CCL:G: Error in termination for a transition state of large organic molecule

[Marcel Swart <marcel.swart[-]icrea.cat>]

Dear Mallika,

why the heck do you specify a maximum of 4092 SCF cycles???

It should be done within 50 max, otherwise you have to try some

other SCF scheme such as EDIIS.

What happens if you take the last geometry, and put that in the

input? Does it stop at the same point, or does it continue for some

steps?

This will help you understand whether this is technical or scientific.

I.e., if it stops and does not go on for not even ONE step, then it

means that there’s something with THAT geometry; if it continues,

then probably you’re running into some memory or disc problem.

Marcel

On 2015-08-15, at 18:36, Mallika Khare mkhare__mtu.edu <owner-chemistry*o*ccl.net> wrote:

Hello Igors

As per your guidance, I tried increasing the optimization cycles,used CalcFC and also used SCF=XQC. But, everytime I am getting the same result (Error in Termination). The output is ending after the rotational constant calculation.

I would be very thankful to you, if you could look at my input output(attached below)and guide me what can be done.

Many Thanks and Regards

Mallika Khare

Here's my input and output:

Input:

%nprocshared=12

%mem=30GB

%OldChk=file_name.chk

%Chk=file_name.chk

#p opt=(qst3,noeigentest,MaxCycle=2048,CalcFC) freq oniom(b3lyp/3-21+g*:b3lyp/sto-3g*:pm3) scf=(xqc,IntRep,maxcycle=4092)

geom=connectivity

Reacrtant

-3 2 0 2 0 2 0 2 0 2 0 2

Output

 90          1           0        5.940783    1.161302   -1.505445

     91          1           0        4.214536    1.555958    0.749945

     92          8           0       -7.269357   -0.588603   -0.813879

     93          1           0       -7.352913   -1.345087   -0.172271

 ---------------------------------------------------------------------

 Rotational constants (GHZ):      0.0701555      0.0201527      0.0192135

 Leave Link  202 at Sat Aug 15 06:39:14 2015, MaxMem=  4026531840 cpu:         0.1

 (Enter /share/apps/gaussian/D01/g09/l9999.exe)

 A people that values its privileges above its principles soon loses both.

 -- Dwight D. Eisenhower

 Error termination request processed by link 9999.

 Error termination via Lnk1e in /share/apps/gaussian/D01/g09/l9999.exe at Sat Aug 15 06:39:14 2015.

 Job cpu time:       1 days 13 hours 27 minutes 31.6 seconds.

 File lengths (MBytes):  RWF=   4865 Int=      0 D2E=      0 Chk=     70 Scr=      2

Error: segmentation violation

   rax 0000000000000000, rbx ffffffffffffffff, rcx ffffffffffffffff

   rdx 000000000001fa18, rsp 00007fffffffa118, rbp 00007fffffffa690

   rsi 000000000000000b, rdi 000000000001fa18, r8  00002aaaaaaaf620

   r9  0000000000000000, r10 00007fffffff9ea0, r11 0000000000000206

   r12 0000000000000000, r13 0000000000000000, r14 00007fffffffa6d8

   r15 00000000000003e6

On Fri, Aug 14, 2015 at 1:16 PM, Mallika Khare mkhare.:.mtu.edu <owner-chemistry!=!ccl.net> wrote:

Hello Igors Mihailovs

Thank you very much for your quick response.

My output didn't terminate at all as per the input I provided. The geometry optimization does not converge as per the output I received. I didn't attach the output because it is really long but, I can show the last RMS Force and Displacement values. They are as follows:

Item                                 Value                        Threshold                          Converged?

 Maximum Force            0.030490                   0.000450                                           NO 

 RMS     Force                0.004366                      0.000300                                               NO 

 Maximum Displacement     0.372605             0.001800                                      NO 

 RMS     Displacement     0.068999                  0.001200                                         NO 

 Predicted change in Energy=-6.415468D-03

 

I am really thankful to you for the information you provided and will try to implement these techniques you told me in your previous email.

Many Thanks and Regards

Mallika Khare

On Thu, Aug 13, 2015 at 2:39 PM, Igors Mihailovs igors.mihailovs0-$-gmail.com <owner-chemistry:_:ccl.net> wrote:

Dear Mallika,

At which exactly part of calculation does it terminate? From Your output (which is just some last lines in file) nobody is able to determine whether this is SCF convergence failure or Geometry optimization failure, or something else...

