer is that calculating the overlap between two > Gaussiantype -Orbitals can be done in closed form. That is given two > GTO's A and B you can write an relatively simple algebraic expression > for the size of the overlap between them. This is not possible with > Slater type orbitals. This statement, of course, is false. Closed-form expressions for overlap integrals in terms of exponential integral-type functions are very well known. For example: C.C.J. Roothaan, J. Chem. Phys. 19, 1445 (1951) J.D. Talman, Phys. Rev. A 48, 243 (1993) Closed-form expressions in terms of "high-school-type" algebraic functions are also readily available, but not as numerically stable. For example: M. McCourt and J.W. McIver, J. Mol. Struct. (THEOCHEM) 163, 343 (1988). On the other hand, calculation of three- and particularly four-centre integrals involving Slater-type functions is much harder. As a result, the usual strategy of evaluating the four-centre two-electron integrals in AO basis is not very computationally efficient for STO-based codes. Other strategies (such as using auxiliary basis sets and/or numerical integration on grid) work rather nicely for STOs, so that STO-based codes are not intrinsically inferiour in terms of computational efficiency. As an example, STO-based ADF (http://www.scm.com/) is very competitive with GTO-based codes - both in terms of science which can be done, and in terms of speed or size of the systems which can be treated. Serguei --- Dr. Serguei Patchkovskii Tel: +1-(613)-990-0945 Fax: +1-(613)-947-2838 E-mail: Serguei.Patchkovskii:+:nrc.ca Coordinator of Modelling Software Theory and Computation Group Steacie Institute for Molecular Sciences National Research Council Canada Room 2011, 100 Sussex Drive Ottawa, Ontario K1A 0R6 Canada From owner-chemistry@ccl.net Tue Sep 6 14:49:22 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29079-050906112621-17295-2NbaQmdzjXHCm+8GjNpz2w[#]server.ccl.net> X-Original-From: Vincent Xianlong Wang Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=iso-8859-1 Date: Tue, 6 Sep 2005 07:26:16 -0700 (PDT) MIME-Version: 1.0 Sent to CCL by: Vincent Xianlong Wang [xloongw[#]yahoo.com] Hi Torsten, The preference of GTOs is due to the relative ease of integral evaluation of GTOs. See Ab initio molecular orbital theory by Hehre, W. J. et al. (Wiley Interscience), page 51 and references therein. Vincent --- CCL wrote: > > Sent to CCL by: "Torsten Bruhn" > [torsten.bruhn(!)mail.uni-oldenburg.de] > Hello, > > > can anyone tell me, why calculations with > Gaussiantype-Orbitals are faster than calculations > with STO? Is there literature where this is > described in detail? > > > > -= 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> > 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> > > > ______________________________________________________ Click here to donate to the Hurricane Katrina relief effort. http://store.yahoo.com/redcross-donate3/ From owner-chemistry@ccl.net Tue Sep 6 15:31:25 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29080-050906120712-29574-2NbaQmdzjXHCm+8GjNpz2w!A!server.ccl.net> X-Original-From: redo Content-Type: TEXT/PLAIN; charset=US-ASCII Date: Tue, 6 Sep 2005 16:47:41 +0200 (CEST) MIME-Version: 1.0 Sent to CCL by: redo [redo!A!thch.unipg.it] On Tue, 6 Sep 2005, CCL wrote: > > Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de] > Hello, Hello Torsten, > can anyone tell me, why calculations with Gaussiantype-Orbitals are > faster than calculations with STO? Is there literature where this is > described in detail? essentially because the integral evalutaion is faster, is faster because of the Gaussian product Theorem (for example the product of two 1s Gaussian functions, each one on different centers, is again a 1s Gaussian function on a different center) any Quantum Chemistry book should speak about it. You can also try searching for the Gaussian product Theorem with some searching engine. best regards. loriano. From owner-chemistry@ccl.net Tue Sep 6 15:31:26 2005 From: "CCL" To: CCL Subject: CCL: W:GTO and STO Message-Id: <-29081-050906144725-24763-2NbaQmdzjXHCm+8GjNpz2w.:.server.ccl.net> X-Original-From: "Ahmed" Content-Type: multipart/alternative; boundary="----=_NextPart_000_0019_01C5B2D8.D9E01AB0" Date: Tue, 6 Sep 2005 11:48:04 -0600 MIME-Version: 1.0 Sent to CCL by: "Ahmed" [ahmed.bouferguene.:.ualberta.ca] This is a multi-part message in MIME format. ------=_NextPart_000_0019_01C5B2D8.D9E01AB0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Hi, Calculations with GTOs are faster since they possess a multiplication theorem (a product of two GTOs is also a GTO). Now, in theory any solution of the shrodinger equation must have: [1] a cusp at the origin (which GTOs do not have) [2] an exponential decrease at infinity (which GTO do not have) The "flip side of the coin" is, multi-center integrals (which is the heart of ab initio calculations) over STOs is much much more difficult than with GTOs. Some people (including myself -but just for the challenge-) are still trying to setup techniques to use STOs Cheers _____ > From: owner-chemistry.:.ccl.net [mailto:owner-chemistry.:.ccl.net] Sent: Tuesday, September 06, 2005 8:35 AM To: Bouferguene, Ahmed Subject: CCL: W:GTO and STO A product of two GAUSSIANS is another GAUSSIAN somewhere in between, resulting in at most 2 center integrals. This is not the case for Slaters. Also I would imgine, that the fact that gaussians fall off more quickly makes integration easier. Joop There must be papers by e.g. V.R.Saunders in proceedinmgs of summerschools. On Sep 6, 2005, at 15:37, CCL wrote: Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de] Hello, can anyone tell me, why calculations with Gaussiantype-Orbitals are faster than calculations with STO? Is there literature where this is described in detail? ========================================= Joop van Lenthe Theoretical Chemistry Group Debye Institute, Utrecht University Padualaan 8 3584 CH Utrecht -31-30-2532733 joop.:.chem.uu.nl ========================================= ------=_NextPart_000_0019_01C5B2D8.D9E01AB0 Content-Type: text/html; charset="us-ascii" Content-Transfer-Encoding: quoted-printable

