One of the nice reactions I remember from graduate school is the Johnson-Corey-Chaykovsky reaction which entails epoxidizing a carbonyl group. The carbonyl is treated with dimethylsulfonium ylide, the ylide attacks the carbonyl, and the negative oxygen then knocks off the sulfur to form the expoxide. This epoxide can of course do a lot of interesting chemistry.
I always wondered if amides can be epoxidized in a similar way and I had not seen an example until now, and for good reason. Epoxyamines are inherently unstable because the nitrogen can act as an internal nucleophile and open the ring and the resulting product can undergo all kinds of unwanted reactions like polymerization. So if we want to form such compounds, the question naturally is how we can tie up that lone pair on the nitrogen so that it won't intervene.
Now it's been known for a long time that bridgehead amides behave more like ketones, and the reason postulated by Anthony Kirby of Cambridge among others (who did a lot of nice work on this topic) was that the C-N bond is twisted, making the lone pair unavailable for delocalization. With this background Jeffery Aube and his colleague have used a bridgehead amide to demonstrate Corey-Chaykovksy epoxidation in amides to generate spiro epoxyamines. They use their model system for illustrating several known reactions including protonation and ring opening with nucleophiles, electrophiles and Lewis Acids. The system behaves like a traditional epoxide in some cases but also shows exceptions (like inertness to ethereal boron trifluoride).
In organic chemistry, anything can be made to undergo a specific reaction if the right conditions and considerations of physical organic principles are judiciously applied. Biological organisms have put this fact to spectacular use. That's probably the most fascinating aspect of the field.
A review of Freeman Dyson's "Dreams of Earth and Sky"
3 hours ago in The Curious Wavefunction