From owner-chemistry@ccl.net Wed Mar 7 03:01:00 2007 From: "Sten Nilsson Lill stenil*o*chem.gu.se" To: CCL Subject: CCL:G: G03 Optimization problem Message-Id: <-33755-070306210626-9989-zAtfsLsAXu3fKNehdoFjMw~!~server.ccl.net> X-Original-From: "Sten Nilsson Lill" Content-Transfer-Encoding: 8bit Content-Type: text/plain;charset=iso-8859-1 Date: Wed, 7 Mar 2007 02:10:28 +0100 (CET) MIME-Version: 1.0 Sent to CCL by: "Sten Nilsson Lill" [stenil{=}chem.gu.se] Dear Aniel, you can try to increase the accuracy in the early phase of the scf by using scf=novaracc. If it doesn't work, let the program make a new guess, i.e. take guess=read away from your input and include also scf=novaracc. If you are running unrestricted DFT just be careful to check that your spin distribution doesn't change too much from your current values when the program makes a new guess. Hope this gives some help! Regards, Sten Nilsson Lill > > Hi All, > I have one problem and one question; first on "optimization" and second on > "IOps". > > 1) I was trying to optimize a small metallic clusters (6-8 atoms) using > SDD basis set and PBE1PBE using G03. However, I encounter some problems > with the optimization and quit the calculation with the following error. > Error termination request processed by link 9999. > > The energy is very close to converge, but the Maximum Force and RMS Force > do not converge. > The error message is: > Item Value Threshold > Converged? > Maximum Force 0.002180 0.000450 NO > RMS Force 0.000952 0.000300 NO > Maximum Displacement 0.000010 0.001800 YES > RMS Displacement 0.000005 0.001200 YES > Predicted change in Energy=-2.455189D-05 > Optimization stopped. > -- Number of steps exceeded, NStep= 31 > -- Flag reset to prevent archiving. > > grep "SCF D" {last 4 steps} > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 3 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 2 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917095 A.U. after 4 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 4 cycles > > I restarted the calculation from chk file; but again, the Max/RMS Force do > not converge. Could you please tell me what modifications I should include > in the "input" file? > (input details for 6 atoms: > #P pbe1pbe/SDD opt=(calcfc,maxstep=50) symm=(PG=Oh) geom=checkpoint > guess=read) > > 2) What exactly is IOps? I checked the G03 manual, but did not understand > the logic behind in choosing the correct numerator and denominator and > integer after the "=" sign. How can I find out the correct/proper > values/numbers to use in IOps? For example, Does the following IOp > statement has any special meaning? > IOp(5/42=10000428,5/43=05720572,5/44=10001000) > > I would really take any feedback from you. Thanks in advance. > > Sincerely, > Aniel > > > > > > Send instant messages to your online friends http://uk.messenger.yahoo.com Ph. D. Sten Nilsson Lill Dep. of Chemistry Göteborg University Kemigården 4 S-412 96 Göteborg, Sweden Phone: +46-31-772 2901 Fax: +46-31-772 3840 E-mail: stenil]^[chem.gu.se Alternative e-mail: slill1]^[lsu.edu From owner-chemistry@ccl.net Wed Mar 7 05:47:00 2007 From: "Ioan SILAGHI-DUMITRESCU isi*_*chem.ubbcluj.ro" To: CCL Subject: CCL:G: G03 Optimization problem Message-Id: <-33756-070307051945-862-vuw0iiSFXXZWZYxzEDJfdg.