Hi,

 

Calculations with GTOs are faster = since they possess a multiplication theorem (a product of two GTOs is also a = GTO). Now, in theory any solution of the shrodinger equation must = have:

 

[1] a cusp at the origin (which = GTOs do not have)

[2] an exponential decrease at = infinity (which GTO do not have)

 

The “flip side of the = coin” is, multi-center integrals (which is the heart of ab initio = calculations) over STOs is much much more difficult than with GTOs. =

 

Some people (including myself = –but just for the challenge-) are still trying to setup techniques to use STOs =

 

Cheers

 

From: = owner-chemistry.:.ccl.net = [mailto:owner-chemistry.:.ccl.net]
Sent: Tuesday, September = 06, 2005 8:35 AM
To: Bouferguene, Ahmed =
Subject: CCL: W:GTO and = STO

 

A product of = two GAUSSIANS is another GAUSSIAN somewhere in between,
resulting in at most 2 center integrals. This is not the case for = Slaters.
Also I would imgine, that the fact that gaussians fall off more quickly = makes
integration easier.
Joop
There must be papers by e.g. V.R.Saunders in proceedinmgs of = summerschools.
On Sep 6, 2005, at 15:37, CCL wrote:


Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de]
Hello,


can anyone tell me, why calculations with Gaussiantype-Orbitals are = faster than calculations with STO? Is there literature where this is described in = detail?

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
Joop van Lenthe
Theoretical Chemistry Group
Debye Institute, Utrecht University
Padualaan 8 3584 CH Utrecht
-31-30-2532733
joop.:.chem.uu.nl
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

