Definition of aromaticity

Hello...





I just realized something weired.


C1=CNNNN1 is considered as aromatic. Considering the fact that all atoms have to have sp2-hybridization and the total system has to have 4n+2 pi-electrons, this should be indeed an aromatic ring. It just looks very weired to me.


What do other people think? Aromatic or not aromatic?





Thanks a lot,





tobias

Hi!





We will run some calculation and tell the results.





Andras

Just two other remarks on that:





1.) C1=CNN1 is not considered aromatic according to chemaxon's definition but aromatic according to daylight's definition.





2.) In benzene you have 6 C atoms and three double bonds (or six aromatic ones). However, you have 6 pz-orbitals containing 1 electron. According to MO-theory those 6 pz's form 3 pi and 3 pi* MO's. In benzene the 3 pi are occupied and the 3 pi* are empty. Therefore you get an energy gain.


In pyrrole the energy gain is less since you get 2 pi, 2 pi* and 1 non-bonding MO and the electrons of the N occupy the nb MO.





So in the case of C1=CNNNN1 you will have a much bigger non-bonding/pi* character in your bonds (with 10 pi-electrons and only 6 pz orbitals). The gain is for sure not very high (and you have to consider the energy you have to invest in the formation of the sp2 hybridised N's).





How does this sound?





tobias

Hi!





I think C1=CNN1 is a rather strained ring that's why I think it is not aromatic.


But to be in conform with the Daylight aromatization you have an option to aromatize the molecule according to the Daylight rules.


See


http://www.chemaxon.com/marvin/doc/api/chemaxon/struc/MoleculeGraph.html#aromatize(int)





On the second point I do agree with you.





Andras

Mmmh...





Ok, I new that I have the option to aromatize according to daylight. But anyway, I think the definition of aromaticity is somehow wrong implemented.


The implemented rules seem to be:


1) 4n+2 pi-electrons


2) only sp2-hybridised atoms (= planar)





O1OOOO1 is also considered as aromatic... I don't know how the implementation gets this idea.





My rules for aromaticity would be:





1) 4n+2 pi-electrons


2) only sp2-hybridised atoms (= planar)


3) floor(k/2) double bonds in a ring with k-atoms





This definition should work... or not?





Anyway, I am pretty sure that non of this compounds (C1=CNNNN1, O1OOOO1, C1=CNN1) would show aromatic behaviour in "real life" like deshielding of the nuclei in NMR.





tobias

I don't understand your third rule.





I have a ring with 'k' atoms, 'n' electrons and 'z' double bonds.


And?

Quote:

I don't understand your third rule. I have a ring with 'k' atoms, 'n' electrons and 'z' double bonds.

I mean the following:


You have a ring with k=6 atoms --> floor(k/2) = 3


So you have to have 3 double bonds in your ring (plus rule 1 and rule 2 fulfilled) that the ring can be considered as aromatic. (True


The same for 5-membered rings:


k=5 --> floor(k/2) = 2


It's just an suggestion... What do you think?





tobias

What about boron compounds?


I think there can be six membered ring which contains Boron that can be aromatic.

Mmmh


I just know it for borazol ([H]1=[N+]([H])([H])=[N+]([H])([H])=[N+]1[H]). It would be aromatic according to all rules, but it is not (at least this is what the literature says). But that's probably a special case.


After all, I don't konw what an appropriate definition of aromatic would be. I just know that the definition at the moment (as implemented) seems at least questionable. Am I completely wrong?


Anyway, I was just curious what other people think of those special cases...





tobias

I'll think about your third rule. If I cannot find a counter-example we can use it.





Thank you for the suggestion

tobias wrote:

Hello... I just realized something weired. C1=CNNNN1 is considered as aromatic. Considering the fact that all atoms have to have sp2-hybridization and the total system has to have 4n+2 pi-electrons, this should be indeed an aromatic ring. It just looks very weired to me. What do other people think? Aromatic or not aromatic? Thanks a lot, tobias

The compound in question is NOT aromatic, please find the attached optimized structure at B3LYP/6-31G(d) level.

Hello...





I spend some thoughts on this whole topic.


Now I wonder if it is possible to use my own definition of aromaticity? Therefore I have some questions:

What kind of flags are set to determine aromatic Atoms/Bonds in the molecule? And are they used to determine an aromatic Atom/Bond when I use a SMARTS in a MolSearch?

At what time are those flags set? The minute I call aromatize() on the molecule. And if so, do they vanish when I call dearomatize()?

I would just like to write a method that sets those flags following my rules and afterwards use MolSearch on that. Is that possible?





Thanks a lot for your help,





Tobias

Hi!





You can definitely write your own aromatization method.


In the method you just have to change the bond type to MolBond.AROMATIC.





You can use the MolSearch API as it doesn't aromatize by default.


The only thing is that you have to take care to call your aromatization method before setQuery and setTarget methods.


See:


http://www.jchem.com/doc/api/chemaxon/sss/search/MolSearch.html

volfi wrote:

You can use the MolSearch API as it doesn't aromatize by default. The only thing is that you have to take care to call your aromatization method before setQuery and setTarget methods. See: http://www.jchem.com/doc/api/chemaxon/sss/search/MolSearch.html

Actually MolSearch currently does not call aromatization at all. (We are planning to add this as an option.) However, JChemSearch, the database search class always does. (For database molecules it is done during standardization at import, and for queries before the search.)





You may find more information about aromatization in the following topic:





http://www.chemaxon.hu/forum/ftopic693.html