Homologating Carbonyls: How easy is that?

Synthetic chemists often have a need to tailor number of carbons in their compound of interest. When increasing this number by adding in new carbon units, we can think of this as homologation. Classically, if you wanted to homologate a carbonyl, you might reach for for a Wittig reagent such as Ph3P=CHOMe and then hydrolyze to the aldehyde that’s one carbon longer.

What König’s advance allows instead is a modular solution, where you can turn one carbonyl into a hydrazone (the group you start with) and then via diazoalkane intermediate react with a second carbonyl to forge that same bond and directly yield the desired homologated carbonyl. Classically, to help give the hydrazone the push off thermodynamic cliff that it needs, we deprotonate it and heat things up. Here, the energy injection comes in the form of 385 nm light.

There’s a lot to like about this method: cheap, easily accessible reagents, simple set up, and the fact that you can iterate to build up lots of complexity quickly. Some of the products that immediately caught my eye were those with basic pyridines and pi bonds, typically ripe for ylide formation or cyclopropanation with carbenes. Additionally, the hydrazone partner can be constructed from cyclopropyl ketones without 1,2-Wagner-Meerwein shift raining on the transformation’s parade; this lets you embed cyclopropanes in your molecule with a bit more peace of mind.

Overall, the method is a really nice practical advance that marries concepts from the storied history of tosylhydrazones as carbene precursors with an eye towards enabling iterative use in synthesis.

Is it quite at the level of say the Matteson homologation or Marty Burke’s MIDA boronate chembot production line right at this moment? Probably not quite yet, but the potential is there! I’ll be watching closely to see where this goes next.

Link to graphical summary Link to the paper