If it is geometry optimization that does not converge, You can probably increase MaxCycle of it, too, not just for SCF: Opt=(QST3,NoEigenTest,MaxCyc=2048). Also (and better), Opt=CalcFC might have helped – as You are requesting frequency calculation after the Opt, I assume You have resources for it. To use it, determine the step where Your calculation was most close to convergence, e. g. by using this command (stolen from "Exploring Chemistry with Electronic Structure Methods" by Foresman and Frisch):

egrep 'out of|SCF Don|Converged| NO | YES | exceeded' your-output-file | grep -v '\\\\'

and choosing the step with the least values; then adding to Your next input file keyword Geom=(Check,Step=N) where N is number of the best step. And Opt=(QST3,NoEigenTest,MaxCyc=2048,CalcFC).

It is also quite convenient to use XQC or YQC instead of QC nowadays: XQC is using quadratic algorithm only if regular SCF does not converge, whereas YQC is doing steepest descent following by regular SCF and only if the last fails does QC (however, YQC does not work in all cases, see, e. g. http://ccl.net/cgi-bin/ccl/message-new?2015+05+30+001). Sometimes SCF=Fermi helps, too. See http://www.ccl.net/cca/documents/dyoung/topics-orig/converge.html for further options.

With best wishes,

Igors Mihailovs (engineer)

Institute of Solid State Physics

University of Latvia

2015-08-13 19:19 GMT+03:00 Mallika Khare mkhare__mtu.edu <owner-chemistry]^[ccl.net>:

Sent to CCL by: "Mallika  Khare" [mkhare- -mtu.edu]
Hello
I am struggling hard to get my job terminated. But, at this point I
couldn't figure out the reason for this error stated below. I checked by
increasing the maxcycle of SCF,increasing the memory and also checked my
checkpoint file but even then it is not optimized.(Using Geom=AllCheck
also makes no difference)
I would be very thankful if anyone can help me with this problem.
Thanks and Regards
Malllika Khare
mkhare[A]mtu.edu

Input
%nprocshared=12
%mem=30GB
%OldChk=filename.chk
%Chk=filename.chk
#p opt=(qst3,noeigentest) freq oniom(b3lyp/3-21+g*:b3lyp/sto-3g*:pm3)
scf=(qc,tight,maxcycle=6500)
geom=connectivity

Output
Error termination request processed by link 9999.
 Error termination via Lnk1e in /share/apps/gaussian/D01/g09/l9999.exe at
Wed Aug 12 09:49:37 2015.
 Job cpu time:       3 days  5 hours 47 minutes  0.4 seconds.
 File lengths (MBytes):  RWF=    150 Int=      0 D2E=      0 Chk=     70
Scr=      2
Error: segmentation violation
   rax 0000000000000000, rbx ffffffffffffffff, rcx ffffffffffffffff
   rdx 000000000002a8ac, rsp 00007fffffffa168, rbp 00007fffffffa6e0
   rsi 000000000000000b, rdi 000000000002a8ac, r8  00002aaaaaaaf620
   r9  0000000000000000, r10 00007fffffff9ef0, r11 0000000000000202
   r12 0000000000000000, r13 0000000000000000, r14 00007fffffffa728
   r15 00000000000003e6
  --- traceback not available



-= This is automatically added to each message by the mailing script =-

E-mail to subscribers: CHEMISTRY]^[ccl.net or use:
      http://www.ccl.net/cgi-bin/ccl/send_ccl_message

E-mail to administrators: CHEMISTRY-REQUEST]^[ccl.net or use
      http://www.ccl.net/cgi-bin/ccl/send_ccl_message

Subscribe/Unsubscribe:
      http://www.ccl.net/chemistry/sub_unsub.shtml

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

Job: http://www.ccl.net/jobs
Conferences: http://server.ccl.net/chemistry/announcements/conferences/

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

RTFI: http://www.ccl.net/chemistry/aboutccl/instructions/

_______________________________________________________
Prof. Marcel Swart

ICREA Research Professor at
Inst. Comput. Chem. Catal. (IQCC)
Univ. Girona (Spain)

Member of Young Academy of Europe
www.yacadeuro.org
Chair COST Action CM1305 (ECOSTBio)
www.ecostbio.eu
Organizer Girona Seminar 2016
www.gironaseminar.com

web
http://www.marcelswart.eu
vCard
addressbook://www.marcelswart.eu/MSwart.vcf