The HAcTive Strategy

I've been working in the field of photocatalyzed Hydrogen Atom Transfer (HAT) for a while now. Here, a photocatalyst absorbs light of a proper wavelength and, when in the excited state, can break homolytically strong aliphatic C-H bonds in a molecule (often referred to as hydrogen donor). The resulting carbon-centered radical can be deployed for synthetic purposes.

The main fact that captured my interest as a fresh MSc student was the fact that with HAT you can activate molecules that are routinely considered solvents (hence, they must be inert!) in traditional catalysis. For example, cyclohexane or tetrahydrofuran are common solvents for transition metal catalysis, however they are ideal substrates for photocatalyzed HAT.

During my PhD at the PhotoGreen Lab, I've contributed developing new synthetic strategies and photocatalysts through this manifold. During my postdoc in Pavia, my colleagues and I have tried combining HAT with emerging technologies such as electrochemistry and biocatalysis. Later on, during my MSCA, I boosted photocatalyzed HAT by using flow chemistry with the goal of bridging the gap between academia and industry.

During these research experiences, which I genuinely enjoyed, I observed that all my methodologies were hampered by the same two limitations: regioselectivity and inefficiency.

• Regioselectivity: all organic molecules contain a plethora of C-H bonds and distinguishing them can be challenging. Although HAT methodologies can display excellent selectivity, the outcome can only be rationalized a posteriori upon a careful evaluation of substrate-, medium-, or photocatalyst-dependent effects.

• Inefficiency: in most synthetic protocols based on photocatalyzed HAT, an excess of the hydrogen donor (up to 20 equivalents) must be used to achieve serviceable conversion. This may not represent a big problem when you need to functionalize cyclohexane or tetrahydrofuran, but it is something you want to avoid if your substrate has been synthesized via a lengthy, tedious 10-step procedure.

As I was returning to Italy from the NRG and settling at the University of Parma for my tenure track, the urgency to find a solution to these problems began to rise. I remembered that during my PhD, intrigued by HAT as a methodology to activate formyl C(sp2)-H bonds to give acyl radicals, I had attempted a radical acylation of methyl vinyl ketone with phenylacetaldehyde via HAT. I had observed both the expected acyl-Giese (the 1,4-diketone) and the hydroalkylation product as a major byproduct. This occurs because the phenylacetyl radical generated via HAT undergoes prompt decarbonylation to unveil the corresponding tolyl radical, which is trapped by the olefin. What I had considered a failed experiment, started becoming an idea.

What if we leverage the decarbonylation of acyl radicals to address the so-far unmet challenges of photocatalyzed HAT?

I started working to this project in Parma with Vittoria (3rd year PhD student) and Serena (one of my first MSc students as PI), who did a great experimental work. Valerio (postdoc) joined later and masterfully helped me in the peer-review phase developing the concept of "umpolung" of HAT. Coincidentally, I discovered that the PhotoGreen Lab at the University of Pavia was working on a similar concept. We chose to collaborate to maximize the potential of our idea.

In the paper, you will learn about how we developed this concept. A hint? Aldehydes work as super hydrogen donors !

Luca, Sere, Vitto and Valè (in the sky) on the day of Sere's MSc graduation day.

Link to publication: https://pubs.acs.org/doi/10.1021/jacsau.5c00530.