------=_NextPart_000_0019_01C5B2D8.D9E01AB0-- From owner-chemistry@ccl.net Tue Sep 6 15:31:26 2005 From: "CCL" To: CCL Subject: CCL: W:Symmetry "Gaussian" Message-Id: <-29082-050906151821-5009-2NbaQmdzjXHCm+8GjNpz2w+*+server.ccl.net> X-Original-From: "Shobe, David" content-class: urn:content-classes:message Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Tue, 6 Sep 2005 14:16:03 -0400 MIME-Version: 1.0 Sent to CCL by: "Shobe, David" [dshobe+*+sud-chemieinc.com] Try symm=loose. That works in G98. Probably Gaussian did not recognize the relevant bond angles as being "close enough" to 120 degrees. --David Shobe, Ph.D., M.L.S. Süd-Chemie, Inc. phone (502) 634-7409 fax (502) 634-7724 Don't bother flaming me: I'm behind a firewall. -----Original Message----- > From: owner-chemistry+*+ccl.net [mailto:owner-chemistry+*+ccl.net] Sent: Monday, September 05, 2005 3:08 PM To: Shobe, David Subject: CCL: W:Symmetry "Gaussian" Sent to CCL by: "Roma Oakes" [r.e.oakes=-=btconnect.com] Hi - the (pg=) keyword is not functional in G98, have a look at my website for some tips on how to get G98 to find symmetry, Roma Dr Roma E Oakes Visiting Research Fellow School of Chemistry The Queen's University of Belfast Visit my Website at; http://home.btconnect.com/reoakes/ -----Original Message----- > From: owner-chemistry+*+ccl.net [mailto:owner-chemistry+*+ccl.net] Sent: 05 September 2005 17:30 To: Oakes, roma Subject: CCL: W:Symmetry "Gaussian" Sent to CCL by: James Kirkpatrick [james.kirkpatrick^_^imperial.ac.uk] hi ali, I am a student too and am happy to give some tips on point group editing. can you use gaussview? It has a convenient point group editor. D3d and C2h are quite different point groups if i am not mistaken, C2h involves a horinzonal reflection plane, whereas D3d, involves a three fold "staggered" symmetry. So I think it is possible that a molecule has both C2h and D3d symmetry. I think that D3d has "more" symmetry than C2h. What I would suggest is to, a) check that your .xyz actually has the symmetry you expect, i.e. the three rotations and the three diehedral rotations. b) use the symm(pg=D3d) keyword This works in g03, I am not sure it will in g98. Hope this is of some use. Good site to visulise point group s is: http://caramel.oc.chemie.tu-darmstadt.de/immel/script/redirect.cgi or make a search for Dr immel's guide to point groups. Hope this is useful! James CCL wrote: >Sent to CCL by: "Ali Salimi" [salimi_ali2002_-_yahoo.com] >Dear CCLers > >I am student. I use G98 for calculations. >I have a problem about symmetry of my strucutre. D3d point group must be for my calculations because I want to obtain normal mode for my compound. >But C2h point group has been done in my caculation. >I want to know, How do I D3d symmetry for my compound. I use XYZ for input file. > >Thanks a lot for your attentions. > >Your sincerely >A. Salimi > >E-mail:salimi_ali2002+*+yahoo.com> > > > > > -- James Kirkpatrick ------------------------------------------- Centre for Electronic Materials and devices Imperial College ------------------------------------------- 020 759 47519 From owner-chemistry@ccl.net Tue Sep 6 15:31:26 2005 From: "CCL" * To: CCL Subject: CCL: W:Any open-source molecular descriptor software for short peptides? Message-Id: <-29083-050906140550-14605-2NbaQmdzjXHCm+8GjNpz2w>* X-Original-From: "Lei Huang" * Sent to CCL by: "Lei Huang" [hxbus]* Dear CCLers: Could anyone recommend a nice open-source molecular descriptor calculation software for short peptides with 8-15 amino acid long? I tried to search the archives but only got the internal error message. Thanks. Lei Huang University of Illinois at Chicago (M/C563) Chicago, IL 60612 Email: hxbus at yahoo.com From owner-chemistry@ccl.net Tue Sep 6 15:31:26 2005 From: "CCL" To: CCL Subject: CCL: Re) W:GTO and STO Message-Id: <-29084-050906121736-32455-2NbaQmdzjXHCm+8GjNpz2w###server.ccl.net> X-Original-From: "Telkuni Tsuru" Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="iso-2022-jp" Date: Wed, 7 Sep 2005 00:25:31 +0900 MIME-Version: 1.0 Sent to CCL by: "Telkuni Tsuru" [telkuni###venus.dti.ne.jp] Hello, Torsten. When GTO and STO are integrated, their scheme are far different. Integrated GTO takes much easier scheme than integrated STO. Therefore GTO calc is faster than STO. Sorry, I don't have English literature for this description(Most of them are Japanese.) But this matter is basic knowledge of quantum chemistry, so you can find this description on beginner's book. Sincerely yours, ---------------------------------------------------- Telkuni Tsuru telkuni###venus.dti.ne.jp ----- Original Message ----- > From: "CCL" To: "Tsuru, Telkuni " Sent: Tuesday, September 06, 2005 10:33 PM Subject: CCL: W:GTO and STO > > Sent to CCL by: "Torsten Bruhn" [torsten.bruhn(!)mail.uni-oldenburg.de] > Hello, > > > can anyone tell me, why calculations with Gaussiantype-Orbitals are faster than calculations with STO? Is there literature where this is described in detail? >