what's in the flask:

Yesterday's mystery flask managed to confound pretty much everyone who guessed. Surprise, surprise: here's the reaction.

Here's a shot of the flask before the reaction with everything in there except the amine.

I know what you're thinking. "Dude, where's my solvent?" Everything is solid! ...At room temperature. The reaction is actually run at between 105 and 140 or so degrees, depending on the boiling point of the amine. Stick a septum on, blow nitrogen through it like crazy and inject the amine. This throws off water vapor, so pressure will build up. If you're not careful, you will hear a *POP* and rush over to see your septum off to the side. Folding it down and using a piece of wire to hold it in place seems to work well, but keep an eye on the bubbler. The lovely white crystals from yesterday's post can clog the needle.

The workup (which is simple but also a pain in the ass) and the rest of the procedure is detailled in this paper. As an added bonus, there's a bit about solar cells in there![1]

Beautiful reaction. It's never failed me. The only downside is that you should keep a LOT of chloroform handy, if you ever want your glassware to be clear instead of reddish in places.[2] Oh, and on this note, be sure to keep R long and greasy.

[1] Sadly, as is stated in the abstract, the solar cells they tried to make with these didn't quite work. Maybe sometime I'll have something to say about electrode interfaces. For the moment, though, I'd rather leave that one to the surface chemists and the engineers.

[2] Particularly if there are ground glass joints involved.


Mitch said...

What's the mechanism?


Russ said...

With an amine reacting with a cyclic anhydride, it's just nucleophilic acyl substitution X2 (first to make the amide, then the imide).

Amide plus acid doesn't normally react to give imide, but we've got a Lewis acid around, and it's an intramolecular reaction, and the equilibrium is driven by constant removal of the water byproduct, and we're heating the bejezus out of it.

Russ said...

Oh, and by 'we' above, I of course mean 'we the people considering the mechanism' and not 'we the person who did the reaction'.

I didn't mean to take any credit for the reaction.


Excimer said...

obvious question: how do you separate the imidazole from the PDI? What is PDI insoluble in that imidazole is not?

Ψ*Ψ said...

HCl/ethanol appears to get rid of the imidazole. Solubility of the product depends very, very strongly on what the side chains are...Most of the PDI derivatives I've made appear to be insoluble (or very, very sparingly soluble) in damn near everything.