@.server.ccl.net> X-Original-From: Ioan SILAGHI-DUMITRESCU Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed Date: Wed, 7 Mar 2007 12:09:12 +0200 (EET) MIME-Version: 1.0 Sent to CCL by: Ioan SILAGHI-DUMITRESCU [isi*chem.ubbcluj.ro] Aniel, Reducing the symmetry of the system might solve the first problem. I encountered similar sitations when the cluster have a C5 axes and small deviations of the coordinates from the perfect symmetry lead to normal tremination of the optimization. For the second question the Technical Support Information of Gaussian.com (http://www.gaussian.com/iops.htm) Hope this helps, isi --------------------------------------------------------------------- Prof.dr. Ioan Silaghi-Dumitrescu Department of Chemistry | Facultatea de Chimie si Inginerie Chimica Babes-Bolyai University | Universitatea Babes-Bolyai 1, Kogalniceanu str. | str. Kogalniceanu, nr.1 RO-400084 Cluj-Napoca, | RO-400084 Cluj-Napoca Roumania | Romania e-mail: isi.:.chem.ubbcluj.ro Fax: +40-264-590818 Tel: +40-264-593833 ext.5665 On Tue, 6 Mar 2007, Aniel Aron anisamron+*+yahoo.com wrote: > Hi All, I have one problem and one question; first on "optimization" and second on "IOps". 1) I was trying to optimize a small metallic clusters (6-8 atoms) using SDD basis set and PBE1PBE using G03. However, I encounter some problems with the optimization and quit the calculation with the following error. Error termination request processed by link 9999. The energy is very close to converge, but the Maximum Force and RMS Force do not converge. The error message is: Item Value Threshold Converged? Maximum Force 0.002180 0.000450 NO RMS Force 0.000952 0.000300 NO Maximum Displacement 0.000010 0.001800 YES RMS Displacement 0.000005 0.001200 YES Predicted change in Energy=-2.455189D-05 Optimization stopped. -- Number of steps exceeded, NStep= 31 -- Flag reset to prevent archiving. grep "SCF D" {last 4 steps} SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 3 cycles SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 2 cycles SCF Done: E(UPBE+HF-PBE) = -1017.54917095 A.U. after 4 cycles SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 4 cycles I restarted the calculation from chk file; but again, the Max/RMS Force do not converge. Could you please tell me what modifications I should include in the "input" file? (input details for 6 atoms: #P pbe1pbe/SDD opt=(calcfc,maxstep=50) symm=(PG=Oh) geom=checkpoint guess=read) 2) What exactly is IOps? I checked the G03 manual, but did not understand the logic behind in choosing the correct numerator and denominator and integer after the "=" sign. How can I find out the correct/proper values/numbers to use in IOps? For example, Does the following IOp statement has any special meaning? IOp(5/42=10000428,5/43=05720572,5/44=10001000) I would really take any feedback from you. Thanks in advance. Sincerely, Aniel Send instant messages to your online friends http://uk.messenger.yahoo.com From owner-chemistry@ccl.net Wed Mar 7 08:33:01 2007 From: "Ulrike Salzner salzner(-)fen.bilkent.edu.tr" To: CCL Subject: CCL:G: G03 Optimization problem Message-Id: <-33757-070307063139-25745-npaEEgkIzIjXbLvJfIMzTQ,,server.ccl.net> X-Original-From: Ulrike Salzner Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset=ISO-8859-1; format=flowed Date: Wed, 07 Mar 2007 13:31:13 +0200 MIME-Version: 1.0 Sent to CCL by: Ulrike Salzner [salzner++fen.bilkent.edu.tr] Hi, it looks to me like you have an extremly flat energy potential energy surface. Your total energy is not changing anymore and the forces are small while there are significant displacements. You can increase the number of steps by defining "MAXCYCLE=N" and choose N larger than 31. Maybe the optimization will converge at Step 35 or so. However, in your case this might lead nowhere since the molecule is floppy. I would check the geometry changes and if they are not very big I would just stop the calculation and accept the physics. The threshold values in the program are reasonable for most cases but might not be suitable for every molecule. Tegards, Ulrike Salzner > Hi All, > I have one problem and one question; first on "optimization" and > second on "IOps". > > 1) I was trying to optimize a small metallic