Unlike its neighbours on the right-hand side of the periodic table, boron barely forms an anion. A new trick has been established that allows it to do so, enabling a highly unusual complex to be prepared.
The history of boron and its compounds is long and distinguished, with the organoboron compounds — in which organic groups are attached to a boron atom — arguably making the biggest splash. These compounds have many crucial roles as reagents in organic synthesis, especially in carbon–carbon bond-forming reactions1. One unusual family of organoboron compounds is the boroles, which contain a relatively unstable ring of atoms (Fig. 1a). These compounds are candidates for a new class of material for organic light-emitting diodes, but synthetic methods for making boroles are rather limited2,3. Reporting in Angewandte Chemie, Braunschweig et al.4 now describe their preparation of a borole anion. The compound takes part in a surprising chemical reaction, and might open up routes to the preparation of other borole compounds.
Boron belongs to the second row of the periodic table, but its anion chemistry is different from that of most of the other p-block elements in that row (fluorine, oxygen, nitrogen and carbon). These elements form anions in alkali–metal salts, such as F− in lithium fluoride, OH− in lithium hydroxide, NH2− in lithium amide and CH3− in methyllithium. For many years, however, there were no direct observations of alkali–metal salts that included the analogous boron anion, R2B− (known as a boryl anion; R can be either hydrogen or an organic group).
This aberrant behaviour is a consequence of boron's inability to fulfil the 'octet' rule of main-group elements (the elements in the periodic table that aren't transition metals). The octet rule states that main-group atoms that have eight electrons in their outermost (valence) shell are particularly stable. These elements therefore tend to form compounds or anions that have eight valence electrons — fluoride, hydroxide, amide and methyl anions are good examples.
But the boron atom of a boryl anion has only six valence electrons. Boryl anions can get around this problem by accepting two electrons from neighbouring atoms, or by forming a complex with another compound. Several such complexes have been reported5, although the evidence for their existence was indirect. The first boryl anion to be isolated and characterized was described four years ago8,9. In that case, the anionic boron atom was stabilized by electron donation from its neighbouring nitrogen atoms.
Braunschweig and colleagues' boryl anion4 is a different kind of beast, which they describe as a π-boryl anion. Remarkably, it contains a borole ring (Fig. 1b). To understand why boroles can be stable as anions, but not as uncharged molecules, we need to consider another rule of chemistry: Hückel's rule of aromaticity. This defines a formula for the number of π-electrons that a planar, cyclic molecule must have to be aromatic, where π-electrons are those electrons that form π-bonds (such as the double bonds of unsaturated hydrocarbons). The formula is 4n + 2, where n can be zero or a positive integer. Aromatic molecules can therefore have two π-electrons, or six, or ten, and so on. Delocalization of the π-electrons in aromatic molecules boosts the thermodynamic stability of these compounds, a phenomenon known as the aromatic stabilization effect. Conversely, cyclic planar molecules that have 4n π-electrons (four, eight, twelve, and so on) are less stable, and are described as anti-aromatic.
Boroles possess four π-electrons, one from each carbon atom in the ring, and so can be thought of as anti-aromatic. But they also have an empty electron orbital on the boron atom. If that orbital can acquire two extra electrons, then the resulting borole dianion will have six π-electrons, thus becoming aromatic. Such borole dianions are known, and have often been incorporated into complexes with metal atoms10,11,12.
But Braunschweig and colleagues' borole4 flouts convention, because it is a monoanion — a new class of boryl anion. The monoanion has only five π-electrons, and so isn't as stable as the aromatic borole dianions. It makes up for this lack of stability by forming a complex with a compound known as an N-heterocyclic carbene (Fig. 1b). Such carbenes are known to donate two electrons to other boron-containing molecules13, but this is the first time that one of these compounds has been used to stabilize a borole species.
The structure of the authors' borole4 can be drawn in several different ways (known as resonance structures; Fig. 1c), which can be used to help to explain electron distribution in the molecule. One of these structures contains a borataalkene group — a negatively charged carbon–boron double bond. These groups usually react with electrophiles (molecules that contain areas of positive charge) so that the electrophile becomes attached to the carbon atom of the group. But Braunschweig et al. found that their compound reacts with an electrophile only at the boron atom. The borole monoanion is therefore the first example of a borataalkene whose typical polarity has been reversed.
Another resonance structure of the borole monoanion might help to explain the reversed reactivity of the borataalkene. In this structure, the borole ring harbours two negative charges (Fig. 1c), rendering it aromatic. Because the overall electronic structure of a compound can be thought of as a hybrid of all of its possible resonance structures, it could be that the dianion-containing resonance structure of Braunschweig and colleagues' borole contributes to the unusual stability of the compound.
There is clearly much more work to be done to explore the chemistry of Braunschweig and colleagues' compound. Perhaps most intriguingly, the borole monoanion might open up new routes for the preparation of other borole derivatives that are otherwise difficult to make. If so, maybe we will finally have a chance to fully investigate the chemistry of this fascinating family of organoboron compounds.
de Meijere, A. & Diederich, F. (eds) Metal-Catalyzed Cross-Coupling Reactions 2nd edn (Wiley, 2004).
Eisch, J. J., Galle, J. E. & Kozima, S. J. Am. Chem. Soc. 108, 379–385 (1986).
Wakamiya, A., Mishima, K., Ekawa, K. & Yamaguchi, S. Chem. Commun. 579–581 (2008).
Braunschweig, H., Chiu, C.-W., Radacki, K. & Kupfer, T. Angew. Chem. Int. Edn 49, 2041–2044 (2010).
Parsons, T. D., Self, J. M. & Schaad, L. H. J. Am. Chem. Soc. 89, 3446–3448 (1967).
Blumenthal, A., Bissinger, P. & Schmidbaur, H. J. Organomet. Chem. 462, 107–110 (1993).
Imamoto, T. & Hikosaka, T. J. Org. Chem. 59, 6753–6759 (1994).
Segawa, Y., Yamashita, M. & Nozaki, K. Science 314, 113–115 (2006).
Segawa, Y., Suzuki, Y., Yamashita, M. & Nozaki, K. J. Am. Chem. Soc. 130, 16069–16079 (2008).
Loginov, D. A., Muratov, D. V. & Kudinov, A. R. Russ. Chem. Bull. 57, 1–7 (2008).
So, C.-W., Watanabe, D., Wakamiya, A. & Yamaguchi, S. Organometallics 27, 3496–3501 (2008).
Herberich, G. E., Wagner, T. & Marx, H.-W. J. Organomet. Chem. 502, 67–74 (1995).
Walton, J. C. et al. J. Am. Chem. Soc. 132, 2350–2358 (2010).
About this article
Chemical Science (2018)
European Journal of Inorganic Chemistry (2018)
Inorganic Chemistry (2017)
Nature Communications (2016)
Chemical Communications (2016)