If organic molecules were trees, then the numerous carbon–hydrogen bonds within them would be leaves. A catalyst that targets one 'leaf' out of many similar other ones looks set to be a huge leap for synthetic chemistry. See Letter p.230
For much of the past century, organic chemists have focused on the reactions of functional groups: arrangements of atoms that have a characteristic, predictable reactivity profile. Such reactions have proved to be powerful tools for assembling structurally complex molecules for use in applications ranging from human medicines to pesticides, but they sometimes introduce inefficiencies that increase waste and costs. To expedite synthetic endeavours, attention has thus turned to a new generation of reactions — known as C–H functionalization reactions — that convert previously neglected carbon–hydrogen (C–H) bonds into useful reactive functional groups. A key question is how to precisely target a single C–H bond for reaction among so many others. On page 230 of this issue, Liao et al.1 report an outstanding advance that addresses this problem.
Carbon–carbon (C–C) and C–H bonds are the most common connecting units in organic molecules; if a molecule is viewed as a tree, then C–C bonds link together to make up the trunk and branches, which C–H bonds adorn as leaves. C–H bonds are so widespread that organic chemists almost always omit them when drawing chemical structures, instead showing only the skeleton of carbon atoms and heteroatoms (those that are neither carbon nor hydrogen). The fact that C–H bonds are not typically drawn also speaks to another, more subtle truth: they do not usually react in classic organic transformations.
This lack of reactivity is because the bonds are strong, are non-acidic and do not have inherent affinities for other chemical entities2,3. Collectively, these issues pose a difficulty: C–H bonds will seemingly be cleaved only by energetic reagents or under 'harsh' conditions (such as high temperatures), but such reactions are unlikely to be selective for a single C–H bond. For example, the combustion of hydrocarbons, a powerful process used to produce heat since the rise of human civilization, non-selectively cleaves all C–H and C–C bonds in the starting material to produce carbon dioxide and water. Reactions that achieve controlled and selective functionalization of C–H bonds are akin to carefully pruning a single leaf from a tree and grafting on a new branch at that position. Such transformations are underdeveloped, yet hold the promise of permanently changing how chemists design and synthesize molecules.
More than 50 years ago, inorganic and organometallic chemists discovered that transition metals interact with C–H bonds in unique ways, enabling ready C–H functionalization2,3. More recently, organic chemists have become interested in this reactivity with an eye to developing selective reactions of C–H bonds4,5. Existing strategies for obtaining such selective reactions generally involve an approach called substrate control, which depends on the intrinsic structural features and reactivity patterns of molecules. It remains a tremendous challenge to develop catalyst-controlled reactions that distinguish the subtle steric and electronic differences between C–H bonds in molecules that lack any functional groups (steric effects are those associated with the spatial crowding of chemical groups and atoms). Liao et al. tackled this issue directly by investigating the selective functionalization of pentane, a simple molecule that contains only C–C and C–H bonds.
Pentane contains a chain of five carbon atoms and, because of molecular symmetry, it has three distinct positions at which reactions can occur: the chain ends; the carbon atoms next to the ends (also known as the C2 positions); and the middle carbon atom. The authors sought to design a catalyst that would promote a highly selective reaction at a C–H bond at C2, a formidable undertaking (Fig. 1).
The same research group had previously pioneered a strategy for taming the reactivity of dirhodium carbenoids, a class of organometallic complex, by attaching electron-donor and electron-acceptor substituents to them6,7. The resulting species are still highly reactive, but are long-lived enough to participate in reactions with other molecules and to selectively target activated C–H bonds (such as those next to a C–C double bond, a benzene ring or an oxygen atom). Because pentane contains only unactivated C–H bonds, a strategy is needed to enable these catalysts to distinguish between the slight differences in steric and electronic properties associated with the possible reaction sites. This means that the size, shape and electronic environment around the reactive metal centres in the complexes need to be finely tuned.
The authors therefore tested a number of catalysts for reactivity and selectivity in the C–H activation of pentane. Synthesizing a library of new catalysts for screening is often a bottleneck in reaction optimization. Catalysts tend to consist of organic ligand molecules bound to metals, and so, for each catalyst, chemists generally synthesize the ligand first and add the metal in a subsequent step.
To expedite this typically tedious process, Liao et al. used an ingenious approach. They first synthesized a versatile dirhodium precursor complex, from which new catalysts could be prepared at a single stroke — streamlining catalyst discovery in a process that parallels methods used for discovering small-molecule drugs. In this way, the authors discovered a catalyst that preferentially functionalizes a C–H bond at the second carbon of pentane rather than C–H bonds at the first or third carbons, forming a new C–C bond at that carbon with more than 95% selectivity. This is a truly remarkable accomplishment, given the similarity between the three C–H bonds.
The authors went on to show that their reaction is similarly effective with other saturated hydrocarbons, and with some simple compounds that contain potentially reactive functional groups such as halides, silanes and esters. Moreover, the transformation is highly enantioselective (it yields only one of the two possible mirror-image isomers of the product). Enantioselective reactions are crucial for the synthesis of many biologically active compounds, including pharmaceuticals.
Although it remains to be seen how generally useful the optimal catalyst for C2 functionalization in pentane will be as a tool for synthesis, the authors' platform for catalyst screening and optimization will potentially allow catalyst structures for any given substrate to be tailored as needed. More broadly, Liao and colleagues' research represents an important step towards achieving high selectivity in catalytic C–H functionalization, even in the most challenging contexts. Reactions of this type would allow organic molecules to be groomed, pruned and shaped with bonsai-like precision, readying them for an array of potential applications.Footnote 1
Liao, K., Negretti, S., Musaev, D. G., Bacsa, J. & Davies, H. M. L. Nature 533, 230–234 (2016).
Labinger, J. A. & Bercaw, J. E. Nature 417, 507–514 (2002).
Arndtsen, B. A., Bergman, R. G., Mobley, T. A. & Peterson T. H. Acc. Chem. Res. 28, 154–162 (1995).
Brückl, T., Baxter, R. D., Ishihara, Y. & Baran, P. S. Acc. Chem. Res. 45, 826–839 (2012).
Neufeldt, S. R. & Sanford, M. S. Acc. Chem. Res. 45, 936–946 (2012).
Davies, H. M. L. & Beckwith, R. E. J. Chem. Rev. 103, 2861–2903 (2003).
Davies, H. M. L. & Morton, D. Chem. Soc. Rev. 40, 1857–1869 (2011).