From owner-chemistry@ccl.net Tue Nov 2 03:43:00 2010 From: "turker akcay turkerakcay(~)hotmail.com" To: CCL Subject: CCL:G: low imaginary freq-IR Message-Id: <-43052-101102034105-6755-v5B1TeLk40DyAuqWPw6+SA^^server.ccl.net> X-Original-From: "turker akcay" Date: Tue, 2 Nov 2010 03:41:03 -0400 Sent to CCL by: "turker akcay" [turkerakcay+*+hotmail.com] Hi everybody; I am study on a phthalocyanine molecule with 210 atoms. I tried oniom method B3LYP/631G:PM3MM for optimization and than freq calculation. I found four imaginary freq: -27.54; -2.74; -2.12; -0.50. I am not sure but I think that the frequencies bellow -50 can ignored. is it true or should I try again with "ultrafine". (I am a gaussian 03 user) could you give me a literature about low imaginary freq. to support my thesis thanks From owner-chemistry@ccl.net Tue Nov 2 04:23:01 2010 From: "partha biswas partha361###gmail.com" To: CCL Subject: CCL:G: BSSE energy difference in G03 and NWChem Message-Id: <-43053-101102042130-2817-6WziWYjpjAt+wlQwOjnxgQ*o*server.ccl.net> X-Original-From: "partha biswas" Date: Tue, 2 Nov 2010 04:21:28 -0400 Sent to CCL by: "partha biswas" [partha361]^[gmail.com] Sir, I have ran the BSSE calculation for water-ammonia cluster in NWCHem as well as in Gaussian03.I have used the dft method m05-2x/6-31g as the basis set. Both the jobs ran successfully with following results: In NWChem BSSE Error= 0.001774669 Corrected energy = -132.89948836 In Gaussian BSSE energy = 0.00165022 Corrected energy = -132.89804157 My concern is that the corrected energies values differ by a value of 3KJ/Mol(0.9kcal/mol) whereas the BSSE errors differ by 0.32kJ/mol(0.077kcal/mol). So what is the reason for huge difference in corrected energies ..!!. Moreover I have ran the BSSE Calculation in NWchem and Gaussian03 with HF method also.There i have found the results quite similar in Gaussian and NWChem. In NWChem(HF/6-31g) BSSE Error= 0.001387039 Corrected energy = -132.162177794 In Gaussian03(HF/6-31g) BSSE Error= 0.0013870449 Corrected energy = -132.162177768 Please see this input file for NWChem scratch_dir /scratch title "waterammonia bsse calc " start waterammonia ##memory global 1024 mb stack 2048 mb heap 2048 mb geometry O -1.54122800 -0.09769000 0.00050800 H -0.58500900 -0.00294800 -0.00853800 H -1.98167500 0.74227700 -0.00278600 N 1.41171900 0.01749500 -0.00026300 H 1.58776600 -0.46249000 0.85508000 H 1.83081200 0.92026600 -0.03925000 H 1.59589900 -0.53804600 -0.80673200 end basis "ao basis" spherical O library 6-31g H library 6-31g N library 6-31g bqO library O 6-31g bqH library H 6-31g bqN library N 6-31g end driver clear end bsse mon water 1 2 3 mon ammonia 4 5 6 7 end task scf optimize From owner-chemistry@ccl.net Tue Nov 2 05:45:00 2010 From: "David Danovich david.danovich^_^huji.ac.il" To: CCL Subject: CCL: dispersion correction in G09 Message-Id: <-43054-101102053311-9058-Gv0dSkACF+VUjYZbmvNKmA[A]server.ccl.net> X-Original-From: "David Danovich" Content-Transfer-Encoding: 7bit Content-Type: text/plain; format=flowed; charset="iso-8859-1"; reply-type=original Date: Tue, 2 Nov 2010 11:33:07 +0200 MIME-Version: 1.0 Sent to CCL by: "David Danovich" [david.danovich-x-huji.ac.il] This is the query about Dispersion correction: I calculated single point calculation for dispersion correction in G09 using iop(3/124=3). I would like to know whether G09 can do geometry optimization with +D correction factor using B3LYP functional. I noticed from the reference of Stefan Grimme that B3LYP functional has a scaling factor of 1.05 to the dispersion correction. Is this scale factor included in optimization with +D? Does the gradients of energy also corrected with dispersion? It will be great if you can clarify me about these points. Thank you for the help, Regards Usha -- D. Usharani c/o Prof. Sason Shaik, Institute of Chemistry The Hebrew University of Jerusalem, Givat Ram Campus, Jerusalem. ushadandamudi.rani : gmail.com FAX: +972-2-6584033 From owner-chemistry@ccl.net Tue Nov 2 08:30:00 2010 From: "Holger Kruse holger.kruse*_*uni-muenster.de" To: CCL Subject: CCL:G: dispersion correction in G09 Message-Id: <-43055-101102074220-20327-djHQ9ISeOlAZ8QJvKJTNnQ]![server.ccl.net> X-Original-From: Holger Kruse Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Tue, 02 Nov 2010 12:42:10 +0100 MIME-Version: 1.0 Sent to CCL by: Holger Kruse [holger.kruse]^[uni-muenster.de] Dear Usha, The current G09 is not capable of doing B3LYP-D calculations (correctly). (Only B97-D and B2PLYP-D I think) However, the manual states: IOp(3/124) Empirical dispersion term. 0 Default (same as 2). 1 Add it regardless. 2 Add it for the DFT functionals for which it has been defined and parameterized and for which a specific name has been defined in Link1. 3 Add it for the DFT functionals for which it has been defined and parameterized. 