I enjoyed seeing this article in Nature Chemistry, where a Bergman cyclization is made to run (forwards and backwards) by pushing the molecules around with an atomic force microscope. It’s still startling to see this sort of thing in action, and I imagine that few find it stranger than Robert Bergman himself, who will be 74 this year.. (He’s already had the experience, back in the 1980s, of seeing that there were natural products that actually had been evolved to do the reaction as part of their mechanism of action).
Here’s a video from the folks at IBM-Zürich about this work. They position the AFM tip above the molecule (which is sitting on a sodium chloride layer) and apply a voltage pulse. You’ll note that at times they can see the radical and diradical intermediates – for instance, in 7, one bromine has departed and the molecule is notably nonplanar around the radical carbon. Diyne product 4 is characterized more thoroughly in the paper than just the image shown above, and it does indeed seem to be the right structure. By giving it a voltage kick in turn, they can send it back to the cyclized diradical intermediate. That, in turn, can then turn back into the structure 4 as shown, or the structure where the diyne ring ends up on the right-hand side of the molecule, and they can watch it sometimes make one, and sometimes make the other. Which is eerie.
At the moment, this sort of work is at the frontier of what can be done with the equipment available. It might always be a laboratory curiosity, even if the price of the hardware comes down – perhaps a day will come when a standard undergraduate experiment is doing the reversible Bergman with the department’s atomic force microscope. Or (as we get better at this kind of thing) it might turn into something more. That will be a mighty engineering challenge, but the rewards of being able to dink around individual molecules would be worth it. One near-term idea is mentioned by one of the paper’s co-authors in this writeup at Chemistry World: what if you use this technique as a platform for reaction discovery? Take some starting material and kick it around with various voltages to see what products you get, and you might find transformations that are worth pursuing under bulk conditions.