From owner-chemistry@ccl.net Thu Jul 6 18:03:00 2017 From: "Sara Elgengehi smahmoud:_:sci.cu.edu.eg" To: CCL Subject: CCL: GAMESS Convergence Problem Message-Id: <-52876-170706174458-400-ok1QO1jr/7O2lUKQkLfXEA^^server.ccl.net> X-Original-From: Sara Elgengehi Content-Type: multipart/alternative; boundary="001a1144f218c0a1790553ad067b" Date: Thu, 6 Jul 2017 23:44:50 +0200 MIME-Version: 1.0 Sent to CCL by: Sara Elgengehi [smahmoud(_)sci.cu.edu.eg] --001a1144f218c0a1790553ad067b Content-Type: text/plain; charset="UTF-8" Dear Brad, Thank you very much for your response. I actually used ICUT=11 and still there is a problem in the convergence. Here is the warning message: THE INPUT BASIS SET CONTAINS APPROXIMATE LINEAR DEPENDENCE. THE SMALLEST EIGENVALUE OF THE OVERLAP MATRIX IS 5.173593E-09 THERE ARE 18 EIGENVALUES LESS THAN QMTTOL= 1.00E-06 EIGENVECTORS BELOW -QMTTOL- ARE DROPPED FROM THE MO SPACE, IN ORDER TO ELIMINATE THE APPROXIMATE LINEAR DEPENDENCE. THIS WILL MOST LIKELY CHANGE THE ENERGY ON THE ORDER OF TENS OF MICROHARTREES, DEPENDING ON THE NUMBER OF ORBITALS REMOVED. THE VALUE OF QMTTOL CAN BE CHANGED IN $CONTRL: JOBS WITH EIGENVALUES BELOW 1.0E-07 PROBABLY WON'T CONVERGE. EIGENVALUES BETWEEN 1.0E-07 AND 1.0E-06 MAY REQUIRE LOOSENING OF THE DENSITY CRITERION (CONV IN $SCF), MORE ACCURATE INTEGRAL EVALUATION (ICUT=11 IN $CONTRL), MORE ACCURATE DIRECT SCF FOCK MATRIX CONSTRUCTION (FDIFF=.FALSE. IN $SCF), OR CHANGING THE SCF CONVERGER (DIIS=.TRUE. IN $SCF). I think that you meant QMTTOL, but how to loosen it without affecting the accuracy? I will also try what you suggested about the diffuse functions. Regards, Sara On Thu, Jul 6, 2017 at 3:31 PM, Brad Rose brose-x-live.com < owner-chemistry**ccl.net> wrote: > > Sent to CCL by: "Brad Rose" [brose{}live.com] > Dear Sara, > > SCF convergence will be a problem with all the diffuse functions. It > takes time and practice to learn how to tell what could be the cause, > something worth your time if you plan on doing this work. The first thing > I would consider is the integral cutoff values, these are ICUT and one > other parameter that I don't recall offhand, I believe there should be a > warning in your output about the integral cutoffs if you are running into > linear dependancies (you likely are with all the diffuse functions > present). To get a reasonable geometry close to the local minima I would > suggest removing the diffuse functions from all but the anion and add > dispersion correction (the dispersion correction is computationally > cheap). > > Regards, > Brad Rose > > > > Sent to CCL by: "Sara Mahmoud" [smahmoud%x%sci.cu.edu.eg] > Sent to CCL by Sara[smahmoud|*|sci.cu.edu.eg] > Dear all, > > I'm trying to study the adsorption of some ions on graphene flakes using > GAMESS-US program. When I optimize the ion-graphene system I get this > error message after several points of optimization. > > ***** FAILURE TO LOCATE STATIONARY POINT, SCF HAS NOT CONVERGED > UPDATED HESSIAN, GEOMETRY, AND VECTORS WILL BE PUNCHED FOR RESTART > **** THE GEOMETRY SEARCH IS NOT CONVERGED! **** > > I am using the DFT functional B3LYP with 6-31+G(d) basis set as a > starting level of calculation, I'm going to justify both the functional > and basis set and then include the suitable dispersion corrections in > single point energy calculations. This is some of the input I used: > > $CONTRL MAXIT=200 ISPHER=0 > $SCF DIRSCF=.T. SOSCF=.F. DIIS=.T. SHIFT=.T. DAMP=.t. > $STATPT OPTTOL=0.00045 NSTEP=300 > > I'm a new student in the field and have no experience in GAMESS-US > program. What I need to know is how can I get my jobs converged? and what > are the keywords that should I use to improve the SCF convergence for > this kind of interaction (ion-pi-system interaction)? > > Any help concerning this point will be highly appreciated. > Thanks in advance for all of you. > Sara> > > -- *Disclaimer:* This E-mail represents the views of its author only and does not necessarily reflect the opinions of Faculty of Science (*SCI *). We do not assume responsibility for any loss, damage (including direct, indirect, punitive, special, or consequential loss or damage), which you may directly or indirectly suffer from this email or its contents or through the ways of handling this email which includes sending, receiving, replying, forwarding, deleting, saving or changing and adjusting its content in any way. --001a1144f218c0a1790553ad067b Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Dear Brad,

Thank you very much for your= response.=C2=A0

I actually used ICUT=3D11 and sti= ll there is a problem in the convergence. Here is the warning message:

THE INPUT BASIS SET CONTAINS APPROXIMATE LINEAR D= EPENDENCE.
=C2=A0THE SMALLEST EIGENVALUE OF THE OVERLAP MATRIX IS= =C2=A0 5.173593E-09
=C2=A0THERE ARE =C2=A0 18 EIGENVALUES LESS T= HAN QMTTOL=3D =C2=A01.00E-06
=C2=A0EIGENVECTORS BELOW -QMTTOL- AR= E DROPPED FROM THE MO SPACE,
=C2=A0IN ORDER TO ELIMINATE THE APPR= OXIMATE LINEAR DEPENDENCE.

