SplendidAcylation
Hazard to Others
Posts: 196
Registered: 28-10-2018
Location: Starving in some deep mystery
Member Is Offline
Mood: No one I think is in my tree.
|
|
Benzamide from benzoic acid + urea - Problems!
Hi,
Inspired by Smuv's write-up here, and wishing to carry out the preparation of benzonitrile from benzamide by dehydration with ammonium sulphamate, I needed to make some
benzamide.
I had already made a small quantity with great difficulty from ethyl benzoate + ammonia solution (the biphasic mixture had to be ampouled and heated
for ~100h at 100c in order to get a 50% yield), the alternatives being this method, and the obvious one employing benzoyl chloride, which I do not
have.
Quite a few people seem to have done this reaction, with two videos on YouTube:
Tom's Lab, who was inspired by the aforementioned ScienceMadness thread, and who used the same procedure with the reactants in the same proportions,
he achieved a 66% yield:
https://www.youtube.com/watch?v=CFcXE3H5NOk
And Catalysed, who used the same procedure as well, but achieved only 51% yield:
https://www.youtube.com/watch?v=swPHdacfxWc
This seems to be the original procedure that Smuv followed, although he used only 15 mol % of boric acid, rather than the 25 mol % prescribed in
the literature...
Anyway, as it turns out, due to my failure to read the whole thread, I hadn't seen the OrgSyn procedure when I decided to do the reaction, however I
had seen this recent publication from May 2021:
https://www.researchgate.net/publication/351585166_SOLVENT-F...
It's free to download.
Having not seen the OrgSyn procedure, I was confused about why Smuv's procedure differed so much from the one linked directly above.
Because of this, my attempt was a sort of hybrid between the two, and is detailed as follows:
61g (500mmol) of benzoic acid was ground into a fine powder, and mixed with ~92g urea (1.5x mass of acid), 16.54g of boric acid was also added (this
amounts to 0.27x the mass of the acid, not quite the 0.4x specified in the publication above, but 54 mol %, much higher than Smuv's quantitty)
This quantity of boric acid was used simply because it was all I had prepared from borax earlier.
The mixture was mixed with a spoon, and then shaken very thoroughly in a jar for as long as I could before I lost the ability to move my arm.
The powdered mixture was then placed in a 3-necked 500mL RBF (it took up about half of the flask), to which an un-cooled Liebig condenser was
attached, through one of the other necks was inserted a thermocouple connected to a PID controller, and the other neck was stoppered.
The top of the condenser was attached via a length of tubing to an inverted funnel trap immersed in dil. HCl.
The flask was placed in an air-bath (i.e. with Al foil wrapped around the flask and the hotplate), and placed on the hotplate with the PID set to
170c.
No stirring was used.
As the temperature reached around ~120c, some water began to condense on the inside of the flask and in the condenser, with some benzoic acid vapour
escaping the HCl trap, but no ammonia was being evolved at this point (as indicated by the lack of smell when the funnel was removed from the acid).
Quite a bit of steam was coming out as well.
Once the temperature reached around 150c, IIRC, the water and benzoic acid vapour evolution seemed to have slackened, and there was now a strong smell
of ammonia, with crystals (assumed to be ammonium carbamate) clogging the condenser, the plastic tube coming from the condenser, and obscuring the
view into the flask.
The reaction was allowed to progress at 170c for around an hour, but there were a few problems:
I couldn't see inside the flask due to the ammonium carbamate
I hadn't sealed the thermocouple sufficiently into the bung through which it was inserted, and a lot of ammonia gas was escaping into the room,
rendering the funnel-trap useless.
Thus, it was decided to move the entire apparatus under the fume-hood, and disconnect the condenser, allowing the reaction to progress in an open
vessel, as specified in the 2021 publication.
It was at this point that I observed something that, to my knowledge, hasn't been documented here or elsewhere;
There was a considerable quantity of solid material at the bottom of the flask, not powder, but rather a solid mass, above which the liquid layer was
to be found.
I had no idea what the solid was, but evidently it had a melting point greater than 170c, so it couldn't be benzoic acid, urea, or benzamide, and
there was too much of it to be boric acid, so I concluded it was likely to be a condensation product of urea, such as biuret, triuret, cyanuric acid,
or melamine.