clusters (6-8 atoms) > using SDD basis set and PBE1PBE using G03. However, I encounter some > problems with the optimization and quit the calculation with the > following error. Error termination request processed by link 9999. > > The energy is very close to converge, but the Maximum Force and RMS > Force do not converge. > The error message is: > Item Value Threshold > Converged? > Maximum Force 0.002180 0.000450 NO > RMS Force 0.000952 0.000300 NO > Maximum Displacement 0.000010 0.001800 YES > RMS Displacement 0.000005 0.001200 YES > Predicted change in Energy=-2.455189D-05 > Optimization stopped. > -- Number of steps exceeded, NStep= 31 > -- Flag reset to prevent archiving. > > grep "SCF D" {last 4 steps} > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 3 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 2 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917095 A.U. after 4 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 4 cycles > > I restarted the calculation from chk file; but again, the Max/RMS > Force do not converge. Could you please tell me what modifications I > should include in the "input" file? > (input details for 6 atoms: #P pbe1pbe/SDD opt=(calcfc,maxstep=50) > symm=(PG=Oh) geom=checkpoint guess=read) > > 2) What exactly is IOps? I checked the G03 manual, but did not > understand the logic behind in choosing the correct numerator and > denominator and integer after the "=" sign. How can I find out the > correct/proper values/numbers to use in IOps? For example, Does the > following IOp statement has any special meaning? > IOp(5/42=10000428,5/43=05720572,5/44=10001000) > > I would really take any feedback from you. Thanks in advance. > > Sincerely, > Aniel > > > > > > Send instant messages to your online friends > http://uk.messenger.yahoo.comhttp://www.ccl.net/chemistry/sub_unsub.shtmlConferences: > http://server.ccl.net/chemistry/announcements/conferences/> > > > > From owner-chemistry@ccl.net Wed Mar 7 10:27:01 2007 From: "Taye Beyene Demisse sene3095.:.yahoo.com" To: CCL Subject: CCL: Sever error message2070 Message-Id: <-33758-070307102159-9521-3Bnb3hT/gN+1MDQz5EXzkg^_^server.ccl.net> X-Original-From: "Taye Beyene Demisse" Date: Wed, 7 Mar 2007 10:21:56 -0500 Sent to CCL by: "Taye Beyene Demisse" [sene3095%x%yahoo.com] Dear subscribers, when I am calculating frequency for a molecule containing 227 alpha and 227 beta electrons, it terminates by reporting "severe error message2070". What is this message? is there any material that clarify error message types? Please help. Thanks. Taye From owner-chemistry@ccl.net Wed Mar 7 11:01:01 2007 From: "Stradella, Omar omar.stradella{=}intel.com" To: CCL Subject: CCL: calculating titration state distributions for multiprotic acids with coupled sites Message-Id: <-33759-070303113411-16175-BtWA30TPkRyyKhhiK5KjLw*_*server.ccl.net> X-Original-From: "Stradella, Omar" Content-class: urn:content-classes:message Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="us-ascii" Date: Sat, 3 Mar 2007 10:58:57 -0500 MIME-Version: 1.0 Sent to CCL by: "Stradella, Omar" [omar.stradella .. intel.com] On Fri, Feb 23, 2007, Seth Olsen s.olsen1,,uq.edu.au wrote: > > I have been trying to find references or algorithms which can estimate > equilibrium distributions of different protonation states of a > multiprotic acid where the sites may be coupled. Can anyone point me in > the right direction? In the past I've used MCCE (Multi-Conformation Continuum Electrostatics) > from Marilyn Gunner at CUNY with very good results. http://www.sci.ccny.cuny.edu/~mcce/ Omar -------------------- Omar Stradella, PhD Life Sciences Applications Engineer Intel Corp. 77 Reed Rd. Hudson, MA 01749 Email: omar.stradella*o*intel.com Phone: +1 978-553-5388 From