The practice of using light to drive catalytic reactions is hotter than ever. The field of photoredox catalysis may trace its origins back to the late 1970s, but its popularity has grown considerably over the past five years. Most recently, David W. C. MacMillan at Princeton University and coworkers have been leading the charge, taking this class of reactions to new heights of synthetic utility.
MacMillan’s group published three high-profile papers about photoredox innovations last month alone. In the first of these, the team reported using photoredox catalysis to nudge nickel into the right oxidation state to catalyze the formation of aryl ethers from alcohols and aryl bromides under mild conditions (Nature 2015, DOI: 10.1038/nature14875; C&EN, Aug. 10/17, page 7). Aryl ethers are important compounds used in the manufacture of pharmaceuticals and cosmetics. Chemists had previously used nickel and other metal catalysts to drive this type of synthesis, but they needed harsh conditions or inefficient reagents to make the reactions work.
MacMillan and coworkers followed up that first paper with two more that demonstrate the power of photoredox catalysis. He discussed the studies on Aug. 19, during a session on “Innovation in Chemical Synthesis” at the American Chemical Society national meeting in Boston.
Each of the studies uses photoredox catalysts to oxidize and/or reduce reagents or other catalysts to accomplish a tough synthetic task. The three reaction schemes are conceptually complex, each involving two or three intersecting catalytic cycles. They therefore might not ever be used by chemists routinely, but their ability to make difficult or impossible reactions possible could offset that drawback for some syntheses.
Drugs, natural products, and other compounds typically contain multiple types of C–H bonds, each with a different inclination to either react or remain aloof. Generally, the weaker C–H bonds are more prone to react, but in the second of the MacMillan team’s three studies, the researchers used photoredox and two other catalysts to turn this proclivity on its head (Science 2015, DOI: 10.1126/science.aac8555).
Their approach alkylates alcohols at strong C–H bond positions, even when weak C–H bonds are present in the same substrates—a long-sought goal in organic synthesis. The three-catalyst strategy removes a hydrogen atom from a strong α-C–H bond (which is on the carbon carrying the alcohol group), making it susceptible to reaction.
Photoredox chemistry drives the reaction when visible light activates an iridium photocatalyst. The iridium catalyst goes on to generate a radical cation from an amine-based reactant. The radical cation then itself acts as a catalyst to abstract the hydrogen atom in the alcohol’s α-C–H bond, a bond that had already been weakened by a third catalyst. From there, the resulting α-hydroxy radical reacts with an allylic ester and, with another boost from the iridium catalyst, ends up cyclizing to yield a lactone. The complex process selectively polarizes and alkylates strong hydroxyalkyl C–H bonds in the presence of weaker allylic, benzylic, α-oxy, and α-acyl C–H groups.
“MacMillan and coworkers continue to demonstrate the power of utilizing cocatalysts in photoredox reaction manifolds to forge bonds in innovative ways,” commented process chemist Danielle M. Schultz at Merck & Co., in Rahway, N.J. The new alkylation scheme “provides a mode of reactivity that is not only thought-provoking but also atom-economical.” Merck recently built a photoredox catalysis reactor that can produce 25 kg or more per day of products for use in drug discovery.
The third study that MacMillan’s team recently unveiled involves a two-catalyst system that alkylates C–H groups in heteroarenes under mild conditions (Nature 2015, DOI: 10.1038/nature14885). The approach uses the Minisci reaction, in which alcohols or other reagents act as sources of alkyl radicals that then add to heteroaromatic compounds. This reaction is important in part because alkyl substituents have a strong influence on how a person’s body takes up and processes heteroarene-based drugs.
Generating alkyl radicals for the Minisci reaction typically requires strong oxidizing agents such as peroxides and/or high temperatures. But strong oxidizing agents are often too expensive and unstable for use in large-scale manufacturing, and harsh reagents and high temperatures can cause unwanted side reactions with sensitive functional groups.
The new method enables methanol and other simple, cheap alcohols to alkylate heteroarenes such as quinolines under much milder conditions. Once again, the reaction begins with visible light activating an iridium catalyst. That catalyst goes on to oxidize an organocatalyst that then abstracts hydrogen from an alcohol, forming an alcohol radical that adds to the heteroarene. The final alkylated product forms after the alcohol-heteroarene radical undergoes a spin-center shift—the transfer of the radical center from one carbon to another—and then loses water. The researchers demonstrated the reaction by alkylating the vasodilation drug fasudil and the heart failure drug milrinone.
“Incorporating a spin-center shift to reduce an alcohol intermediate to an alkyl group is clever, allowing direct alkylation with readily available alcohols,” commented C–H activation expert Jin-Quan Yu of Scripps Research Institute California.
Catalysis specialist Marco Bandini of Italy’s University of Bologna said that “the study fills an important gap in the literature—the direct alkylation of arenes with unactivated alcohols under mild conditions.”
What’s next for photoredox catalysis isn’t yet known. But groups like MacMillan’s are shining a strong light on its ability to make organic synthesis more versatile than ever.