=C2=A0THIS WILL MOST LI= KELY CHANGE THE ENERGY ON THE ORDER OF TENS OF MICROHARTREES,
=C2= =A0DEPENDING ON THE NUMBER OF ORBITALS REMOVED.=C2=A0 THE VALUE OF QMTTOL C= AN BE=C2=A0
=C2=A0CHANGED IN $CONTRL: JOBS WITH EIGENVALUES BELOW= 1.0E-07 PROBABLY WON'T
=C2=A0CONVERGE.=C2=A0 EIGENVALUES BET= WEEN 1.0E-07 AND 1.0E-06 MAY REQUIRE LOOSENING OF
=C2=A0THE DENSI= TY CRITERION (CONV IN $SCF), MORE ACCURATE INTEGRAL EVALUATION
= =C2=A0(ICUT=3D11 IN $CONTRL), MORE ACCURATE DIRECT SCF FOCK MATRIX
=C2=A0CONSTRUCTION (FDIFF=3D.FALSE. IN $SCF), OR CHANGING THE SCF CONVERG= ER
=C2=A0(DIIS=3D.TRUE. IN $SCF).

= I think that you meant QMTTOL, but how to loosen it without affecting the a= ccuracy?

I will also try what you suggested about = the diffuse functions.=C2=A0

Regards,
Sa= ra


On Thu, Jul 6, 2017 at 3:31 PM, Brad Rose brose-x-live.com <owner-chemistry**ccl.net>= wrote:

Sent to CCL by: "Brad=C2=A0 Rose" [brose{}live.com]
Dear Sara,

SCF convergence will be a problem with all the diffuse functions. It
takes time and practice to learn how to tell what could be the cause,
something worth your time if you plan on doing this work. The first thing I would consider is the integral cutoff values, these are ICUT and one
other parameter that I don't recall offhand, I believe there should be = a
warning in your output about the integral cutoffs if you are running into linear dependancies (you likely are with all the diffuse functions
present). To get a reasonable geometry close to the local minima I would suggest removing the diffuse functions from all but the anion and add
dispersion correction (the dispersion correction is computationally
cheap).

Regards,
Brad Rose



Sent to CCL by: "Sara=C2=A0 Mahmoud" [smahmoud%x%sci.cu.edu.eg] Sent to CCL by Sara[smahmoud|*|sci.cu.edu.eg]
Dear all,

I'm trying to study the adsorption of some ions on graphene flakes usin= g
GAMESS-US program. When I optimize the ion-graphene system I get this
error message after several points of optimization.

***** FAILURE TO LOCATE STATIONARY POINT, SCF HAS NOT CONVERGED
=C2=A0 =C2=A0 =C2=A0UPDATED HESSIAN, GEOMETRY, AND VECTORS WILL BE PUNCHED = FOR RESTART
=C2=A0**** THE GEOMETRY SEARCH IS NOT CONVERGED! ****

I am using the DFT functional B3LYP with 6-31+G(d) basis set as a
starting level of calculation, I'm going to justify both the functional=
and basis set and then include the suitable dispersion corrections in
single point energy calculations. This is some of the input I used:

$CONTRL MAXIT=3D200 ISPHER=3D0
$SCF DIRSCF=3D.T. SOSCF=3D.F. DIIS=3D.T. SHIFT=3D.T. DAMP=3D.t.
$STATPT OPTTOL=3D0.00045 NSTEP=3D300

I'm a new student in the field and have no experience in GAMESS-US
program. What I need to know is how can I get my jobs converged? and what are the keywords that should I use to improve the SCF convergence for
this kind of interaction (ion-pi-system interaction)?

Any help concerning this point will be highly appreciated.
Thanks in advance for all of you.
Sara



-=3D This is automatically added to each message by the mailing script =3D-=
E-mail to subscribers: CHEMISTRY**ccl.n= et or use:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/= ccl/send_ccl_message

E-mail to administrators: CHEM= ISTRY-REQUEST**ccl.net or use
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/cgi-bin/= ccl/send_ccl_message

Subscribe/Unsubscribe:
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/chemistry/s= ub_unsub.shtml

Before posting, check wait time at: http://www.ccl.net

Job: http://www.ccl.net/jobs
Conferences: http://server.ccl.net/ch= emistry/announcements/conferences/

Search Messages: http://www.ccl.net/chemistry/searchccl/index.shtml
=C2=A0 =C2=A0 =C2=A0 http://www.ccl.net/spammers.txt

RTFI: http://www.ccl.net/chemistry/aboutcc= l/instructions/




D= isclaimer:

This E-mail represents the views of its author only a= nd does not necessarily reflect the opinions of Faculty of Science (SCI)= . We do not assume responsibility for any loss, damage (including direct, indirect, punitive, special, or consequential loss or damage), which you may directly or indire= ctly suffer from this email or its contents or through the ways of handling this email which includes sending, receiving, replying, forwarding, deleting, sa= ving or changing and adjusting its content in any way.

=C2= =A0

--001a1144f218c0a1790553ad067b--