taffy - 12-10-2014 at 01:05
The following experimental procedure is based of the attached patent reference United States patent US3663589. The patent basically states that
urea, amidosulfonic acid (sulfamic acid), and a halobenzoic acid such as 2-chlorobenzoic acid are dry distilled to form the corresponding nitrile in
good yields. The ref. also goes on to mention that good yields are also possible with similar halobenzoic acids, and many other benzoic acids in
general (but not limited to just phenyl radicals, even nitrogenous substrates may be used within certain parameters).
From the patent:
"EXAMPLE 2
31.4 parts of 2-chlorobenzoic acid, 36 parts of amido
sulphonic acid and 16.8 parts of urea are mixed in a
vessel equipped with an air-cooled condenser and the mix
ture is reacted for 1 hour at an inner temperature of 220°-
230°. The 2-chlorobenzonitrile thus formed is distilled oil
under normal or reduced pressure and, if desired, is crys
talllized from petroleum ether.
In place of 2-chlorobenzoic acid, an equivalent amount
of one of its alkali metal or ammonium salts or the anhy
dride of the acid can be employed with equally good suc
cess."
Based off this patent example I was able reproduce the procedure several different times, all with similar yields. A typical run goes as
follows:
2-chlorobenzonitrile
5g 2-chlorobenzoic acid, 5.8g sulfamic acid, and 2.7g urea are combined in a 250ml round bottom flask on a sand bath or aluminum foil bath. The flask
is then shaken by hand for several minutes to ensure adequate, intimate mixing of solids. The single neck flask is then arranged for simple
distillation with an attached liebig condenser(no circulating water necessary). The high temps might crack the glass. I was unwilling to find out so a
basic air condenser of any kind will do just fine...the longer the better.
The mixture is gradually heated and within 5-15 minutes the mixture forms a melt. Shortly after the melt is formed effervescence is observed. Around
30-60 minutes depending on heat gradients in the flask the thermometer will read approx. 220-230C. At which point brilliant white solids will form in
the receiving flask. This temp. will remain consistent for 5-15 minutes and then the distillation temp will begin to gradually fall.
At this point the flask is allowed to cool to ~30-80C and ~100ml water is added along with a mag. stir bar. The somewhat charred solids after the
distillation may appear at first glance to be stuck like tar to the flask but go into solution fairly easily upon heating with mag. stirring. The
contents of the flask are then steam distilled until most of the volume of the solution is removed. If all the water is removed the residue is easily
removed with hot water. The residue consists of what appears to be several grams at most of a light brown solid.
The distillate is triturated to remove it from the flask as it clings to the walls and then is buchner filtered to yield 3.8g of a white, brittle
solid with a strong odor of sweet almonds and cinnamon The crude solid mp=43-45C via a solid aluminum melting block apparatus consistent with the
lit. mp. value.
Any ideas on a mechanism?
Attachment: US3663589 Process for the production of nitriles.pdf (357kB)
This file has been downloaded 870 times
CuReUS - 12-10-2014 at 01:13
this reaction reminds me of a similar one called the ugh G(ugi reaction) in which an aldehyde/ketone ,amine and cyanide are brought together to make a
bis amide
Boffis - 12-10-2014 at 12:24
This is effectively a one-pot version of two reaction previously discussed on this forum, namely the formation of benzamide from benzoic acid and urea
and the conversion of amides to nitriles by heating with sulphamic acid. I seem to remember there are links to the relevant papers in both case, I
certainly them both but I can't say I have ever tried them.
taffy - 18-1-2015 at 23:54
When scaling this up beyond 15g with the same substrate clogging of the condenser and the vacuum adapter occurs to a major extent...this is easily
remedied with use of a heat gun or hair dryer...otherwise scaling up isn't a problem.