owner-chemistry@ccl.net Wed Mar 7 11:36:01 2007 From: "Soren Eustis soren _ jhu.edu" To: CCL Subject: CCL:G: G03 Optimization problem Message-Id: <-33760-070307112435-14737-zP4Of7hiYRzip7Yg9i7+iw_._server.ccl.net> X-Original-From: "Soren Eustis" Content-language: en-us Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Wed, 7 Mar 2007 10:24:06 -0500 MIME-Version: 1.0 Sent to CCL by: "Soren Eustis" [soren]|[jhu.edu] This does not look like an SCF convergence problem, rather it is an optimization convergence problem. You need to increase the number of optimization cycles. Add OPT=MAXCYCLE=100 to your input. Also, You can try OPT=(NRScale,Maxcycle=100) if you still have issues. Soren N. Eustis Graduate Research Assistant Department of Chemistry Johns Hopkins University Remsen Hall 3400 N Charles Street Baltimore, MD 21218 (410) 516-4675 (office) (410) 925-5167 (cell) (410) 516-8420 (fax) soren() jhu.edu -----Original Message----- > From: owner-chemistry() ccl.net [mailto:owner-chemistry() ccl.net] Sent: Tuesday, March 06, 2007 8:10 PM To: Eustis, Soren Subject: CCL:G: G03 Optimization problem Sent to CCL by: "Sten Nilsson Lill" [stenil{=}chem.gu.se] Dear Aniel, you can try to increase the accuracy in the early phase of the scf by using scf=novaracc. If it doesn't work, let the program make a new guess, i.e. take guess=read away from your input and include also scf=novaracc. If you are running unrestricted DFT just be careful to check that your spin distribution doesn't change too much from your current values when the program makes a new guess. Hope this gives some help! Regards, Sten Nilsson Lill > > Hi All, > I have one problem and one question; first on "optimization" and second on > "IOps". > > 1) I was trying to optimize a small metallic clusters (6-8 atoms) using > SDD basis set and PBE1PBE using G03. However, I encounter some problems > with the optimization and quit the calculation with the following error. > Error termination request processed by link 9999. > > The energy is very close to converge, but the Maximum Force and RMS Force > do not converge. > The error message is: > Item Value Threshold > Converged? > Maximum Force 0.002180 0.000450 NO > RMS Force 0.000952 0.000300 NO > Maximum Displacement 0.000010 0.001800 YES > RMS Displacement 0.000005 0.001200 YES > Predicted change in Energy=-2.455189D-05 > Optimization stopped. > -- Number of steps exceeded, NStep= 31 > -- Flag reset to prevent archiving. > > grep "SCF D" {last 4 steps} > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 3 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 2 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917095 A.U. after 4 cycles > SCF Done: E(UPBE+HF-PBE) = -1017.54917094 A.U. after 4 cycles > > I restarted the calculation from chk file; but again, the Max/RMS Force do > not converge. Could you please tell me what modifications I should include > in the "input" file? > (input details for 6 atoms: > #P pbe1pbe/SDD opt=(calcfc,maxstep=50) symm=(PG=Oh) geom=checkpoint > guess=read) > > 2) What exactly is IOps? I checked the G03 manual, but did not understand > the logic behind in choosing the correct numerator and denominator and > integer after the "=" sign. How can I find out the correct/proper > values/numbers to use in IOps? For example, Does the following IOp > statement has any special meaning? > IOp(5/42=10000428,5/43=05720572,5/44=10001000) > > I would really take any feedback from you. Thanks in advance. > > Sincerely, > Aniel > > > > > > Send instant messages to your online friends http://uk.messenger.yahoo.com Ph. D. Sten Nilsson Lill Dep. of Chemistry Göteborg University Kemigården 4 S-412 96 Göteborg, Sweden Phone: +46-31-772 2901 Fax: +46-31-772 3840 E-mail: stenil|*|chem.gu.se Alternative e-mail: slill1|*|lsu.eduhttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt From owner-chemistry@ccl.net Wed Mar 7 12:11:00 2007 From: "Shobe, David David.Shobe!