Catalysts turn racemic mixtures into single enantiomers

In chemistry, breaking a mirror isn’t always a harbinger of bad luck. Using a trio of catalysts and some blue light, researchers have transformed a racemic mixture of mirror-image molecules into a single enantiomer of the same compound (ChemRxiv 2019, DOI: 10.26434/chemrxiv.8206634.v1).

Many molecules exist as right- or left-handed enantiomers, reflections of each other than cannot be superimposed. Designated as (R) or (S), they often have distinct biological activities—one enantiomer of a pharmaceutical may be far more effective than its counterpart, for example.

Chemists often rely on chiral catalysts to guide a reaction so that it makes just one of its product’s enantiomers. It’s far more unusual to take a racemic mixture, containing equal numbers of left- and right-handed molecules, and use a catalyst to convert all of the lefties into the righty versions of the same molecule.

Catalytic deracemization, as it’s known, is uncommon because it is difficult. Turning a racemic mixture into a single enantiomer decreases the entropy of the system, making it a thermodynamic slog. In addition, every intermediate step of the reaction from left-handed to right-handed molecule is as likely to go backwards and forwards, a principle called microscopic reversibility. As a consequence, the reaction generally tends to fall back to a racemic equilibrium.

Robert R. Knowles of Princeton University and Scott J. Miller of Yale University reasoned that they could smash through this racemic mirror by giving the reaction a jolt of energy, and then use chiral catalysts to steer each intermediate step so that one enantiomer was preferred over the other. Knowles presented the work at the Synthesis in Organic Chemistry conference in Cambridge, England, in July. (He was unable to speak to C&EN due to the policy of the journal where the work is under review.)

In their process, the researchers start with a racemic mixture of a cyclic urea, and aim to finish with just the (R)-enantiomer. First, they use blue light to activate an iridium catalyst so that it removes an electron from one of the urea’s nitrogen atoms, producing a radical cation of both (R) and (S) forms.

Then a chiral phosphate base plucks a proton from the C–H bond next to the nitrogen atom, forming an achiral carbon radical. Crucially, the base’s chirality ensures that this happens more readily with the (S) radical cation. The slower-reacting (R) intermediate is more likely to avoid deprotonation and recover an electron from the iridium catalyst, reverting to the desired (R)-enantiomer of the starting material. Finally, a chiral peptide containing a thiol group adds a hydrogen atom to one face of the achiral carbon radical, preferentially producing the (R)-enantiomer. Although the proton removal and hydrogen transfer reactions are only modestly stereoselective on their own, their effects multiply to convert almost all of the (S)-enantiomer to (R). After some fiddling with reaction conditions, the researchers were able to reach an enantiomeric ratio of 94:6, nearly in quantitative yield. The reaction worked just as well for cyclic ureas with a variety of substituents.

“This is a really clever way to get around the challenges associated with deracemization,” says David A. Nicewicz at the University of North Carolina at Charlotte, who saw Knowles present the work in Cambridge. “And it’s much more practical than a lot of prior deracemization strategies,” he adds. Thorsten Bach of the Technical University of Munich, for example, has previously carried out deracemizations on allenes and cyclopropylquinolones, but these involve molecules that are rather more esoteric than cyclic ureas, Nicewicz says.

Another advantage of Knowles’ and Miller’s approach is that the three key processes—light activation, proton removal and hydrogen transfer—all involve separate catalysts. In principle, this should make it easier to tweak each part of the process so that it can be adapted to different molecules. “It’s going to get people thinking about how they can implement these types of systems in other contexts,” says Sarah E. Reisman, who works on organic synthesis at California Institute of Technology. “I think this is an awesome paper.”

Still, deracemization is not about to displace asymmetric synthesis. If a catalytic reaction is available to transform a precursor into a single-enantiomer product, taking it one step closer to the ultimate goal of a synthetic sequence, most researchers would choose that over a deracemization, says Tomislav Rovis of Columbia University, who also saw the work at Cambridge. But if there was no asymmetric reaction available, or if a racemic mixture was particularly easy to make, then deracemization could be competitive, he says. “This isn’t a synthetic solution yet, but it is a big conceptual advance,” Rovis says. “In time, it could color the way many of us think about doing science.”