Many valued reagents and catalysts used in the preparation of organic compounds are highly reactive and are therefore incompatible with exposure to the open atmosphere — in some cases, dangerously so. The instrumentation required to carry out reactions with these compounds, such as high-vacuum apparatus and gloveboxes (isolation chambers), is costly and demands special training. Chemical processes that do not have such requirements are therefore more likely to make a big impact on how chemists synthesize molecules, be it for the development of new materials, pharmaceuticals or agrochemicals. The reactions reported by Fu and co-workers1in Nature are a case in point. Not only are they simple to carry out, but they also deliver a variety of useful products that are otherwise much more difficult to make.
Parallels are often drawn between the fields of organic synthesis and architecture2: aliphatic carbon–carbon (C–C) bonds are the architect’s ‘girders’ on which many structurally complex molecules are built. Installing these girders is challenging, and necessitates the use of highly reactive reagents. To add to the challenge, the orientation in which new C–C bonds are installed — the stereochemistry of the reaction — affects the overall shape of the final molecule3, which in turn can affect the molecule’s function in applications.
Fu and colleagues’ advance addresses these challenges. The authors describe a new C–C bond-forming reaction, known as a cross-coupling reaction, that produces one isomer of the reaction product to the near exclusion of the product’s mirror-image isomer (in chemists’ terms, the reaction is said to be enantioselective). Moreover, the process does not require the use of highly reactive and fragile reagents.
The authors’ approach requires three reagents: an alkene, a silane and an alkyl halide (Fig. 1a). Alkenes are not sensitive to air, which distinguishes Fu and colleagues’ reactions from the majority of cross-coupling reactions4, in which the alkene is replaced by an air-sensitive organometallic compound, either as a reagent or as the precursor to a reagent. The new reactions seem to involve an orchestrated set of events wherein the alkene first attaches to a catalytic nickel complex, which is generated in situ by a process that involves the silane reagent. The attachment of the alkene produces a transient reactive species, which then reacts with the alkyl halide to form the new C–C bond.
Fu and co-workers’ nickel-catalysed process is related to one reported5 by another team in 2016, but enhances the usefulness of that approach by addressing two key challenges. First, a catalyst had to be identified that not only promotes stereoselective C–C bond formation for an array of different substrates, but also activates the alkene without promoting side reactions between the silane and either the alkyl halide or the alkene. Second, reaction conditions had to be identified that allowed a base to drive catalytic cycles — which is difficult in this context, because bases often interconvert mirror-image isomers.
Not only is the use of alkenes as replacements for reactive organometallic reagents appealing in terms of its practical simplicity, but it also broadens the range of substrates that can be used in Fu and colleagues’ reactions. Alkenes are widely available, many are produced industrially on a large scale, and they can be generated by a variety of chemical processes. Alkenes are also especially attractive as reagents for chemical synthesis: they are chemically inert to a range of reagents, but can be induced to react in the presence of the right catalyst and under the right set of conditions. Impressively, the catalytic conditions used by the authors allow alkenes to react without interference when a variety of other common organic groups are also attached to the alkene (see Fig. 2a of the paper1).
Intriguingly, when Fu and colleagues used internal alkenes — in which the characteristic carbon–carbon double bond of the alkene is in the middle of a chain of carbon atoms — in their reactions, they observed a phenomenon called chain-walking6, which causes the double bond to migrate to the end of the carbon chain before reacting. This observation means that products obtained from an increasingly used type of reaction known as olefin cross-metathesis7 (which produces internal alkenes) might be suitable substrates. It will also be exciting to find out whether the step in which the alkene attaches to the nickel complex can be made to be enantioselective, because this would allow products containing multiple stereogenic centres (carbon atoms to which three different groups are attached by carbon atoms) to be generated enantioselectively.
A particularly notable feature of Fu and co-workers’ strategy is that a considerable array of alkyl halides can be used, some of which are not effective substrates for cross-coupling reactions with organometallic reagents. For example, the authors use alkyl halides known as secondary α-halo amides in their reactions, and show that this provides a simple and enantioselective route to prepare compounds that contain a carbonyl (C=O) group next to a stereogenic centre (Fig. 1b). Such compounds are potentially versatile intermediates for chemical synthesis, and have most commonly been prepared using a much less efficient approach based on the use of compounds called chiral auxiliaries8. The researchers also demonstrate that they can use their enantioselective reactions to make certain fluorine-containing compounds (see Fig. 1c, for example), which might be useful in medicinal chemistry. Moreover, the chemistry can be used to make compounds that contain quaternary stereocentres (Fig. 1d) — carbon atoms to which four different groups are attached by carbon atoms, which are some of the most difficult structures to prepare enantioselectively.
Overall, this advance is a much-needed method for the enantioselective synthesis of an impressive assortment of versatile small organic molecules, many of which will be of value to research at the frontiers of chemistry.
Nature 563, 336-337 (2018)