From owner-chemistry@ccl.net Mon Jan 30 14:00:00 2012 From: "Jim Kress ccl_nospam[A]kressworks.com" To: CCL Subject: CCL:G: Unusual imaginary frequencies in ORCA Message-Id: <-46209-120130121912-23024-nZvo4YgxuPBLulvEAcQ1hQ,,server.ccl.net> X-Original-From: "Jim Kress" Content-Language: en-us Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset="iso-8859-1" Date: Mon, 30 Jan 2012 12:19:02 -0500 MIME-Version: 1.0 Sent to CCL by: "Jim Kress" [ccl_nospam*_*kressworks.com] Did you try v 2.9? Jim > -----Original Message----- > From: owner-chemistry+ccl_nospam==kressworks.com|a|ccl.net > [mailto:owner-chemistry+ccl_nospam==kressworks.com|a|ccl.net] On Behalf > Of Jason D Acchioli jdacchio%x%gmail.com > Sent: Sunday, January 29, 2012 3:21 PM > To: Kress, Jim > Subject: CCL:G: Unusual imaginary frequencies in ORCA > > > Sent to CCL by: "Jason D'Acchioli" [jdacchio^^^gmail.com] Rudolf, > > Thanks for the suggestion. We're still trying to figure this one out. > Dmytro Bykov from the ORCA team is assisting, since this could also be an > mpi problem. I won't know for sure, since the jobs with %scf convcheckmode > 0 are still queuing on a remote server. My local calculations are the ones I'm > trying to get to work. > > Jason > > On Jan 26, 2012, at 11:21 AM, Rudolf Herrmann > rudolf.herrmann*|*physik.uni-augsburg.de wrote: > > > > > Sent to CCL by: "Rudolf Herrmann" > > [rudolf.herrmann/./physik.uni-augsburg.de] > > Hello Jason, > > > >> > >> I'm having fits over the results of vibrational analyses of some > >> organometallic species in ORCA 2.8. I've been running calcs on > >> eta5-coordinated cyclopentadienyl and indenyl species, and keep > >> coming > > up > >> with 12 imaginary vibrational modes. I've used Grid6 and Grid7, both > > with VeryTightSCF, but I keep getting the same results. It is > > interesting > >> to note that in the ORCA output, I get the message: > >> > >> The first frequency considered to be a vibration is 20 The total > >> number of vibrations considered is 31 > >> > >> Frequency number 20 is the first real frequency in the file. So do I > >> ignore the imaginary vibrational frequencies? > >> > >> As an aside, I don't get the same vibrational results using > >> Gaussian09 > > or > >> Gamess (Guassian calculates frequencies analytically, and Gamess > >> using > > semi-numerical methods). Please help! I'd like to be able to use ORCA > > for > >> these calcs, but I'm not sure why I keep getting imaginary modes! > >> > >> Jason > >> > > > > I recently had a similar problem (due to diffuse basis functions) > > which I reported to the ORCA list. Christoph Riplinger from the ORCA > > team noted that I did not have good SCF convergence (Energy was > > converted but not > > DP-MAX) which created the awkward frequency results. This could also > > be your case. Maybe you can get better convergence by enforcing > > convergence of all criteria with > > > > %scf convcheckmode 0 > > maxiter 1000 > > end > > > > Hope this helps, > > > > R. Herrmann. > > > > -- > > Dr. Rudolf Herrmann > > Institut für Physik > > Universität Augsburg > > > >> PS The input file is: > >> > >> #Geometry optimization of eta-5 cp Fe complex; ! PAL8 RKS B3LYP/G > >> ECP{SDD=Fe, SD(10,MDF)} def2-SVP VeryTightSCF Grid6 > > moread NoFinalGrid > >> > >> %moinp "eta5_cp_fe_cat.inp_374.rat1.uwsp.gbw" > >> > >> !NumFreq > >> > >> %freq CentralDiff true end > >> > >> *xyz 1 1 > >> C 0.971689 0.719105 1.589637 > >> C 0.977762 -0.702168 1.606407 > >> C -0.400171 1.150321 1.594023 > >> C -1.221910 -0.005615 1.599239 > >> C -0.374426 -1.161652 1.578488 > >> H -0.749285 2.182860 1.606905 > >> H -2.312277 -0.012064 1.604646 > >> H -0.701906 -2.200605 1.595688 > >> C -1.475017 -0.471787 -1.147495 > >> C 0.341691 1.519012 -1.134741 > >> C 1.139269 -1.047892 -1.139251 > >> O -2.398822 -0.777188 -1.727932 > >> O 0.566217 2.472894 -1.703417 > >> O 1.854371 -1.711141 -1.715752 > >> H 1.866252 -1.334418 1.615166 > >> H 1.849699 1.364293 1.609103 > >> Fe -0.002414 0.002238 -0.169933 > >> * > >> > >> > >> ******************************************* > >> Jason D'Acchioli > >> Assistant Professor of Chemistry > >> University of Wisconsin-Stevens Point > >> 2001 Fourth Avenue > >> Stevens Point, WI 54481 > >> http://chemdac.uwsp.edu> > > > > ******************************************* > Jason D'Acchioli > Assistant Professor of Chemistry > University of Wisconsin-Stevens Point > 2001 Fourth Avenue > Stevens Point, WI 54481 > http://chemdac.uwsp.edu> From owner-chemistry@ccl.net Mon Jan 30 17:03:00 2012 From: "Billy Wong wongzhi(-)myvuw.ac.nz" To: CCL Subject: CCL:G: Gaussian Error l103.exe: Error in internal coordinate system Message-Id: <-46210-120130163404-18145-s7+WoDWt1ZmoLwJORpTQSw+/-server.ccl.net> X-Original-From: "Billy Wong" Date: Mon, 30 Jan 2012 16:34:02 -0500 Sent to CCL by: "Billy Wong" [wongzhi||myvuw.ac.nz] Hi, I'm using g09 to optimize my structure to TS. My input keywords are as follows: #p freq=raman opt=(cartesian,qst3,maxstep=300,maxcycle=300) SCF=(xqc,maxcycle=512) nosymm bp86/genecp All of my calculations ended with the same error which is: Eigenvalue 90 is 2.36D-07 should be less than 0.000000 Eigenvector: R9 A22 A20 D15 D14 1 0.36553 -0.36274 0.25441 0.20931 0.16765 R6 A11 R13 A16 D17 1 0.15094 -0.14743 -0.14323 0.13206 0.13147 Eigenvalue 91 is 2.03D-07 should be less than 0.000000 Eigenvector: A44 D32 R6 D33 A41 1 0.24371 0.23806 -0.23420 0.21638 0.20473 D15 A5 D29 D30 A57 1 0.18902 -0.17993 -0.17854 -0.16646 -0.16610 NTrRot= -1 NTRed= 172 NAtoms= 29 NSkip= 91 IsLin=F Error in internal coordinate system. Error termination via Lnk1e in /home/scifachpc01/groups/scps/Gaussian09//g09/l103.exe at Mon Jan 30 20:29:45 2012. Job cpu time: 1 days 17 hours 36 minutes 38.8 seconds. File lengths (MBytes): RWF= 613 Int= 0 D2E= 0 Chk= 54 Scr= 1 I've tried using a smaller basis set and using Opt=Z-matrix too. But none of this works. My previous optimization calculations which ended with this error can be easily solved by adding the Cartesian keyword into Opt. However, the error persists in TS calculation. Can anyone help me with this please? Thanks in advance. Cheers, Billy Email: wongzhi . myvuw.ac.nz