Organic chemistry

Forgotten hydrocarbons prepared

Article metrics

Dendralene hydrocarbons have a reputation for being difficult — it seemed that these molecules couldn't easily be made. A practical synthesis of dendralenes opens them up for study, and reveals some surprises.

The presence of unsaturation — double or triple bonds — in molecules often largely determines their structural properties and chemical behaviour. Unsaturated hydrocarbons are well known to chemists, but one class, known as dendralenes, has been neglected, in part because the compounds were thought to be unstable. In Angewandte Chemie, Payne et al.1 report a practical synthesis of dendralenes, and find that they are stable after all. Intriguingly, the physical and chemical properties of the compounds depend on whether there is an odd or even number of double bonds in the molecules.

There are six different ways in which carbon–carbon double bonds (C=C bonds) can be assembled to form unsaturated hydrocarbons2. The simplest is to connect these C=C bonds using carbon–carbon single bonds, to produce chain-like molecules known as acyclic conjugated polyolefins (Fig. 1a). Some crucial naturally occurring compounds, such as vitamin A and β-carotene, are derivatives of this class, whereas polymeric versions are familiar to materials scientists as 'organic metals' — so-called because of their conducting properties.

Figure 1: Six varieties of unsaturation.
figure1

a–f, Carbon–carbon double bonds can be assembled in six different ways to construct distinct families of unsaturated hydrocarbons, examples of which are shown. In f, the first member of the family is known as allene, whereas the others are known as cumulenes. Carbon atoms shared by two double bonds are represented as bold dots.

Alternatively, C=C bonds may be connected together using single bonds to form ring-shaped, or cyclic, molecules (Fig. 1b). The resulting hydrocarbons, known as annulenes, are either aromatic (for those with an odd number of C=C bonds), or antiaromatic (for flat molecules that have an even number of C=C bonds). Annulenes have had pivotal roles in the development of theories of the structure and reactivity of organic molecules. More practically, the aromatic hydrocarbons benzene, toluene and xylene are important feedstocks for the chemical industry.

Then there are the radialene3 and fulvene2 classes of hydrocarbons, which have more exotic-looking structures. In radialenes, C=C bonds 'radiate' from a central ring of carbon atoms that is formed from carbon–carbon single bonds (Fig. 1c), whereas, in fulvenes, C=C bonds radiate from annulene-like rings (Fig. 1d). Dendralenes4 (Fig. 1e) are chain-like versions of radialenes — on paper, dendralene structures look like radialenes in which a single bond has been broken (although it isn't actually possible to convert a radialene to a dendralene). The final structural arrangement of hydrocarbons occurs when two or more C=C bonds share a common carbon atom; allenes contain two C=C bonds, whereas cumulenes contain three or more5 (Fig. 1f).

Although dendralenes exhibit a curious electronic phenomenon known as cross-conjugation — a feature also found in numerous organic dyestuffs — they have long been neglected by organic chemists. The reason is simple: the compounds could not be made readily in sufficient amounts for further study, and were assumed to be too unstable to handle under normal laboratory conditions (on the basis of what was known from the few reported examples of dendralenes4). Payne et al.1 show that this assumption is wrong. They devised a general method for the preparation of dendralenes, and used it to make the first six members of the family. Their synthetic route provides more than enough material for further studies of the reactivity and structures of these mysterious compounds.

Payne et al. constructed their compounds from molecular building blocks that already contained one or more C=C bonds. For example, they used a magnesium-containing reagent (easily made from a commercially available compound) as the source of a diene fragment, which contains two C=C bonds connected by a single bond (Fig. 2a). They reacted this with other double-bond-containing compounds — various vinyl halides — in a nickel-catalysed process6,7 that 'stitched' together the unsaturated hydrocarbon groups. In this way, Payne and colleagues prepared dendralenes containing three to five C=C bonds in good yields. The authors prepared higher oligomers (containing up to eight C=C bonds) using similar processes, providing each member of the series in gram quantities and as analytically pure substances. Previously, only milligram quantities could be made.

