Clipping the carbon–carbon bond

Article metrics

The bonds between two carbon atoms tend to be hard to break. But careful manipulation of the starting material can make the process remarkably easy.

Some chemical reactions are simple to state but difficult to do. One such reaction is the rapid cleavage of a carbon– carbon bond by inserting a metal atom between the two carbons. This is an important process, as it 'activates' the bond and allows potentially more useful molecules to be made from relatively unreactive starting materials.

A few years ago, Milstein and co-workers from the Weizmann Institute at Rehovot, Israel, successfully demonstrated this reaction1. Now, in the Journal of the American Chemical Society2, they report a full characterization of this unusual process, which reveals that it happens so readily it even occurs at temperatures as low as −70 °C. This is remarkable because previous C–C bond-breaking reactions3 have generally required temperatures of at least 100 °C.

In the past, such reactions have led to valuable applications in which a small amount of the metal — acting as a catalyst — triggers the conversion of large amounts of starting materials (the substrate) into useful products. Catalysis is used in a wide variety of areas from petroleum refining to medicinal drug synthesis.

Metal catalysis occurs naturally in the enzyme reactions of biochemistry, as well as in the organic reactions of synthetic and industrial chemistry. An important goal in the chemistry of the transition metals is to understand why these metals — found in the central part of the periodic table — so effectively catalyse such a wide variety of reactions. Typical metal catalysts consist of a transition-metal atom surrounded by a set of ligands that determine the catalytic behaviour of the central metal.

In many catalytic reactions, the first step of the pathway is the insertion of a metal atom into a single bond of one of the starting materials, a reaction known as 'oxidative addition'4. In one of the earliest successes in this area, molecular hydrogen, H2, a relatively unreactive molecule, underwent rapid oxidative addition (see Fig. 1a).

Figure 1: Using metal complexes to break bonds.

a, An early example of bond cleavage using a transition-metal catalyst. The unreactive bond in molecular hydrogen is broken when the metal (M), attached to ligands (Ln), is inserted between the two hydrogens. b, In the reaction studied by Milstein and colleagues2, the C–C bond in the substrate (red) is impossible to break with conventional reagents, but it can be cleaved using the rhodium complex [(C2H4)2RhCl]2. Incorporating phosphorous groups (P(C4H9)2) and/or nitrogen groups (N(C2H5)2) at positions close to the C–C bond (Q and Q′) causes the substrate to bind tightly to the metal reagent (source of the Rh–Cl fragment). If the phosphorus groups are present at both Q and Q′, then one of the C–H bonds can be cleaved to give product 1, or — much harder — the C–C bond can be cleaved to yield product 2. But if a nitrogen group is introduced at Q′, then only the C–C bond is broken and product 2 generated.

As catalysis developed, less reactive — chemically strong — bonds were tackled. For example, the carbon–hydrogen bond in an alkane (such as ethane, CH3–CH3) is particularly strong, making the molecule very stable. In fact, the bond is even less reactive than the H–H bond in molecular hydrogen, and resisted cleavage by oxidative addition for many years. In the 1970s, such reactions were found and used in catalytic dehydrogenation reactions5, which convert unreactive alkanes into more chemically reactive alkenes (such as ethene, CH2 = CH2) by removing H2. In the past decade, ways have also been found to break the extremely inert fluorocarbon C–F bonds in the same way, and this has been used to convert chlorofluorocarbons into more environmentally friendly hydrochlorofluorocarbons, for example4,6.

Perhaps the most challenging of all such oxidative additions involves the C–C bond, not only because it is relatively strong, but also because it is always buried so deep within the molecule. The reaction studied by Milstein and colleagues2 is shown in Fig. 1b. The C–C bond to be broken, shown in red, is first enticed into close proximity with the metal reagent (rhodium) by attaching groups containing phosphorus or nitrogen atoms to the substrate (at points Q and Q′). These atoms initially bind very strongly to rhodium, and in doing so draw the central ring with its target C–C bond close to the metal. Because the phosphorus atom is larger than nitrogen, the orientation of the C–H and C–C bonds in the substrate is affected. This allows the substrate containing two phosphorus atoms to undergo both the easier C–H bond cleavage to yield product 1, and the more difficult C–C cleavage to give product 2. But when the smaller nitrogen atom replaces one phosphorus, the substrate orientation is altered so that rhodium only has ready access to the C–C bond. Rapid breaking of this bond occurs and only structure 2 is produced3,4.

The results of Milstein and colleagues' investigations2 indicate that if the metal reagent can be brought close enough to the C–C bond, the actual barrier to the oxidative addition is relatively low. The next step is to investigate whether this concept can be developed for use in industrial processes. One promising possibility would be a technique for rearranging the carbon skeleton of the substrate by selectively breaking its C–C bonds. This sort of chemical surgery can be achieved through catalytic cracking — widely used in the petrochemical industry — but only at temperatures of around 500 °C. In contrast, Milstein and colleagues' reaction occurs rapidly at room temperature. But it relies on the phosphorus and nitrogen groups to direct the rhodium catalyst to the desired bond. To become useful in industrial catalysis, such reactions must work without needing these directing groups. This remains a formidable challenge for the future.


  1. 1

    Gandelman, M. et al. Organometallics 16, 3981– 3986 (1997).

  2. 2

    Gandelman, M., Vigalok, A., Konstantinovski, L. & Milstein, D. J. Am. Chem. Soc. 122, 9848–9849 (2000).

  3. 3

    Crabtree, R. H. Chem. Rev. 85, 245–269 ( 1985).

  4. 4

    Crabtree, R. H. The Organometallic Chemistry of the Transition Metals 3rd edn 149–361 (Wiley, New York, 2001).

  5. 5

    Jensen, C. M. Chem. Commun. 2443–2449 ( 1999).

  6. 6

    Burdeniuc, J. et al. Chein. Ber. 130, 145– 154 (1997).

Download references

Author information

Correspondence to Robert H. Crabtree.

Rights and permissions

Reprints and Permissions

About this article

Further reading


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.