Running ONIOM


Clicked A Few Times
Hi,
I use the exact same example given in the documentation to try the ONIOM calculation, but it gave me an error saying :

oniom: model does not have correct symmetry        0
------------------------------------------------------------------------
There is an error related to the specified geometry
------------------------------------------------------------------------


This is the input file:

start
  
basis spherical
H library 6-311++G**; C library 6-311++G**
end

title "ONIOM Me-CH2"

geometry autosym
H -0.23429328 1.32498565 0.92634814
H -0.23429328 1.32498565 -0.92634814
C -0.13064265 0.77330370 0.00000000
H -1.01618703 -1.19260361 0.00000000
H 0.49856072 -1.08196901 -0.88665533
H 0.49856072 -1.08196901 0.88665533
C -0.02434414 -0.71063687 0.00000000
end

scf; uhf; doublet; thresh 1e-6; end
mp2; freeze atomic; end

oniom
high mp2
low scf
model 3 3 7 0.724
end

task oniom

title "ONIOM Me-Me"

geometry # Note cannot use full D3D symmetry here
H -0.72023641 0.72023641 -1.16373235
H 0.98386124 0.26362482 -1.16373235
H -0.26362482 -0.98386124 -1.16373235
C 0.00000000 0.00000000 -0.76537515
H 0.72023641 -0.72023641 1.16373235
H -0.98386124 -0.26362482 1.16373235
H 0.26362482 0.98386124 1.16373235
C 0.00000000 0.00000000 0.76537515
end

scf; rhf; singlet; end

oniom
high mp2
low scf
model 4 4 8 0.724
end

task oniom


Has anyone encountered the same problem ?
Thanks in advance for your help.
Regards,

Forum Regular
Hi Mhiranya,

Thank you for spotting this problem. Actually it turns out that the example was out-of-date. Judging by the presence of the "autosym" keyword on the first geometry this example must date back to a time when automatic symmetry detection was not the default behavior of the code. Hence in those days no symmetry would be used with the second geometry.

This behavior has been changed years ago. Now the code always tries to automatically detect symmetry, unless instructed otherwise. This leads it to find the incorrect D3D symmetry for the second geometry. So, now we need to explicitly intervene and lower the symmetry to get the correct behavior. Specifying the second geometry as

   H   -0.72023641     0.72023641    -1.16373235
H 0.98386124 0.26362482 -1.16373235
H -0.26362482 -0.98386124 -1.16373235
C 0.00000000 0.00000000 -0.76537515
H 0.72023641 -0.72023641 1.16373235
H -0.98386124 -0.26362482 1.16373235
H 0.26362482 0.98386124 1.16373235
C1 0.00000000 0.00000000 0.76537515

does just that. With this change the input will correctly.

I have updated the manual accordingly.

Best wishes, Huub

Clicked A Few Times
Hi Huub,

So, this means you have changed the label or so ...

C1 0.00000000 0.00000000 0.76537515

Thanks


Forum >> NWChem's corner >> QMMM