Hi all,
I fail to see why a Forward IRC calculation fails when I used the results from a successful opt+freq calculation. The original opt+freq was:
hf/3-21+g* opt=(tight,calcfc) SCFC=(solvent=water) freq=noraman
0 3 (charge 0, spin 3)
My first set of frequencies:
Frequencies -- -637.6381 23.0570 29.2226
But when I try and perform an IRC using:
# irc=(RCFC,tight,forward,maxpoints=150,maxcycle=1028) hf/3-21+g* scrf=(checkpoint) geom=allcheck
I get the following error:
Restarting incremental Fock formation.
Restarting incremental Fock formation.
Restarting incremental Fock formation.
Restarting incremental Fock formation.
Restarting incremental Fock formation.
Restarting incremental Fock formation.
>>>>>>>>>> Convergence criterion not met.
Error on total polarization charges = 0.02200
SCF Done: E(UHF) = -1170.87352619 A.U. after 129 cycles
NFock=128 Conv=0.15D-04 -V/T= 2.0041
= 0.0000 = 0.0000 = 1.0000 = 2.0924 S= 1.0305
= 0.000000000000E+00
Annihilation of the first spin contaminant:
S**2 before annihilation 2.0924, after 2.0024
Convergence failure -- run terminated.
Error termination via Lnk1e in /usr/local/g09_d01/l502.exe at Fri May 1 08:27:28 2015.
Any help would be appreciated.
Thanks
Anthony
Note the following lines:
SCF Done: E(UHF) = -1170.87352619 A.U. after 129 cycles
NFock=128 Conv=0.15D-04 -V/T= 2.0041
You are far from converging the energy at the new geometry. You need more SCF cycles. You may also try SCF=qc to get quadratic convergence. If the solvent is not critical in understanding the mechanism, I would do this in "gas phase" at least initially.
Hi,
Try running IRC with less number of steps, and then optimized the final geometry with either calcall or calcfc.
Hi Dipankar,
I forgot to mention that it fails after 2 steps. So what I am seeing is quite a large vibration; visually it is very indicative of the motion of the reaction, but the moment I try using IRC it fails after a couple of steps. I have tried using calcall during the IRC (in calcfc and RCFC, and iop=(1/7=10) and tight).
Thanks
Hi Vladimir,
I've tried GDIIS and it still results in the same error, failing to converge after only two steps. Seeing the only two steps in GView the nucleophilic addition of nitrogen to the carbon of the ketone only just 'begins' to work. I am pretty certain that I expect a + charge once the NHR group attacking the RHC=O ketone carbon forced a bond by an elimination of the C=O to a C-O(-) This, I think should have a spin multiplicity of 1, but unfortunately I never see an increase in energy during a modredundant bond scan unless I use a spin of 3.
Anthony,
I think that option SCFC=(solvent=water) makes the problem. Removing this option and preparing your input wirh counterpoise=3 (or some other number of groups) would give you an additional insight about bond energies. I suppose you have Gaussview 5, which makes atom grouping easy to prepare. When you, hopefully, get optimized grouped structures with counterpoise, you can with that result try to repeat calculation without counterpoise but with option SCFC=(solvent=water).
Hi Vladimir, I'll have a look into this now and see how I get on with it.
Thanks again.
Note the following lines:
SCF Done: E(UHF) = -1170.87352619 A.U. after 129 cycles
NFock=128 Conv=0.15D-04 -V/T= 2.0041
You are far from converging the energy at the new geometry. You need more SCF cycles. You may also try SCF=qc to get quadratic convergence. If the solvent is not critical in understanding the mechanism, I would do this in "gas phase" at least initially.
Hi B. R. Ramachandran ,
Using SCF=qc made a very big difference. I was able to successfully find minima on either side of the IRC. Thanks again.
Awesome! I have run into a couple of cases where I exhausted all the tricks I know. I was worried this may be one of those cases.
- Badges
- Science topic