I have seen in a lot of pictures that benzyne looks like:


But aren't $\ce{C_6}$ and $\ce{C_5}$ $-$ $\ce{sp}$ hybridized?

And hence bond angle should be $180^{\circ}$ or at least near about it?

Something like :





  • $\begingroup$ Please don’t ask two unrelated questions at once. I feel that the cycloallene question could stand by its own right. (But please ask ‘do they exist’ rather than ‘why do they not exist’. If need be, add that you never heard of them.) $\endgroup$
    – Jan
    Oct 1 '15 at 16:08
  • $\begingroup$ Benzyne is but a short-lived reaction intermediate. I'm not sure its geometry is known with any certainty. $\endgroup$ Oct 1 '15 at 16:22
  • 2
    $\begingroup$ Wikipedia has a good summary of the situation here. Remeber that geometry determines hybridisation, not the other way round - i.e just because a triple bond is depicted doesn't mean there is $sp$ hybridisation. $\endgroup$
    – bon
    Oct 1 '15 at 16:35
  • $\begingroup$ related chemistry.stackexchange.com/questions/215/… $\endgroup$
    – Mithoron
    Oct 1 '15 at 19:55
  • $\begingroup$ To address an implicit question, it is standard practice to draw line structures with simple angles and bond lengths and leave it to the reader to interpret what the real structure might actually look like based on their knowledge and intuition. $\endgroup$
    – Aaron
    Oct 2 '15 at 7:29

Yes, but.

Yes, in the conventional low-level models, one would consider the two carbons in benzyne you mentioned as sp-hybridised. And that does mean that their orbitals seem to be pointing the wrong way.

Crude description of the p orbitals that make benzyne’s triple bond

However, a better picture would be to use an orbital which is a lot closer to sp²-hybridisation, and an even better picture would calculate orbitals computationally (I can’t do that, but Orthocresol can).

The thing is, all these pictures show, that the triple bond in benzyne is rather different from those in linear molecules or large cycloalkynes such as cyclodecyne ($\ce{C10H16}$). It contains a large angular strain, not unlike very small saturated rings such as cyclopropane. The strain causes the triple bond to be rather weak and easily cleaved/attacked — which paves the way for all reactions in which benzyne is an intermediate.

Larger cycloalkynes with eight or nine carbons have been isolated, but for benzyne only in situ studies have been made according to my quick check on SciFinder two seconds ago. In layman’s terms, you can think of the angle strain by saying ‘I am a triple bond, I want a $180°$ angle, but all I can have is $120°$ so I’m rather unhappy!’ A straight-up $180°$ angle is not possible, however, because of the other two atoms in the benzyne ring.

Kukolich, McCarthy and Thaddeus published two papers (one together with Tanjaroon) on microwave-spectrum measurements to determine the structure and compare it to in silico structures. Their results give a $126°$ angle along $\ce{C^1-C^6-C^5}$ (your numbering scheme). This is more obtuse than in benzene ($120°$) but still far away from $180°$. However, the major constraint is the angle $\ce{C^6-C^1-C^2}$, which is compressed to a mere $111°$.[1,2]

Before the objection is raised and as ron pointed out in the comments: The triple bond is indeed a bond and not a diradical as can be derived from its behaviour. E.g, it concertedly and stereospecificly adds to dienes.

[1] S. G. Kukolich, C. Tanjaroon, M. C. McCarthy, P. Thaddeus, J. Chem. Phys. 2003, 119, 4353. DOI: 10.1063/1.1593015
[2] S. G. Kukolich, C. Tanjaroon, M. C. McCarthy, P. Thaddeus, J. Phys. Chem. A 2004, 108, 2645. DOI: 10.1021/jp031344p

  • $\begingroup$ Nice answer. I would just add that although the aryne bond is strained, there is still enough overlap between the two orbitals for it to be considered a bond rather than a biradical. For example, o-benzyne reacts in a concerted, stereospecific manner with dienes. $\endgroup$
    – ron
    Oct 1 '15 at 16:51
  • $\begingroup$ @ron Added a line to implement that. $\endgroup$
    – Jan
    Oct 1 '15 at 16:54
  • 2
    $\begingroup$ If you are interested in the computational bits: chemistry.stackexchange.com/a/114929/16683 $\endgroup$
    – orthocresol
    May 6 '19 at 16:18
  • $\begingroup$ @orthocresol I’m always interested in the computational bits; thanks for linking =3 $\endgroup$
    – Jan
    May 6 '19 at 18:01

Not the answer you're looking for? Browse other questions tagged or ask your own question.