Project 1:
This work was published in Science magazine: Science 2025, 387, 1167–1174.
In order to achieve chemoselective oxidative cleavage of arenes in the presence of olefins, it was necessary to devise a new reagent capable of this transformation. Previously (Nature 2022, 610, 810), the group had established that purple light excited nitroarenes were able to oxidatively cleave olefins under mild conditions. While this exact reactivity was undesired in our case, we wondered whether photoexcited nitroarenes could also have a role in oxidatively cleaving other π-systems, such as arenes. Consequently, we screened nitroarenes for their ability to oxidatively cleave arenes and olefins (under identical conditions), and it became evident that nitrobenzene derivatives with lowest-lying π,π* triplet excited-states, where not only competent at effecting the desired arene cleavage reaction, but also exhibited no reactivity toward olefins. Nanosecond transient absorption spectroscopy confirmed that the π,π* triplet excited state of 3,4-dimethoxynitrobenzene is not quenched by olefins (e.g. cyclooctene, styrene), but is strongly quenched by arenes such as 1-methoxynaphthalene. Competition experiments revealed that arenes featuring tethered olefins (terminal, disubstituted, trisubstituted alkenes and styrenes) could be oxidatively cleaved by photoexcited 3,4-dimethoxynitrobenzene with full selectivity. Finally, a scope evaluation determined that electron rich naphthalenes, anisoles, indoles and pyrroles, could be selectively cleaved by this reagent in good yeidls, in the presence of a number of sensitive functional groups, as demonstrated by the late-stage transformations of 10 drug molecules.
Project 2:
We postulated on the basis of pioneering work by Roberts (J. Chem. Soc. Chem. Commun., 1988, 480) that ammonia borane could serve as a solid hydrogen source for radical hydrogenation. Indeed, our optimisation campaign revealed that alkenes (1.0 equiv) in the presence of ammonia borane (1.0 equiv.), a photocatalyst (anthraquinone, 0.2 equiv.), and a thiol precursor (p-tolyl disulfide, 0.1 equiv.) in CHCl3 (0.1 M), with purple light irradiation, were hydrogenated to alkanes in good to excellent yields. Pleasingly, functional groups typically prone to hydrogenolysis or reduction under traditional hydrogenation conditions (e.g. azides, benzyl ethers, epoxides, N-Cbz, nitriles) were tolerated under the reaction conditions, and compatibility with complex molecules was demonstrated by the late-stage hydrogenation of xx molecules. Experimental mechanistic studies and computation indicated the reaction proceeds through a complex mechanism involving HAT with a hydridic B-H of ammonia borane giving a boron-centred radical. This species undergoes XAT with the chloroform solvent (up to 2 times if a subsequent HAT then XAT occurs) delivering ammonia mono- or dichloroborane. These species undergo hydroboration with alkenes, providing B-alkylated ammonia borane species. Finally, H-atom abstraction of an N-H of these species (e.g. by boryl radical), subsequent beta-scission and HAT with the thiol catalyst delivers the final product.
