From owner-chemistry@ccl.net Thu Nov 25 01:36:00 2010 From: "Horkel, Ernst ehorkel|-|ioc.tuwien.ac.at" To: CCL Subject: CCL:G: AW: G: Gaussian 09 - Problems with TDDFT geometry optimization Message-Id: <-43222-101125013413-22525-qnFGuHQYivgPpzzCA9hPVQ\a/server.ccl.net> X-Original-From: "Horkel, Ernst" Content-class: urn:content-classes:message Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Thu, 25 Nov 2010 07:35:50 +0100 MIME-Version: 1.0 Sent to CCL by: "Horkel, Ernst" [ehorkel_._ioc.tuwien.ac.at] Hello, yes, this seems to bet he solution, I started a separated job last night and the frequencies are still alive and running :-). Thanks to you and Theis (private email communication) for help! Ernst _______________________________________________________________ Univ.Ass. Dipl.-Ing. Dr.techn. Ernst Horkel Institute of Applied Synthetic Chemistry, Vienna University of Technology Tel.: +43-1-58801-15411 Getreidemarkt 9/163OC, +43-644-60588-7122 A-1060 Vienna, Austria Fax: +43-1-58801-15499 email: ehorkel##ioc.tuwien.ac.at -----Ursprüngliche Nachricht----- Von: owner-chemistry+ehorkel==ioc.tuwien.ac.at##ccl.net [mailto:owner-chemistry+ehorkel==ioc.tuwien.ac.at##ccl.net] Im Auftrag von Jamin Krinsky krinsky.jamin%gmail.com Gesendet: Mittwoch, 24. November 2010 21:28 An: Horkel, Ernst Betreff: CCL:G: Gaussian 09 - Problems with TDDFT geometry optimization Sent to CCL by: Jamin Krinsky [krinsky.jamin#,#gmail.com] Dear Ernst, Nobody has replied to this so I'll give it an educated guess. It could be something about specifying what is actually two-part job, the "opt freq" keyword combination. Gaussian must do numerical frequencies for TD jobs and maybe the keywords generated automatically for the second (frequency) part are not compatible with that method. I would do the two parts separately anytime the first part works and the second doesn't. You already have the optimization part done, so make a copy of the .chk file from that job with a new name (newfile.chk for example below) and make a new input like this: %nprocshared=8 %mem=4GB %chk=newfile.chk # freq td=(singlets,nstates=6,root=1) hf/6-31g(d,p) geom=checkpoint guess=read frequency part of calc 0 1 {make sure to put a few blank lines here!!!} You could also include the "read" option in the TD keyword to save some time but I can't remember exactly how that ends up behaving in these kinds of jobs. Regards, Jamin On Mon, Nov 22, 2010 at 4:42 AM, Ernst Horkel ehorkel++ioc.tuwien.ac.at wrote: > > Sent to CCL by: "Ernst  Horkel" [ehorkel=-=ioc.tuwien.ac.at] > Dear all, > > I intend to simulate electronic spectra of organic compunds using TDDFT level of theory. For this purpose I want to perform a geometry optimization of the 1st excited state (of benzene as an example) and an vibrational analysis of the minimum found. Unfortunately, the job fails and I have no idea why. IMHO the first job (geometry optimization) succeeds, but the second one doesn't. > > I am using G09 on a 8 core Opteron machine (all 8 cores, 4 GB RAM) > > So this is the (partial) output: > > -----SNIP------- > >   D24       -0.00016   0.00000   0.00000   0.00007   0.00007  -0.00009 >         Item               Value     Threshold  Converged? >  Maximum Force            0.000048     0.000450     YES >  RMS     Force            0.000017     0.000300     YES >  Maximum Displacement     0.000107     0.001800     YES >  RMS     Displacement     0.000032     0.001200     YES >  Predicted change in Energy=-1.673574D-08 >  Optimization completed. >    -- Stationary point found. >                           ---------------------------- >                           !   Optimized Parameters   ! >                           ! (Angstroms and Degrees)  ! >  --------------------------                            -------------------------- >  ! Name  Definition              Value          Derivative Info.                ! >  -------------------------------------------------------------------------------- >  ! R1    R(1,2)                  1.4138         -DE/DX =    0.0                 ! > > -----SNAP----- > > so this seems to tell me geometry optimization is OK; > > -----SNIP------- > > ZZZY=              0.0003 XXYY=            -89.8781 XXZZ=            -63.1837 YYZZ=            -63.1836 >  XXYZ=             -0.0003 YYXZ=             -0.0010 ZZXY=              0.0000 >  N-N= 2.014646440196D+02 E-N=-9.396066542997D+02  KE= 2.300858570874D+02 >  1\1\GINC-MASTER\FOpt\RTD-HF-FC\6-31G(d,p)\C6H6\HANMIK\22-Nov-2010\0\\# >  opt freq td=(singlets,nstates=6,root=1) hf/6-31g(d,p) geom=connectivi >  ty\\benzol-test tddft\\0,1\C,6.5168316274,1.2693566194,0.001738384\C,7 >  .9306171845,1.2692925681,0.0021804376\C,8.637562195,2.4936128114,0.001 >  6217232\C,7.930730751,3.7180250171,0.0006395457\C,6.5169664187,3.71808 >  94956,0.0001753444\C,5.8100105637,2.4937526246,0.000783884\H,5.9799138 >  434,0.3394581721,0.0021016335\H,8.467444507,0.3393399532,0.0029641075\ >  H,9.7113385438,2.4935793196,0.0019875469\H,8.4676875942,4.6479015869,0 >  .0002526738\H,5.980104819,4.6480239866,-0.0007186233\H,4.7362355123,2. >  4938321253,0.0004983426\\Version=AM64L-G09RevA.02\State=1-A\HF=-230.70 >  92354\RMSD=2.128e-09\RMSF=3.164e-05\Dipole=0.0000005,0.,-0.0000154\PG= >  C01 [X(C6H6)]\\_._ > > >  The arm of the moral universe is long, but it bends toward justice. >  -- Martin Luther King, Jr. >  Job cpu time:  0 days  0 hours 12 minutes  2.7 seconds. >  File lengths (MBytes):  RWF=     47 Int=      0 D2E=      0 Chk=      5 Scr=      1 >  Normal termination of Gaussian 09 at Mon Nov 22 13:16:17 2010. >  Link1:  Proceeding to internal job step number  2. >  ---------------------------------------------------------------------- >  #N Geom=AllCheck Guess=TCheck SCRF=Check GenChk RTD-HF(FC)/6-31G(d,p) >  Freq >  ---------------------------------------------------------------------- >  QPErr --- A syntax error was detected in the input line. >  Check SCRF=Check GenChk RTD-HF(FC)/6-31G >                               ' >  Last state="GCL" >  TCursr= 1046 LCursr=   54 >  Error termination via Lnk1e in /opt/g09/l1.exe at Mon Nov 22 13:16:17 2010. >  Job cpu time:  0 days  0 hours  0 minutes  0.2 seconds. >  File lengths (MBytes):  RWF=     47 Int=      0 D2E=      0 Chk=      5 Scr=      1 > > ------SNAP------- > > This are the last lines of the log file; This tells me there is a syntax error, which I do not understand... > > Here is the (complete) input: > > > %nprocshared=8 > %mem=4GB > %chk=benzene_tddft.chk > # opt freq td=(singlets,nstates=6,root=1) hf/6-31g(d,p) geom=connectivity > > benzol-test tddft > > 0 1 >  C                  6.52617188    1.28567244    0.00167800 >  C                  7.92133188    1.28567244    0.00167800 >  C                  8.61886988    2.49342344    0.00167800 >  C                  7.92121588    3.70193244    0.00047900 >  C                  6.52639088    3.70185444    0.00000000 >  C                  5.82878988    2.49364844    0.00099600 >  H                  5.97641288    0.33335544    0.00212800 >  H                  8.47083988    0.33315944    0.00299300 >  H                  9.71854988    2.49350344    0.00231200 >  H                  8.47141588    4.65407544    0.00042000 >  H                  5.97626888    4.65413544   -0.00095300 >  H                  4.72918588    2.49383144    0.00081600 > >  1 2 1.5 6 1.5 7 1.0 >  2 3 1.5 8 1.0 >  3 4 1.5 9 1.0 >  4 5 1.5 10 1.0 >  5 6 1.5 11 1.0 >  6 12 1.0 >  7 >  8 >  9 >  10 >  11 >  12 > > Btw, the input was generated using GaussView 5.0; so it should be OK? > > So my questions are: > > 1.) is it possible to perform the job in the way I intend to? > 2.) is the input correct? > 3.) what else can be wrong? > > Any ideas or hints are really welcome. > > Thank you in advance, > > Ernst Horkel, > > ehorkel_._ioc.tuwien.ac.at>      http://www.ccl.net/cgi-bin/ccl/send_ccl_message>      http://www.ccl.net/cgi-bin/ccl/send_ccl_message>      http://www.ccl.net/chemistry/sub_unsub.shtml>      http://www.ccl.net/spammers.txthttp://www.ccl.net/cgi-bin/ccl/send_ccl_messagehttp://www.ccl.net/chemistry/sub_unsub.shtmlhttp://www.ccl.net/spammers.txt