Randomized radical reaction leads to selective cyclizations

A new, convenient strategy to build common nitrogen-containing rings balances the rates of competing reactions (J. Am. Chem. Soc. 2026, DOI: 10.1021/jacs.6c01294). Dubbed “radical sampling,” the unusual approach converts simple aldehyde and amine substrates into valuable pharmaceutical motifs, such as piperidines and morpholines, and leverages differences in the rates of cyclization to selectively form these six-membered rings.

These functional groups “play a fundamental role in pharmaceuticals and agrochemicals, and access to these structures represents one of the most important challenges for synthetic chemists,” says Mattia Silvi, an organic chemist at the University of Nottingham who wasn’t involved in the new work. These motifs have become especially prevalent in small-molecule therapeutics in recent years, thanks in part to their pH-tunable properties and ability to increase the solubility of large drug structures.

But synthetic methods to form these rings have not kept up with this rising demand. Despite the rings’ deceptively simple appearance, chemists typically need to carefully engineer their starting materials to direct the reaction to a single product, adding steps and shrinking the pool of potential substrates for the reaction. The new work gets around these limitations by effectively separating the substrate activation from the cyclization step, explains co-first-author Qinyan “Anna” Cai, a PhD student in Dave MacMillan’s lab at Princeton University.

The team began by combining aldehyde and amine starting materials and treated the resulting intermediate with a light-activated catalyst. This radical-forming step randomly pulls hydrogen atoms from C–H bonds on the intermediate’s backbone.

The resulting radical intermediates are primed for cyclization, but only those in positions able to react quickly—crucially, faster than the competing quenching process—lead to a cyclic product. The remaining radicals get quenched and simply revert to the starting substrate, ready to begin the process again. As a result, a single product is mainly formed. The approach essentially pits cyclization against quenching, and the relative rates of these competing processes determine the fate of each individual radical, Cai explains.

“You essentially allow the substrate to decide which product it’s going to reach by which barrier is the lowest,” co-first-author Noah Bissonnette explains. The selectivity for six-membered rings rather than four-, five-, or seven-membered ones is therefore an intrinsic part of the mechanism.

“You want to set up your different rates so that you have product formation being the fastest. Next fastest needs to be back donation of hydrogen to regenerate the precursor and finally, all of your deleterious steps. Only when these three processes are set up in that exact kinetic order does it achieve selectivity,” Bissonnette says.

The reaction exhibits good functional-group tolerance and likely has the potential to be scaled up, Silvi says. He also admires how the building blocks can be mixed and matched. “It can be achieved simply by starting from appropriately substituted aldehydes or amines, which are generally readily available materials,” he says.

Radical sampling will undoubtedly streamline access to these valuable ring structures, particularly for medicinal chemists looking to rapidly generate libraries of compounds for drug discovery, says Armido Studer, an organic chemist at the University of Münster. “You can basically do radical chemistry at a position where, at first glance, you would never expect to see a reaction occur through intermolecular hydrogen abstraction,” he says. “A stereoselective version would be an extremely valuable [extension]—for example, if you could use a chiral Brønsted acid to induce selectivity in the cyclization.”