Solvent-based syntheses have long been normative in all areas of chemistry, although mechanochemical methods (specifically grinding and milling) have been used to good effect for decades in organic, and to a lesser but growing extent, inorganic coordination chemistry. Organometallic synthesis, in contrast, represents a relatively underdeveloped area for mechanochemical research, and the potential benefits are considerable. From access to new classes of unsolvated complexes, to control over stoichiometries that have not been observed in solution routes, mechanochemical (or ‘M-chem’) approaches have much to offer the synthetic chemist. It has already become clear that removing the solvent from an organometallic reaction can change reaction pathways considerably, so that prediction of the outcome is not always straightforward. This Perspective reviews recent developments in the field, and describes equipment that can be used in organometallic synthesis. Synthetic chemists are encouraged to add mechanochemical methods to their repertoire in the search for new and highly reactive metal complexes and novel types of organometallic transformations.

Unsolvated tris(allyl)aluminum, (C₃H₅)₃Al, has never been isolated, although adducts with THF, OPPh₃, and pyridine are known. Attempts to make a base-free derivative with the bulky 1,3-bis(trimethylsilyl)allyl anion (A′) from the reaction of aluminum halides and K[A′] in ethers or by deprotonation of HA′ with R₃Al (R = Me, Et) were not successful. However, grinding AlX₃ and K[A′] in the solid state, followed by extraction of the mixture with hexanes, produces A′₃Al in high yield. A single-crystal X-ray structure demonstrated that the monomeric molecule contains σ-bonded allyl ligands (average Al–C 1.964(3) Å) and, like the known gallium analogue, is fluxional in solution. A DFT study of the parent (C₃H₅)₃Al indicates that the ligands would be of mixed hapticity, i.e., (η¹-C₃H₅)₂(η³-C₃H₅)Al, although the trimethylsilylated ligands evidently prevent that conformation in A′₃Al. A′₃Al reacts immediately with benzophenone in hexanes to produce Al(OCPh₂A′)₃ but negligibly with the bulky alcohol HOCPh₃. Grinding A′₃Al with ScCl₃ yields no observable product, although the same reaction with K[A′] produces the hexane-soluble A′₃Sc, whose X-ray crystal structure reveals three π-bound A′ ligands.

Solution-based reactions are a staple of synthetic chemistry—but what happens mechanochemically, when there is no solvent?

Owing to the strength of the C–F bond, the ‘direct’ preparation of Grignard reagents, i.e., the interaction of elemental magnesium with an organic halide, typically in an ethereal solvent, fails for bulk magnesium and organofluorine compounds. Previously described mechanochemical methods for preparing Grignard reagents have involved ball milling powdered magnesium with organochlorines or bromines. Activation of the C–F bond through a similar route is also possible, however. For example, milling 1- and 2-fluoronaphthalene with an excess of magnesium metal for 2 h, followed by treatment with FeCl₃ and additional milling, produces the corresponding binaphthalenes, albeit in low yields (ca. 20%). The yields are independent of the particular isomer involved and are also comparable to the yields from corresponding the bromonaphthalenes. These results may reflect similar charges that reside on the α-carbon in the naphthalenes, as indicated by density functional theory calculations.