Alex Granovsky
gran@classic.chem.msu.su
take a second look at your system. Take into account that there are
no periodic boundary conditions (as they are not available in Firefly).
You'll see that the molecular structure you have suggested has some
atoms with unsaturated bonds and there are several atoms in unusual
oxidation states.
You need to fix these problems, and start with chemically realistic
structure.
Kind regards,
Alex Granovsky
On Mon Sep 30 '13 12:09pm, David G. wrote
-----------------------------------------
>>Hi
>In addition we have checked the convergence for lower (mult=4) and higher (mult=8) spin polarizations and the convergence is still unsuccessful. any comment ?
>
>
>
>On Thu Sep 26 '13 7:28pm, David G. wrote
>----------------------------------------
>>>Dear Alex
>>sorry for the long delay ! we were testing different things to reply your comment.
>>We have revised the system as you proposed and deleted one Cobalt atom to make the system more realistic from chemistry point of view. Also since the H2O ligands may interact with the transition metals like van der Waals interaction we have used B3LYP1 method. The spin of the system has increased too. Fortunately we are seeing some improvement in the SCF procedure but the problem has been not solved yet. Attached please find the output and basis set files.
>>we really appreciate your comments on this issue. Sincerely Yours, David
>>
>>
>>
>>
>>
>>
>>
>>On Wed Sep 4 '13 11:43pm, Alex Granovsky wrote
>>----------------------------------------------
>>>Dear David,
>>>sorry for asking but... is this structure really the one you
>>>are interested in? Could you please reexamine your Firefly's
>>>output file in ChemCraft of any other visualization-capable
>>>software?
>>>Is the structure really a monomer i.e. not a part of polymer?
>>>It does not seem to be chemically reasonable, if so.
>>>Kind regards,
>>>Alex Granovsky
>>>
>>>
>>>
>>>
>>>
>>>On Wed Sep 4 '13 1:39pm, David G. wrote
>>>---------------------------------------
>>>>>Also we have examined the same structure using Gaussian with the following commands and the scf surprisingly converged just in 66 steps.
>>>>#p b3lyp/gen scf=(maxcycle=250,conver=5) pseudo=read
>>>>O H C 0
>>>>3-21g
>>>>****
>>>>Co 0
>>>>LANL2DZ
>>>>****
>>>>Co 0
>>>>LANL2DZ
>>>>any comment to solve this issue using FF ?
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>On Mon Sep 2 '13 9:57am, David G. wrote
>>>>---------------------------------------
>>>>>>Dear Richard
>>>>>In practice we tried different multiplicities for this system which is neutral. Making it positive is just to get a guess for the wave-function. As you proposed we can passivate the dangling cites but in fact the system is a monomer so we would like to study the electronic structure of the real system at first.
>>>>>
>>>>>
>>>>>
>>>>>On Mon Sep 2 '13 1:49am, Richard wrote
>>>>>--------------------------------------
>>>>>>Anybody's guess what the charge and multiplicity of this should be - assuming you can get say a trans bisbenzoate tetra-water Co (2/3?) model to work, then you need to cap the dangling valencies of your target in some way.
>>>>>>Richard
>>>>>>
>>>>>>
>>>>>>On Sun Sep 1 '13 8:38pm, David G. wrote
>>>>>>---------------------------------------
>>>>>>>>Hi
>>>>>>>We have used 3-21g and LANL2DZ basis sets for our case and we have followed the manual instructions to get convergence even we tried to get convergence for the positively charged case but our struggle with it was unsuccessful. In the following you can see some parts of the input and the complete input and output files
>>>>>>>were attached.
>>>>>>>Best Wishes, D.