VMD-L Mailing List

From: Mcguire, Kelly (klmcguire_at_UCSD.EDU)
Date: Sat Aug 31 2024 - 17:54:33 CDT

For those of you using FFTK in VMD, during the Calc. Bonded steps you might get an error during the second job:

ERROR !!! The optimization did not converge but reached the maximum number of optimization cycles. As a subsequent Frequencies calculation has been requested ORCA will abort at this point of the run. Please restart the calculation with the lowest energy geometry and/or a larger maxiter for the geometry optimization.

This happens before any optimization starts in the second job. Changing MaxIter does not solve the issue. It's the way the file and jobs are written from FFTK. Instead of using the $new_job method, change your ORCA input script to (example, FES is my resname, it will be different for your molecule):

%maxcore 10000

%pal
   nprocs 24
end

*xyz 0 1
 Fe -0.1532990038394928 -0.14298400282859802 0.47529301047325134
  S 1.481408953666687 1.37191903591156 0.05584799870848656
  S -1.5341229438781738 0.34771600365638733 -1.2598680257797241
 Fe 0.2724350094795227 1.5886199474334717 -1.853674054145813
  S 0.8466820120811462 -1.9920909404754639 -0.38588398694992065
 Fe 2.2999210357666016 -0.40101000666618347 -1.0997960567474365
 Fe -0.2597309947013855 -1.263314962387085 -2.2285239696502686
  S 1.3637059926986694 0.05114499852061272 -3.1183950901031494
*
%Compound

New_Step

! B3LYP def2-SVP TightSCF opt NumFreq
%base"FES_hess_initial"

STEP_END

New_Step

! B3LYP def2-SVP TightSCF opt NumFreq

%base"FES_hess_job2"

%geom
  MaxIter 1000
  inhess read
  inhessname "FES_hess_initial.hess"
  PRINTINTERNALHESS true
end

STEP_END

end

The main difference now is the %Compound method. This works for me using ORCA 5.0.3 and VMD 1.9.4a57. Also, when you get a FES_hess_job2_internal.hess as an output, change the first underscore to a period (i.e. FES.hess_job2_internal.hess).

Kelly McGuire PhD
Postdoc - Herzik Lab
University of California - San Diego