^!sud-chemie.com" To: CCL Subject: CCL: Sever error message2070 Message-Id: <-33761-070307113231-18408-9N2ibROBQDE+5nCBqjQjIg#,#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: Wed, 7 Mar 2007 17:32:17 +0100 MIME-Version: 1.0 Sent to CCL by: "Shobe, David" [David.Shobe^sud-chemie.com] What comp-chem program and operating system are you using? Regards, --David Shobe Süd-Chemie, Inc. Attention to detial is essentail. -----Original Message----- > From: owner-chemistry : ccl.net [mailto:owner-chemistry : ccl.net] Sent: Wednesday, March 07, 2007 10:22 AM To: Shobe, David Subject: CCL: Sever error message2070 Sent to CCL by: "Taye Beyene Demisse" [sene3095%x%yahoo.com] Dear subscribers, when I am calculating frequency for a molecule containing 227 alpha and 227 beta electrons, it terminates by reporting "severe error message2070". What is this message? is there any material that clarify error message types? Please help. Thanks. Tayehttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txtThis e-mail message may contain confidential and / or privileged information. If you are not an addressee or otherwise authorized to receive this message, you should not use, copy, disclose or take any action based on this e-mail or any information contained in the message. If you have received this material in error, please advise the sender immediately by reply e-mail and delete this message. Thank you. From owner-chemistry@ccl.net Wed Mar 7 13:36:00 2007 From: "Jose Gamez joseantonio.gamez|a|uam.es" To: CCL Subject: CCL:G: Restart a scan from a point not converged Message-Id: <-33762-070307112957-17253-8q9ITy1fTy7pnOdSkMNtgg],[server.ccl.net> X-Original-From: "Jose Gamez" Date: Wed, 7 Mar 2007 11:29:54 -0500 Sent to CCL by: "Jose Gamez" [joseantonio.gamez,uam.es] Hi, everyone. I would like to ask for some help. I was doing a scan to follow the reaction path of a reaction with Gaussian 03. The eighth point didn't converge as it exceeded the limit number of optimisation cycles, having then the ninth and so forth really wierd strucutres. Is there any way in which i could restart the scan from the eighth point? I mean, could i choose the point from which i can restart it? That makes me think if there is some options or anywith to tell Gaussian to stop a scan when one of the points is not fully converged Thanks for your help Cheers Jos A. Gmez joseantonio.gamezuam.es Departamento de Qumica. Universidad Autnoma de Madrid. Spain From owner-chemistry@ccl.net Wed Mar 7 14:11:00 2007 From: "Peeter Burk peeter.burk ~ ut.ee" To: CCL Subject: CCL: NH4NO3 Structure Message-Id: <-33763-070304033107-16631-me43TqNcOclUSZPuJTISnQ-*-server.ccl.net> X-Original-From: Peeter Burk Content-Type: TEXT/PLAIN; charset=US-ASCII Date: Sun, 4 Mar 2007 09:54:45 +0200 (EET) MIME-Version: 1.0 Sent to CCL by: Peeter Burk [peeter.burk]~[ut.ee] Hi, It seems that HNO3 is a too weak acid (and NH3 is too weak base) to give you proton transfered product (i.e. salt). That means, if the hydrogen bonded acid-base pair is more stable than ion pair, thats what you get. The suggestion of adding some explicit water molecules might help as it usually stabilizes both anion and cation... Best regards Peeter On Fri, 2 Mar 2007, Cory Pye cpye|*|crux.smu.ca wrote: > > Sent to CCL by: Cory Pye [cpye]~[crux.smu.ca] > Hi, > > My quess is either that you will need to add a few explicit water molecules to > solvate the cluster and allow the ions to form, or to implicitly add a solvent > using a variant of PCM, perhaps. Glycine zwitterion has the same problem in > vacuo. > > -Cory Pye > > On Fri, 2 Mar 2007, Young