The vapour exiting the flask at this point seemed to be less ammoniacal, and there wasn't a lot of ammonium carbamate forming on the neck of the
flask; Rather, a fine white powder was condensing on the neck of the flask, which I thought might be benzamide.
Not wishing to lose my benzamide by evaporation, and considering that the reaction had been ongoing for around 2 hours at this point (the 2021 paper
saying only 10 - 30 minutes was needed), heating was at this point discontinued, and the liquid phase was poured out of the flask into a glass dish.
The liquid took about a minute to solidify, even in a cold dish; It eventually solidified as a slightly waxy off-white solid smelling slightly of
benzonitrile.
The flask was returned to the hot-plate, the temp. was increased to 180c (the thermocouple being inserted into the small quantity of liquid remaining
therein), and this liquid was then poured off into the dish too;
Only a few bits of the solid phase fell out of the flask as well.
The total mass of presumed benzamide (hereafter referred to as benzamide, despite its composition not yet having been verified) solidified in the
glass dish was 59g.
The flask with the solid material was put aside and left to cool.
The benzamide was scraped off of the dish and
A: dissolved in the minimum quantity of boiling water, around 120mL; Only a tiny quantity of undissolved material remained at the
bottom of the slightly yellow solution, but nonetheless the solution was hot-filtered to remove this impurity.
To the filtrate was added ~30g (I don't have the exact figures with me at this moment but it doesn't really matter) of 25% ammonia solution, in order
that the entire solution would be a 5% ammonia solution, this solution was then heated in the microwave until everything was fully dissolved (as some
benzamide had begun to crystallise out after hot-filtering).
The flask was swirled a bit, and then cooled in the fridge to crystallise out the benzamide, which was then vacuum filtered; The process was then
repeated from A above, the only difference being the quantities of water and ammonia solution were slightly decreased.
After this second ammonia extraction (the purpose of which is to remove any unreacted benzoic acid), the vacuum-filtered benzamide crystals were
washed a couple of times with a small quantity of cold water, vacuum-filtered, and dried.
Final yield of odourless, slightly off-white benzamide: 25g (41% yield)
I don't have a melting point apparatus or any capillaries with which to attempt a melting point determination, but the following tests were done:
A small quantity of benzamide added to some aqueous sodium hydroxide in a test tube didn't dissolve immediately in the cold, but the benzamide
dissolved when heated; When the liquid was heated to boiling, the smell of ammonia was observed.
A small quantity of benzamide added to sodium hypochlorite solution resulted in the evolution of a gas which wasn't chlorine, I believe the gas is
CO2, as the Hofmann would be taking place.
No aniline was isolated, however.
A small quantity of benzamide placed in a test tube along with some phosphorus pentoxide, gently heated, was converted to benzonitrile, which
condensed as clear oily droplets on the side of the tube, smelling pleasantly similar to benzaldehyde.
The first ammonia extraction filtrate was acidified to pH 1, precipitating unreacted benzoic acid, which was vacuum-filtered, recrystallized from
boiling water, vacuum-filtered, and dried:
5.1g of benzoic acid was thus recovered.
This would bring the yield up to 45% if we base the yield on unrecovered benzoic acid... But this is still a very low yield.
I wish to try this reaction again, but hopefully improve the yield somewhat, I'm just not sure how to do that.
A few mysteries:
Why, according to the 2021 paper, is good mixing necessary, when the reaction occurs in the liquid, molten, phase?
Perhaps the thorough mixing is necessary as the reaction is supposed to take place so quickly, and 30 minutes wouldn't be long enough to ensure that
the melted reaction mixture was homogenous, thus it might be necessary to ensure it is homogenous prior to heating.
If the mechanism of the reaction involved the formation of a mixed anhydride, then why is carbon dioxide and ammonia evolved, rather than water?
Why did I end up with solids in the reaction flask, when everyone else seems to have had a monophasic liquid reaction mixture at the end of the ~2.5h
period?
Perhaps the larger amount of boric acid catalyst used in my experiment (4x as much as Smuv, Tom's Lab, and Catalysed used) expedited the reaction,
however this doesn't really explain the solids, unless boric acid also happens to catalyse the condensation of urea...