Figure 2: Preparation and reactions of selected dendralenes.
figure2

Payne et al.1 have prepared dendralenes by stitching together unsaturated hydrocarbon fragments from other compounds. a, In these examples, the diene fragment (red) of a magnesium-containing compound is coupled in nickelcatalysed reactions to hydrocarbon fragments (various colours) of halogen-containing compounds, to make the first three members of the dendralene family. b, The authors also investigated the reactivities of dendralenes in Diels–Alder additions. In these reactions, a diene fragment (red) reacts to form a six-membered ring. Another diene is formed in the product, which can, in principle, take part in another Diels–Alder reaction. The cycle continues until no more dienes are formed.

Like annulenes, the physical and chemical properties of the newly prepared dendralenes depend on the number of C=C bonds in the molecule: the properties of the even-numbered members of the series are distinctly different from those of its odd-numbered members. A good example is the thermal stability of the compounds. Dendralenes that have an even number of C=C bonds can be kept at room temperature for weeks without any significant decomposition, whereas their odd analogues have much shorter half-lives. A similar dichotomy occurs for the electronic spectra of these compounds, and in their chemical behaviour.

Perhaps the most likely initial use of dendralenes will be in organic synthesis, acting as sources of dienes in 'cycloaddition' reactions. The most widely used cycloaddition reaction is the Diels–Alder addition, because this is the best method for preparing rings of six carbon atoms. When dendralenes are used in Diels–Alder additions, the reaction product will contain a new diene fragment, which can in principle undergo another Diels–Alder addition, and so on, until no more diene units can be generated (Fig. 2b). Such 'diene-transmissive Diels–Alder processes'8 allow the rapid generation of molecular complexity from relatively simple starting materials in a one-pot operation.

Payne et al.1 found that the reactivity of dendralenes in Diels–Alder additions again depends on the number of C=C bonds in the molecule: odd-numbered dendralenes react faster than their even-numbered counterparts. Furthermore, only the endmost dienes of odd-numbered dendralenes take part in reactions, whereas diene subunits throughout the even-numbered dendralenes react. The authors rationalized this surprising chemical effect using quantum mechanical calculations, which suggest that the geometries of the bonds in the dendralenes are at least partly responsible. In the odd-numbered dendralenes, the endmost diene subunits adopt a conformation that has long been known to be optimal for Diels–Alder reactions. These subunits therefore react quickly, and preferentially to the other diene subunits. But all of the diene subunits in the even–numbered dendralenes adopt an unfavourable conformation for Diels–Alder additions; their reactions are therefore slower than in the odd-numbered dendralenes, and no particular diene subunit reacts preferentially to the others.

With the dendralenes now available in sufficient amounts for further study, we can expect the discovery of many new reactions. The resulting products should show interesting chemical and structural properties, and would not have been available using conventional methods of synthesis.

References

  1. 1

    Payne, A. D., Bojase, G., Paddon-Row, M. N. & Sherburn, M. S. Angew. Chem. Int. Edn 48, 4836–4839 (2009).

  2. 2

    Hopf, H. Classics in Hydrocarbon Chemistry (Wiley-VCH, 2000).

  3. 3

    Maas, G. & Hopf, H. Chemistry of Dienes and Polyenes Vol. 1 (Ed. Rappoport, Z.) 927–977 (Wiley, 1997).

  4. 4

    Hopf, H. Angew. Chem. Int. Edn 23, 948–960 (1984).

  5. 5

    Krause, N. & Hashmi, A. S. K. (Eds) Modern Allene Chemistry Vol. 1 & 2 (Wiley-VCH, 2004).

  6. 6

    Corriu, R. J. P. & Masse, J. P. J. Chem. Soc. Chem. Commun. 144a (1972).

  7. 7

    Tamao, K., Sumitani, K. & Kumada, M. J. Am. Chem. Soc. 94, 4374–4376 (1972).

  8. 8

    Tsuge, O., Wada, E. & Kanemasa, S. Chem. Lett. 12, 1525–1528 (1983).

Download references

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.