clearly_not_atara - 5-7-2022 at 17:16
Using the very lax definition of superbase which is any neutral base stronger than pyrrolidine ![:D](./images/smilies/biggrin.gif)
DBN is typically produced from 2-pyrrolidone by reaction with acrylonitrile, a highly toxic petrochemical byproduct. It is a very useful reagent in
organic synthesis and a component (with AcOH) of a proposed distillable ionic liquid for 'green' cellulose dissolution developed by researchers at
Aalto in Finland:
http://ioncell.fi/
The use of acrylonitrile for DBN synthesis is a blot on its green reputation and prevents us from being able to make it, which is annoying. It also
makes it harder for individuals to experiment with the process (though DBN is commercially available). It might come as a surprise to find
that all of the carbon atoms can be snapped together by the condensation of lactic acid and GABA, as shown in the attached image:
![dbn.png - 51kB](https://www.sciencemadness.org/whisper/files.php?pid=674871&aid=93862)
The final reduction of the amide might hopefully be achieved with NaBH4, which is known to reduce tertiary amides, but not secondary amides, and
converts primary amides to the nitriles (which are apparently not further reduced). There are relatively few studies that mention the simultaneous use
of borohydride and DBU or DBN, but the amidine carbon would be expected to be significantly less electrophilic than the amide carbon, which is
promising for the use of nucleophilic reducing agents.
The cyclization hopefully generates a 5-membered ring imide rather than an 8-membered ring ester. 8-membered rings are typically not very stable, so
hopefully the ester will eliminate anyway if it forms. Alternatively, the alcohol could be methylated before addition, and would then eliminate MeOH
with H+/heat. Boric acid seems like a possible catalyst.
The acylation of GABA would hopefully happen spontaneously by the reaction of ethyl (or methyl) lactate. The similar (usually slow) reaction of other
esters with primary amines is well-known.
Hopefully, ammonia adds to the double bond much faster than it reacts with the imide; this seems relatively likely, since it is not used to cleave
other imides, and the rather noxious hydrazine must be used instead.