Leh youngleh * gmail.com wrote: > > > > > Sent to CCL by: "Young Leh" [youngleh*|*gmail.com] > > Dear CCLer, > > > > I have problem while optimizing the structure of NH4NO3. It looks like no matter what my initial structure is, the structure always ends up with CH3 & HNO3. Could somebody please provide me some help on this, a xyz or pdb file will greatly help? > > > > Thanks all > > > > Young Leh > > > > ************* ! Dr. Cory C. Pye > ***************** ! Associate Professor > *** ** ** ** ! Theoretical and Computational Chemistry > ** * **** ! Department of Chemistry, Saint Mary's University > ** * * ! 923 Robie Street, Halifax, NS B3H 3C3 > ** * * ! cpye++crux.stmarys.ca http://apwww.stmarys.ca/~cpye > *** * * ** ! Ph: (902)-420-5654 FAX:(902)-496-8104 > ***************** ! > ************* ! Les Hartree-Focks (Apologies to Montreal Canadien Fans)> > > > > > -- > This message was scanned for spam and viruses by BitDefender. > For more information please visit http://linux.bitdefender.com/ > > --- Peeter Burk Jakobi 2, 51014 Tartu, Estonia Institute of Chemical Physics Phone (372-7) 375-258 Tartu University Fax (372-7) 375-264 Estonia E-mail peeter..chem.ut.ee -- This message was scanned for spam and viruses by BitDefender. For more information please visit http://linux.bitdefender.com/ From owner-chemistry@ccl.net Wed Mar 7 14:47:01 2007 From: "Michael Potter potterm]|[verachem.com" To: CCL Subject: CCL: Free tools for conformer generation and partial charge assignment Message-Id: <-33764-070307114150-25619-kiEfkebHFd61QOx/G0NPzQ|server.ccl.net> X-Original-From: "Michael Potter" Date: Wed, 7 Mar 2007 11:41:47 -0500 Sent to CCL by: "Michael Potter" [potterm^^verachem.com] VeraChem is pleased to announce the release of three new products. All VeraChem products are now licensed free of charge to academic and other non-profit users! Vconf, conformational search for drug-like molecules. In "prep" mode, Vconf quickly generates 3D structures suitable for docking and other torsional search applications, with multiple outputs for distinct ring-puckers. In "search" model, Vconf carries out thorough conformational searches and provides high-quality, low-energy conformations. Special features include: - filtering to remove repeat conformations from the output, accounting for internal molecular symmetries - option to hold a set of atoms rigidly fixed in the input conformation - detailed control of chiral centers and double bond geometries. For more information and download, go to http://www.verachem.com/vconf.html. Vfilter, conformational filtering for drug-like molecules. Processes the output of Vconf or any other conformational analysis program to remove repeat conformations based upon energy or RMSD, accounting for symmetry. For more information and download, go to http://www.verachem.com/vfilter.html. Vrms, calculation of RMSDs of conformations of a drug-like molecule, accounting for symmetries. RMSD with superposition is suitable for free ligands in solution; RMSD without superposition is suitable for docked ligands. For more information and download, go to http://www.verachem.com/vrms.html Vcharge, rapid generation of RESP-like partial charges, is also available. For more information and download, go to http://www.verachem.com/vcharge.html. Please feel free to contact me for additional information. Regards, Michael J. Potter VeraChem LLC potterm()verachem.com From owner-chemistry@ccl.net Wed Mar 7 19:38:00 2007 From: "=?UTF-8?B?w5Zkw7ZuIEZhcmthcw==?= farkas!=!chem.elte.hu" To: CCL Subject: CCL:G: G03 Optimization problem Message-Id: <-33765-070307130645-15135-wsnrIz4h9JBtcKiA0WIDVg|*|server.ccl.net> X-Original-From: =?UTF-8?B?w5Zkw7ZuIEZhcmthcw==?