4 Do not add it regardless. This IOp is not working correctly! iop(3/124=3) adds the dispersion correction with a S6 scaling factor of 1.00 (the default), no matter if there exist a parameterized functional. You can see that if you add IOp33(3=3) to you input. The output will show somewhere: -- R6Disp: E=T FX=F FFX=F UseDF=T IUseCN= 2 R6DPar: UseDF=T IExCor= 402 ScaHFX= 0.200000 S6= 1.000000 D= 20.000000 -- Fortunately, this can be fixed in the source code very easily (One logical is 'wrong'). If you have access to the source that is. Feel free to contact me for help. Maybe it would also help to complain about this to the Gaussian people. For single point calculations you currently have to scale the dispersion energy by hand, too. Regards, Holger Kruse On 11/02/2010 10:33 AM, David Danovich david.danovich^_^huji.ac.il wrote: > > Sent to CCL by: "David Danovich" [david.danovich-x-huji.ac.il] > This is the query about Dispersion correction: > > I calculated single point calculation for dispersion correction in G09 using > iop(3/124=3). I would like to know whether G09 can do geometry optimization with +D > correction factor using B3LYP functional. I noticed from the reference of > Stefan Grimme that B3LYP functional has a scaling factor of 1.05 to the > dispersion correction. Is this scale factor included in optimization with > +D? Does the gradients of energy also corrected with dispersion? > It will be great if you can clarify me about these points. > > Thank you for the help, > > Regards > Usha > From owner-chemistry@ccl.net Tue Nov 2 09:05:00 2010 From: "Chenghua Zhang zchua126.com]^[126.com" To: CCL Subject: CCL: low imaginary freq-IR Message-Id: <-43056-101102055035-29440-aG3Y9N5jKyF1D8pdWlWCcQ~~server.ccl.net> X-Original-From: "Chenghua Zhang" Content-Type: multipart/alternative; boundary="----=_Part_115140_25846690.1288691414351" Date: Tue, 2 Nov 2010 17:50:14 +0800 (CST) MIME-Version: 1.0 Sent to CCL by: "Chenghua Zhang" [zchua126.com+*+126.com] ------=_Part_115140_25846690.1288691414351 Content-Type: text/plain; charset=gbk Content-Transfer-Encoding: 7bit Hi, as far as I know, it doesn't matter in the mechanism calculation if there are some low imaginary. ref: 1: JCP,1993,98,5648 2: JPCA,2003,107,11445 -- Sincerely Chenghua Zhang College of Chemistry Sichuan University, China. ------=_Part_115140_25846690.1288691414351 Content-Type: text/html; charset=gbk Content-Transfer-Encoding: 7bit Hi,
 as far as I know, it doesn't matter in the mechanism calculation if there are some low imaginary.
ref: 1: JCP,1993,98,5648
2: JPCA,2003,107,11445

--
Sincerely
Chenghua Zhang
College of Chemistry
Sichuan University, China.




------=_Part_115140_25846690.1288691414351-- From owner-chemistry@ccl.net Tue Nov 2 10:10:00 2010 From: "Marcel Swart marcel.swart[#]icrea.cat" To: CCL Subject: CCL:G: dispersion correction in G09 Message-Id: <-43057-101102100842-20358-fVfUhFkupBPKBJcErUR73g^-^server.ccl.net> X-Original-From: Marcel Swart Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=iso-8859-1 Date: Tue, 2 Nov 2010 15:08:30 +0100 Mime-Version: 1.0 (Apple Message framework v1081) Sent to CCL by: Marcel Swart [marcel.swart-*-icrea.cat] This has been corrected in version B.01. On Nov 2, 2010, at 12:42 PM, Holger Kruse holger.kruse*_*uni-muenster.de wrote: > > Sent to CCL by: Holger Kruse [holger.kruse]^[uni-muenster.de] > Dear Usha, > > The current G09 is not capable of doing B3LYP-D calculations (correctly). > (Only B97-D and B2PLYP-D I think) > > However, the manual states: > > IOp(3/124) > Empirical dispersion term. > 0 Default (same as 2). > 1 Add it regardless. > 2 Add it for the DFT functionals for which it has been defined and parameterized > and for which a specific name has been defined in Link1. > 3 Add it for the DFT functionals for which it has been defined and parameterized. > 4 Do not add it regardless. > > This IOp is not working correctly! > iop(3/124=3) adds the dispersion correction with a S6 scaling > factor of 1.00 (the default), no matter if there exist a parameterized functional. > You can see that if you add IOp33(3=3) to you input. The output will show somewhere: > -- > R6Disp: E=T FX=F FFX=F UseDF=T IUseCN= 2 > R6DPar: UseDF=T IExCor= 402 ScaHFX= 0.200000 S6= 1.000000 D= 20.000000 > -- > > Fortunately, this can be fixed in the source code very easily (One logical is 'wrong'). If you have access to the source that is. > Feel free to contact me for help. > > Maybe it would also help to complain about this to the Gaussian people. > > For single point calculations you currently have to scale the dispersion energy by hand, too. > > Regards, > Holger Kruse =================================== dr. Marcel Swart ICREA Research Professor at Institut de Química Computacional Universitat de Girona Parc Científic i Tecnològic Edifici Jaume Casademont (despatx A-27) Pic de Peguera 15 17003 Girona Catalunya (Spain) tel +34-972-183240 fax +34-972-183241 e-mail marcel.swart,,icrea.cat marcel.swart,,udg.edu web http://www.marcelswart.eu ===================================