I build from text O-N+O (the SMILES of this is O=[N+][O-]), I delete the hydrogen atom added automatically on N, then I start MM2 Minimization. Both the bond length and the angle differ from what I see in the top right image here: https://en.wikipedia.org/wiki/Nitrogen_dioxide
$\begingroup$
$\endgroup$
3
-
1$\begingroup$ If you are referring to PerkinElmer's product, I'm afraid we cannot help you since it's a proprietary software and the implementation for the structure optimization algorithms cannot be validated. And even if you contact them, I'd recommend to provide more details as currently it is not clear what exactly the deviation is (at least a screenshot could help). Other than that, this question seems to be off-topic on Chemistry.SE. $\endgroup$– andselisk ♦Commented Apr 16, 2020 at 11:55
-
$\begingroup$ @andselisk why exactly is this off-topic? $\endgroup$– user1271772Commented Apr 18, 2020 at 21:43
-
$\begingroup$ There are too many reasons, why this happens; your question is not specific enough. The general area is of course on-topic for chemistry.se. However, without knowing the set-up of your calculation, it is impossible to really help you. E. g.: Did you adjust spin and charge to reflect the actual molecule? What is the deviation? Chem3d uses - like stated in the question - the MM2 force field; I think it was designed for neutral hydrocarbons predominantly, no surprise it fails for a very complicated molecule like $\ce{NO2}$. $\endgroup$– Martin - マーチン ♦Commented Apr 27, 2020 at 9:59
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
2
By the looks of it, Chem3D uses generalised classical force fields to describe the molecular interactions.
Such force fields are optimised to provide a reasonable fit to multiple properties for lots of different molecules. For example, they reproduce reasonable bond lengths not only for $\ce{NO2}$ but also for the various bonds in organic molecules.
They are also constructed to reasonably reproduce solvation properties which depend on the molecule-specific partial charges, which in turn alter bond lengths and angles.
So it is unlikely that you have done anything wrong.
If you require a more accurate model, either apply more accurate force fields (e.g. those fitted specifically to $\ce{}$) or switch to quantum chemistry methods such as DFT.
-
1$\begingroup$ Thanks for taking the time to explain this! $\endgroup$ Commented Apr 16, 2020 at 16:12
-
$\begingroup$ Yes, indeed, the option explicitly states MM2. This is designed for hydrocarbons and small organic molecules. NO2 is a very complex molecule, it is no surprise that a force field doesn't describe it well. So it is unlikely that you have done anything wrong. There are still plenty of settings that could be wrong though. And I believe this is the case. It makes a huge difference if the program choses a neutral molecule with a doublet, or a singlet cation, or anion. The OP is not specific about that. While I generally agree with your post, this sentence caused me to down-vote. $\endgroup$– Martin - マーチン ♦Commented Apr 27, 2020 at 10:26