= Content-Type: multipart/alternative; boundary="----=_NextPart_000_0015_01C760EB.737AFED0" Date: Wed, 7 Mar 2007 19:04:36 +0100 MIME-Version: 1.0 Sent to CCL by: =?UTF-8?B?w5Zkw7ZuIEZhcmthcw==?= [farkas!^!chem.elte.hu] This is a multi-part message in MIME format. ------=_NextPart_000_0015_01C760EB.737AFED0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Dear Aniel, =20 My guess is different from what the others suggested, the rest of the = output may contain the real answer. It is really unusual that the step = size converges in an optimiyation but the gradient not, so I think this = problem is related to symmetry. You might have indication that the = wavefunction breaks symmetry during the SCF. If this is the case, than = the optimization will still produce a symmetric step, so you should turn = of symmetry and try if it converges or not. =20 The IOps are integer arrays to control the Gaussian links. At the = beginning of the output you can find the route table. Containing the = sequence of links and the assigned IOps, which is created automatically = > from your command line. Using IOps you can alter these values or even = run Gaussian using a non-standard route table (NonStd command). =20 Best wishes, =20 =C3=96d=C3=B6n =20 =C3=96d=C3=B6n Farkas Associate professor Deparment of Organic Chemistry, Institute of Chemistry, E=C3=B6tv=C3=B6s Lor=C3=A1nd University, Budapest Address: 1/A P=C3=A1zm=C3=A1ny P=C3=A9ter s=C3=A9t=C3=A1ny, H-117 = Budapest, Hungary Phone: +36-1-372-2570 Cell phone: +36-30-255-3111 Fax: +36-1-372-2620 URL: http://organ.elte.hu/farkas =20 -----Original Message----- > From: owner-chemistry-,-ccl.net [mailto:owner-chemistry-,-ccl.net]=20 Sent: Tuesday, March 06, 2007 11:07 PM To: Farkas, XdXXn Subject: CCL:G: G03 Optimization problem =20 Hi All, I have one problem and one question; first on "optimization" and second = on "IOps". 1) I was trying to optimize a small metallic clusters (6-8 atoms) using = SDD basis set and PBE1PBE using G03. However, I encounter some problems = with the optimization and quit the calculation with the following error. = =20 Error termination request processed by link 9999. The energy is very close to converge, but the Maximum Force and RMS = Force do not converge. The error message is:=20 Item Value Threshold = Converged? Maximum Force 0.002180 0.000450 NO RMS Force 0.000952 0.000300 NO Maximum Displacement 0.000010 0.001800 YES RMS Displacement 0.000005 0.001200 YES Predicted change in Energy=3D-2.455189D-05 Optimization stopped. -- Number of steps exceeded, NStep=3D 31 -- Flag reset to prevent archiving. grep "SCF D" {last 4 steps} SCF Done: E(UPBE+HF-PBE) =3D -1017.54917094 A.U. after 3 = cycles SCF Done: E(UPBE+HF-PBE) =3D -1017.54917094 A.U. after 2 = cycles SCF Done: E(UPBE+HF-PBE) =3D -1017.54917095 A.U. after 4 = cycles SCF Done: E(UPBE+HF-PBE) =3D -1017.54917094 A.U. after 4 = cycles I restarted the calculation from chk file; but again, the Max/RMS Force = do not converge. Could you please tell me what modifications I should = include in the "input" file? (input details for 6 atoms:=20 #P pbe1pbe/SDD opt=3D(calcfc,maxstep=3D50) symm=3D(PG=3DOh) = geom=3Dcheckpoint guess=3Dread) 2) What exactly is IOps? I checked the G03 manual, but did not = understand the logic behind in choosing the correct numerator and = denominator and integer after the "=3D" sign. How can I find out the = correct/proper values/numbers to use in IOps? For example, Does the = following IOp statement has any special meaning? IOp(5/42=3D10000428,5/43=3D05720572,5/44=3D10001000) I would really take any feedback from you. Thanks in advance. Sincerely, Aniel Send instant messages to your online friends = http://uk.messenger.yahoo.com=20 ------=_NextPart_000_0015_01C760EB.737AFED0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable

Dear Aniel,

 

My guess is different from what the others suggested, the = rest of the output may contain the real answer. It is really unusual that the = step size converges in an optimiyation but the gradient not, so I think this = problem is related to symmetry. You might have indication that the wavefunction = breaks symmetry during the SCF. If this is the case, than the optimization will still = produce a symmetric step, so you should turn of symmetry and try if it converges = or not.

 

The IOps are = integer arrays to control the Gaussian links. At the beginning of the output you = can find the route table. Containing the sequence of links and the assigned = IOps, which is created automatically from your command line. Using IOps you can = alter these values or even run Gaussian using a non-standard route table (NonStd = command).

 

Best = wishes,

 

=C3=96d=C3=B6n

 

=C3=96d=C3=B6n = Farkas

Associate = professor

Deparment of Organic Chemistry, = Institute of Chemistry,

E=C3=B6tv=C3=B6s Lor=C3=A1nd = University, Budapest

Address: 1/A P=C3=A1zm=C3=A1ny = P=C3=A9ter s=C3=A9t=C3=A1ny, H-117 Budapest, Hungary

Phone: = +36-1-372-2570

Cell phone: = +36-30-255-3111

Fax: = +36-1-372-2620

URL: = http://organ.elte.hu/farkas

 

-----Original = Message-----
From: = owner-chemistry-,-ccl.net [mailto:owner-chemistry-,-ccl.net]
Sent: Tuesday, March 06, = 2007 11:07 PM
To: Farkas, XdXXn =
Subject: CCL:G: G03 = Optimization problem

 


Hi All,
I have one problem and one question; first on "optimization" = and second on "IOps".

1) I was trying to optimize a small metallic clusters (6-8 atoms) using = SDD basis set and PBE1PBE using G03. However, I encounter some problems with = the optimization and quit the calculation with the following error.  =
Error termination request processed = by link 9999.

The energy is very close to converge, but the Maximum Force and RMS = Force do not converge.
The error message is:
      =            Item                     Value           = Threshold   Converged?
 Maximum = Force                0.002180     = 0.000450         NO
 RMS        Force            =     0.000952     0.000300         NO
 Maximum Displacement     0.000010     0.001800       =   YES
 RMS     Displacement        0.000005     0.001200       =   YES
 Predicted change in Energy=3D-2.455189D-05
 Optimization stopped.
    -- Number of steps exceeded,  NStep=3D  = 31
    -- Flag reset to prevent archiving.

grep "SCF D" {last 4 steps}
 SCF Done:  = E(UPBE+HF-PBE) =3D  -1017.54917094     A.U. after    3 = cycles
 SCF Done:  E(UPBE+HF-PBE) =3D  -1017.54917094     A.U. after    2 = cycles
 SCF Done:  E(UPBE+HF-PBE) =3D  -1017.54917095     A.U. after    4 = cycles
 SCF Done:  E(UPBE+HF-PBE) =3D  -1017.54917094     A.U. after    4 = cycles

I restarted the calculation from chk file; but again, the Max/RMS Force = do not converge. Could you please tell me what modifications I should include = in the "input" file?
(input details for 6 atoms:
#P pbe1pbe/SDD = opt=3D(calcfc,maxstep=3D50) symm=3D(PG=3DOh) geom=3Dcheckpoint guess=3Dread)

2) What exactly is IOps? I checked the G03 manual, but did not = understand the logic behind in choosing the correct numerator and denominator and = integer after the "=3D" sign. How can I find out the correct/proper values/numbers to use in IOps? For example, Does the following IOp = statement has any special meaning?
IOp(5/42=3D10000428,5/43=3D05720572,5/44=3D10001000)

I would really take any feedback from you. Thanks in advance.

Sincerely,
Aniel


Send instant messages to your online friends = http://uk.messenger.yahoo.com

------=_NextPart_000_0015_01C760EB.737AFED0--