A synthesis combining radical methods old and new

Free radicals! Everything about them sounds scary and extreme – and for a long time it was. In the early days of radical chemistry, high temperatures, UV light and dissolving metals were the only way to get at these often highly reactive intermediates. These days, while these species are still a bogeyman in adverts for many dubious health products, they have become an indispensable tool to the modern synthetic chemist. One can only imagine the dismay of radical chemistry pioneer Moses Gomberg, who tried to reserve their study for himself back in 1901.

During my PhD studies, I ran a range of textbook radical reactions, including Birch-type reductions, benzylic brominations and even some organotin chemistry, which I can still smell to this day. These were typical old-school radical reactions – that is to say, fairly dangerous and unappealing. However, in the 10 years or so since I completed my graduate education, some of the hottest areas of research in organic synthesis have been those that allow us to more easily generate and harness open-shell species, which is the essence of most photoredox catalysis. With modern methods to generate radicals, compute bond strengths and explain their reactivity, we can now leverage these reactive but predictable species to perform powerful and unique chemistry – without getting sunburn from mercury lamps or having to bleach tin off all of our glassware!

A recent total synthesis that showcases a range of old and new radical methodology is the total synthesis of norzoanthamine by Shuanhu Gao and co-workers at East China Normal University in Shanghai.1 One of the redoubtable zoanthamine alkaloids, this complex target has attracted considerable attention, but the group manages a stand-out synthesis thanks to its deft use of radical chemistry to set both the C12 and C22 quaternary stereocentres.

It never fails to make me feel like an alchemist – or a magician

We pick up the synthesis as the team is about to set the final quaternary stereocentre at C12. Radical chemistry is uniquely suitable for forming often-tricky all-carbon quaternary centres as it easily forms tertiary radicals and is not slowed down much by sterics. Here, a recently reported cobalt-catalysed hydrogen atom transfer (HAT) protocol generates the desired radical under mild conditions, which rapidly adds into the adjacent arene to give a single product diastereomer. Next up, a favorite reaction of mine, the Birch reduction, brings the aryl ring down to the correct enone oxidation state (figure 1). The inky blue–black colour that occurs as sodium dissolves in liquid ammonia is perhaps the closest that we can get to seeing unpaired electrons with the naked eye and never fails to make me feel like an alchemist – or a magician.

An image showing a reaction scheme

Figure 1. Something old, something new, something basic and something blue

Although that ring does now contain the ketone and alkene found in the target, both are initially in the wrong place. The first step to remedy this is to reduce the double bond, but the challenge here is that tetra-substituted alkenes are notoriously unreactive. Furthermore, classical hydrogenation methods effect syn addition of hydrogen, which would give the undesired cis decalin ring junction. Here, radical chemistry again steps up with a recently reported HAT reaction that I’ve mentally filed as the ‘Shenvi anti-hydrogenation’ (figure 2). This unique transformation involves stepwise addition of hydrogen atoms to an alkene and allows the configurationally flexible radical intermediate to find its thermodynamically most comfortable position, selectively delivering the desired trans-stereochemistry.

An image showing a reaction scheme

Figure 2. I can’t believe it’s not hydrogenation, but it is (sort of)

From here, there’s still a lot of work to do on the lower half of the molecule, but with the tricky carbocyclic framework established, the team is well set up for success.