Sciencemadness Discussion Board

3,4,5-trimethoxyphenylacetonitrile hydrogenation problems

qt314 - 25-1-2019 at 05:22

Hello. I havent posted here in a *very very* long time but find myself in need of help ...

I have been trying to reduce 3,4,5-trimethoxyphenylacetonitrile to the amine. I have employed several methods and encountered problems throughout.
At first I followed the general protocol with CoCl2/NaBH4 (0.1eq/2eq) with thf/water
Results were acceptable at the beginning, roughly 50% yield.
I then moved to IPA/water (I try to avoid thf when I can). Just replacing the thf with IPA gave comparable results.
During optimization studies I reached 90% yield twice by employing reverse addition (borohydride in ipa/water, nitrile added slowly). Apart from these 2 experiments, all times yield was consistently 20-30%.
So, I moved to (20% w/w with nitrile) Pd/c and hydrogen gas under balloon pressure with 1-2 eq of mineral acid.
Best yield so far is 35% after 8 hours. 48 hours gave 10%.

I am at a loss. The nitrile is clean (98.28% assay, nmr is perfect), solvents are clean, palladium is of good quality ....

Can anyone advise or give tips?

myr - 25-1-2019 at 06:55

Tetrabutylammonium borohydride will work (but doing so is 100% illegal.)

qt314 - 25-1-2019 at 07:06

TAB witg CoCl2 should work though I haven't tried yet. Or do you claim it works without the cobalt?

qt314 - 25-1-2019 at 07:13

Oh, forgot to mention - I tried cth with ammonium formate and pd/c (6eq, 20% w/w respectively) for 12 hours = ~30% yield

unionised - 25-1-2019 at 10:14

Just a thought; O2 goes through rubber quite well

qt314 - 26-1-2019 at 23:01

Quote: Originally posted by unionised  
Just a thought; O2 goes through rubber quite well


So, you think that a long time with a balloon may let O2 in and inhibit reaction?
I have been using double layered balloons quite well for hydrogenation of other compounds and it works quite well

clearly_not_atara - 27-1-2019 at 00:00

What kind of mineral acid are you using? The studies I see use sodium dihydrogen phosphate. I get the feeling an excessively strong acid will hydrolyse the nitrile.
https://www.sciencedirect.com/science/article/abs/pii/S09268...
Quote:
During optimization studies I reached 90% yield twice

This was not reproducible?

qt314 - 27-1-2019 at 02:03

Quote: Originally posted by clearly_not_atara  
What kind of mineral acid are you using? The studies I see use sodium dihydrogen phosphate. I get the feeling an excessively strong acid will hydrolyse the nitrile.
https://www.sciencedirect.com/science/article/abs/pii/S09268...
Quote:
During optimization studies I reached 90% yield twice

This was not reproducible?


I am using HCl (37%, AR). Some hydrolysis does take place. I do the reaction in an ice/salt bath - the rate of hydrolysis is not that big. Using sodium dihydrogen phosphate does make sense. I will give it a try tomorrow.

I today noticed that the solubility of the nitrile drops appreciatly when the acid is added, which may account for my low yields. I am testing this hypothesis now.

Regarding the CoCl2/NaBH4 - I dont know what to say. I am not able to reproduce these results no matter what I try. It is very weird. TLC shows either of 2 results -
Case A - ~50% of nitrile reacted, ~30% primary amine product, the rest being secondary and tertiary amines.
Case B - ~75% of nitrile reacted, major product is unknown, runs very very close to the nitrile. ~25% primary amine yield.
These results repeat no matter what variation I take (more borohydride, different solvents, different temps, different concentrations, different reagents).
The 2 times 90% primary amine yield was achieved tlc showed a very clean reaction, mostly primary amine, some leftover nitrile, some tertiary amine. For the life of me, i do not know what to make of this.

S.C. Wack - 27-1-2019 at 16:30

Quote: Originally posted by qt314  
During optimization studies I reached 90% yield twice by employing reverse addition (borohydride in ipa/water, nitrile added slowly). Apart from these 2 experiments, all times yield was consistently 20-30%.


Doesn't it sound like a lower concentration of nitrile and intermediates and/or lower temperature is helping? Dimerization is expected to be a problem even more so with phenylacetonitriles, which may explain the unpopularity of this route to PEAs (by amateurs). PS instead of increasing the borohydride it would not be uninteresting to increase the Co from those low levels and see if perhaps a larger amount of the boride is a good thing.

[Edited on 28-1-2019 by S.C. Wack]

qt314 - 28-1-2019 at 01:55

Quote: Originally posted by S.C. Wack  
Quote: Originally posted by qt314  
During optimization studies I reached 90% yield twice by employing reverse addition (borohydride in ipa/water, nitrile added slowly). Apart from these 2 experiments, all times yield was consistently 20-30%.


Doesn't it sound like a lower concentration of nitrile and intermediates and/or lower temperature is helping? Dimerization is expected to be a problem even more so with phenylacetonitriles, which may explain the unpopularity of this route to PEAs (by amateurs). PS instead of increasing the borohydride it would not be uninteresting to increase the Co from those low levels and see if perhaps a larger amount of the boride is a good thing.

[Edited on 28-1-2019 by S.C. Wack]


Lowering the nitrile concentration has crossed my mind. I will try it.

Increasing the Co eq does indeed improve things considerably but it becomes uneconomic... BTW, Co boride can be supported on carbon easily and gives for me comparable yields to unsupportee Co (this simplifies workup considerably).

Low temperatures (as in, lower than 10C) will halt reduction and only allow for hydrogen generation through catalysed borohydride hydrolysis. Too high temperatures increase the rate of borohydride hydrolysis too much also (above 55C ime).

While dimerization does pose a real problem, in theory hydrogenation of nitrile is straightforward...

I was thinking of trying carbon supported Co boride under hydrogen atmosphere with ammonia according to Barnett (https://doi.org/10.1021/i360030a009) but I insist to avoid pressurised hydrogen reaction so it remains to be seen if balloon hydrogen will be sufficient. Ill post results

qt314 - 28-1-2019 at 10:37

--update--

I followed the progress of reaction on tlc (MeOH/Conc. Ammonia, 100:1) every hour instead of checking after 12 hours on all my experiments due to error in my logic regarding the time required for completion.
After about 5 hours ~10% starting nitrile remains, only primary amine product is formed. 1 hour later secondary amine product begins to accumulate and most of nitrile is gone.
Workup gave 70% molar yield of primary amine hcl along with ~20% of secondary amine.
Reaction conditions were :
300ml MeOH, 7.5g 37% HCl, 6g pd/c 10% (60% water content), fitted with hydrogen balloon. Reaction was conducted in an ice bath on a magnetic stirrer.

Can still optimize a bit, but im pleased with the results :)