Choosing One Among Many

Selectively modifying specific C–H positions in a molecule containing lots of similar bonds remains one of the largely unsolved problems in organic synthesis. Recent efforts suggest that scientists are beginning to better understand how it can be achieved.

The latest one comes from grad student Mark S. Chen and associate professor of chemistry M. Christina White of the University of Illinois, Urbana-Champaign, who report that it’s feasible to modify a specific methylene bond (a secondary C–H bond, R₂HC–H) in a molecule containing a number of similar bonds, even in a complex natural product (Science 2010, 327, 566).

Using an iron catalyst called Fe(PDP) that they developed to selectively oxidize tertiary C–H bonds (R₃C–H) (Science 2007, 318, 783), the researchers selectively convert methylene groups to hydroxyls or ketones. Methylene bonds are much harder to functionalize selectively than tertiary C–H bonds, which are less common. Chen and White earlier proposed that electronic, steric, and stereoelectronic properties of specific tertiary C–H sites can be used to predict those most prone to oxidation, and they now successfully employ those factors to functionalize methylenes selectively.

One of the most complex molecules they transform is dihydropleuromutilone. The tricyclic diterpenoid has 18 oxidizable C–H bonds—14 secondary C–H bonds and four tertiary ones—and a primary alcohol that’s highly susceptible to oxidation. Yet the researchers oxidize a single methylene site diastereoselectively to form a hydroxyl or ketone group in combined 62% yield.

This kind of selectivity can normally be achieved only by enzymes, White notes. “The ability to use the same simple rules that govern reactivity of traditional organic functional groups to direct site selectivity in high-energy oxidations of inert methylene bonds will change the way chemists view the hydrocarbon skeletons of complex molecules in synthesis,” she says. She adds that the iron catalyst can “operate under synthetically useful conditions of limiting organic substrate and achieve preparatively useful yields, on average 50% of monooxidized products.”

White’s iron catalyst is one of several C–H functionalization reagents developed in recent years. A group led by chemistry professor Huw M. L. Davies, now at Emory University, reported a way to functionalize C–H bonds selectively with carbenes (J. Am. Chem. Soc. 2000, 122, 3063). Stanford University associate professor of chemistry Justin du Bois’s group devised a technique for aminating C–H bonds with rhodium-nitrene reagents (Angew. Chem. Int. Ed. 2001, 40, 598). Associate professor of chemistry Melanie S. Sanford of the University of Michigan, Ann Arbor, and coworkers developed a palladium catalyst for C–H oxidation (J. Am. Chem. Soc. 2004, 126, 9542). And Yale University chemistry professors Robert H. Crabtree and Gary W. Brudvig and coworkers used a dimanganese catalyst to oxygenate C–H bonds (Science 2006, 312, 1941).

In a recent study, postdoc Ke Chen and chemistry professor Phil S. Baran of Scripps Research Institute, in collaboration with chemistry professor Albert Eschenmoser of the Swiss Federal Institute of Technology, Zurich, used some of these techniques to carry out selective transformations of several secondary and tertiary C–H groups (Angew. Chem. Int. Ed. 2009, 48, 9705). For example, they used du Bois’s rhodium-nitrene catalyst to selectively aminate a specific methylene of the natural product sclareolide, which contains a number of other methylenes and tertiary C–H bonds.

They report that three factors—the release of strain energy in transition-state formation, nucleophilicity, and spatial hindrance to the approach of a reagent—account for the tendency of equatorial C–H bonds, such as the one oxidized in sclareolide, to react more readily than axial ones. The challenge for the future of selective C–H functionalization, Baran says, “is to use directing groups or reagents that can override the inherent preferences of molecules” to be more susceptible to functionalization at some C–H groups than at others.

To further advance C–H functionalization capabilities, Davies is setting up a National Science Foundation center on stereoselective C–H functionalization. Participants include Emory assistant professor of chemistry Simon B. Blakey and computational scientist Djamaladdin (Jamal) G. Musaev and Scripps associate professor Jin-Quan Yu, in addition to du Bois and White.

Asked to comment on the work of Chen and White, Davies says some of the reported reactions “show some of the most impressive selectivities one has seen in the C–H oxidation field,” exemplifying the promise of C–H oxidation, whereas others generate product mixtures, don’t go to completion, or require large amounts of catalyst, demonstrating that the field still has a long way to go. Nevertheless, the work should help “spur a lot of interest in the development of new catalysts and new systems, both to enhance selectivities and more-complete conversions and to generate different selectivities,” Davies says.

“It’s been a long haul getting C–H activation to be useful, but now it seems to be moving faster,” Crabtree points out. “Maybe we’ve reached a point where things start to happen in rapid succession. Let’s hope so.”