That's all I can think of at the moment, all in all I suppose it's a success, but any suggestions about how to improve the reaction would be greatly
appreciated (especially since the 2021 paper has been published since the first ScienceMadness thread on this topic, so perhaps it will be helpful to
anyone who has tried/wishes to try this reaction but who hasn't seen that paper)
[Edited on 24-3-2022 by SplendidAcylation]
|
|
|
SplendidAcylation
Hazard to Others
Posts: 196
Registered: 28-10-2018
Location: Starving in some deep mystery
Member Is Offline
Mood: No one I think is in my tree.
|
|
Second attempt
Hi,
Since my original post, I have re-attempted this reaction, with great success!
I shall describe the process in less detail this time, with emphasis on the parts that I changed:
This time, 64g (525mmol) of benzoic acid was used, this was mixed very thoroughly with 96g of urea, and 26g of boric acid.
This was the first difference compared with my first attempt, I used the "correct" proportion of boric acid here; 0.4x the mass of acid, as specified
by the aforementioned paper, whereas 0.27x the mass of acid was used in my first attempt.
Simpler apparatus was used this time; the powdered mixture was placed in a 500mL single-necked RBF with a thermocouple immersed in the powder, the top
of the flask was loosely covered with Al foil.
The flask was then heated on a sand-bath with the PID controller set to 170c, again the mixture began to melt at around 100c with evolution of steam,
whereupon the temperature rose further and ammonia began to evolve.
The reaction was left alone for 30 minutes, whereupon heating was discontinued, at this point the mixture in the flask consisted of a solid and a
liquid phase; It was desired to find out whether or not the reaction had completed by this point, as the aforementioned paper stated that 10-30
minutes of heating should be sufficient...
A small quantity of the liquid phase was poured out into a beaker, (~4g), to this was added a quantity of 5% ammonia solution (perhaps 20mL) and it
was heated to boiling to dissolve all the solids.
Upon cooling in the fridge, no crystals precipitated.
When the beaker was placed in the freezer, eventually some crystals did precipitate, most of which re-dissolved upon warming to R.T.
The crystals appeared qualitatively different from benzamide, it is suspected that they were ammonium benzoate, although various sloppily performed
tests to determine this were all inconclusive.
Indeed, when the solution was heated in the microwave to re-dissolve all of the crystals, and then acidified, benzoic acid immediately precipitated.
Due to the lack of benzamide precipitate upon cooling of the solution, I conclude that the liquid decanted into the beaker contained very little
benzamide, and significant quantities of unreacted benzoic acid, so I would tentatively suggest that the reaction was not complete within 30 minutes.
The benzoic acid precipitated from the solution was filtered, dried, and returned to the reaction flask, whereupon heating was continued for another ~
2 hours.
Near the end of the 2 hour period, the temp. was increased to 180c in an attempt to melt the solids in the flask to render the mixture monophasic;
This was unsuccessful.
After the 2 hours had elapsed, heating was discontinued, and the liquid phase (only amounting to 30g this time) was decanted off, the solids in the
flask were then broken up carefully and shaken out of the flask.
The solid material was then ground up (with great effort) using a pestle & mortar; The solidified liquid phase decanted off earlier was scraped up
and added to the powdered solids, (A) whereupon a minimum quantity of 5% ammonia solution was added, the mixture was heated to
boiling, and hot-filtered to remove considerable quantities of "insoluble crap", probably condensation products of urea.
The filtrate crystallized upon cooling in the fridge, the crystals were filtered out, and recrystallized a second time from 5% ammonia solution,
filtered, washed with cold water, and dried.
41g of off-white benzamide was thus obtained!
This corresponds to a 64% yield!
It seems likely that the significant error made in my first attempt was not the lack of boric acid, but the failure to extract the amide from the
solids in the flask, and the naive assumption that the liquid phase would contain all of the benzamide.
Fortunately, I had kept the solids from the original reaction, so the same process from A, above, was carried out using the solids I
had retained; 21g of benzamide, was thus obtained!
This brings the yield of the initial attempt to 76%, which is pretty good!
Of course, I have no proof that my benzamide is pure, it passes various qualitative tests, but no quantitative determination of its purity has been
performed, I suppose we shall find out when it is used for its destined purpose; The dehydration into benzonitrile!
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
