Sciencemadness Discussion Board

Reduction of Methyl 3-Nitrobenzoate to Methyl 3-Amino Benzoate

Archy12345 - 4-3-2016 at 03:38

So over the past week I have been trying to convert methyl 3-nitrobenzoate into methyl 3-aminobenzoate. I have not been very successful, though.

I am taking the iron powder/acetic acid route, and I read a paper not too long ago which described the reduction of nitro groups to amine with Fe/AcOH using sonication instead of reflux. I will attach a link at the end of the post.

I am using a 5mL of a 2/2/1 Ethanol/Acetic Acid/Water mix, with 0.15g of my nitro compound and 0.23g of reduced iron powder (5 equivalents). I put all this in a test tube and expose it to ultrasound for 1h. I occasionally stir the mixture vigorously. My ultra sound bath delivers 40kHz and anywhere from 80W-130W of power. I'm not the owner of the machine, and I'm not entirely sure.

I can see that iron powder is being oxidized because of a brown powder forming in the test tube. So there is SOMETHING happening.

After the ultrasound exposure, I pour the contents of the test tube into a beaker and add about 20mL of aqueous 3M NaOH. This is for the purpose of precipitating out all of the dissolved iron ions and increasing the pH to about 10 or 11. The addition of hydroxide creates something akin to a thin slurry. It is notable to say that when made basic the solution acquired a grape-like smell. The target compound is a regioisomer of methyl anthranilate, so perhaps this is a good sign. I pass this through a frit (which takes forever as you'd expect), then I wash the filtrand with 25mL of ethyl acetate. The filtrate I get is clear and colorless and has two layers. This gets transferred into a separatory funnel and shaken/vented until pressure no longer audible when venting. I then proceed to extract three more times with 25mL of ethyl acetate. The organic layers are combined and dried with anhydrous sodium sulfate. The ethyl acetate is removed using a rotatory evaporator.

This step yields two out comes. One outcome is that there is nothing left once the ethyl acetate has been removed.

The second outcome is that there is a very viscous yellow liquid residue in the flask. This only happened once. I added some DCM to redissolve the yellow liquid and transferred it into a vial. I ran an liquid film IR on my sample, which is attached. The link to an authentic IR is also provided at the bottom of this post.

The IR isn't perfect, but it wasn't awful either. I set up an hydrogen chloride gas generator and gassed the contents of the vial. A very fine solid precipitate did form. I centrifuged out the solid (the solid formed a pellet at the top of the solution, which I thought was interesting, but I suppose it's not hard to be less dense than DCM). the DCM was pipetted into a beaker. I added more DCM to rinse the solid, and this was also centrifuged. I pipetted as much of the DCM as I could, then transferred the contents of the vial into a round bottom flask and used the rotatory evaporator to get rid of all the DCM. At this point I wish I had taken an IR of the solid by KBr pellet. I didn't, but I did weigh it, though. It's mass was roughly 0.07g. Which corresponds to a terrible yield, assuming I had the right compound anyway.

I decided I wanted to try to convert the compound back into its freebase form. The solid readily dissolved in water, which is a good sign. I tried to extract the salt via ethyl acetate, but upon evaporation the flask was empty.

So! Does anyone have any thoughts on any of this? I've run the reaction 5 times so far. I have obtained yield 2/5 times, and only one of those times gave a decent IR. If anyone who has experience with this type of reaction, help is greatly appreciated. I thank you in advance.


Paper about Ultrasound:
http://ro.uow.edu.au/cgi/viewcontent.cgi?article=1174&co...

Authentic IR:
http://www.chemicalbook.com/Spectrum/4518-10-9_IR2.gif

Methyl 3-Aminobenzoate IR.JPG - 1.7MB

Dr.Bob - 4-3-2016 at 08:24

I seem to remember that some HCl helps this reaction. You could also try Zinc powder, granules or dust (even the dust from inside a large box of galvanized nails might be suffucient), tin has also been used. All have their pros and cons. If you can get sodium dithionite, found in some OTC rust removers, that also can reduce nitro groups. The simple way is Pd/C in ethanol and H2, but that is not very OTC, however, many variations exist.

UC235 - 4-3-2016 at 08:45

If you give up on saving the ester, 3-nitrobenzoic acid is an easy reduction using sodium dithionite or thiourea dioxide in aq. NaOH/NH3. The water solubility of the sodium salt will make for a rapid reduction in homogenous solution. See the other new thread about luminol prep. The reduction methods should be quite suitable for both.

Nicodem - 4-3-2016 at 09:23

You can easily reduce methyl m-nitrobenzoate with hydrogen over Pd-C in methanol. The yield is quantitative and the isolation could hardly be easier. At atmospheric pressure you might need some 2 mol% of 5% Pd-C. At >3 bar <0.5 mol% Pd-C will do. I don't remember exactly if I ever tried it on this plain methyl m-nitrobenzoate or maybe some additionally substituted methyl m-nitrobenzoate, but I'm pretty sure it works. Esterification of m-benzoic acid with methanol also gives good yields.

Using ethanol for solvent, sonicating and quenching with 3M NaOH up to pH 11 are not good ideas for your attempts with iron based reductions.

clearly_not_atara - 4-3-2016 at 12:00

Quote:
If you give up on saving the ester, 3-nitrobenzoic acid is an easy reduction using sodium dithionite or thiourea dioxide in aq. NaOH/NH3. The water solubility of the sodium salt will make for a rapid reduction in homogenous solution.


You might be able to make dithionite in situ with sodium bisulfite and zinc in ethylene glycol. Sodium dithionite is usually produced by Na2S2O5 + Zn >> Na2S2O4 + ZnO, but it is extremely flammable and thus difficult to isolate. NaHSO3 has some limited solubility in alcohols; glycol happens to do better than most alcohols at dissolving inorganic salts.

(First, make sure that ethylene glycol will dissolve both your substrate and sodium bisulfite!)

(Yes, I searched for the solubility of Na2S2O5 in ethylene glycol; couldn't find anything definitive)

Edit: this patent seems to agree that bisulfite dissolves in Et(OH)2.

[Edited on 4-3-2016 by clearly_not_atara]

Archy12345 - 4-3-2016 at 14:14

Quote: Originally posted by Nicodem  


Using ethanol for solvent, sonicating and quenching with 3M NaOH up to pH 11 are not good ideas for your attempts with iron based reductions.



If you read the paper I attached a link for in my original post, there has been good success with using ethanol and sonication for this sort of reduction.

I am fairly certain my issue resides in my work up. I can't find any good instructions on how to isolate amines from their reactions mixtures for this kind of reaction. So I'm just going off my own sep schemes. Any ideas for how to better isolate the compound?

Archy12345 - 4-3-2016 at 14:19

Quote: Originally posted by UC235  
If you give up on saving the ester, 3-nitrobenzoic acid is an easy reduction using sodium dithionite or thiourea dioxide in aq. NaOH/NH3. The water solubility of the sodium salt will make for a rapid reduction in homogenous solution. See the other new thread about luminol prep. The reduction methods should be quite suitable for both.


As much as I would like to scrap the idea of keeping the ester around, it would unfortunately defeat the overall purpose of my reaction series.

Nicodem - 4-3-2016 at 15:10

Quote: Originally posted by Archy12345  
If you read the paper I attached a link for in my original post, there has been good success with using ethanol and sonication for this sort of reduction.

I can't see methyl m-nitrobenzoate among the checked substrates. You must have misread something.
I seriously doubt that any serious chemist would use ethanol as a solvent for a methyl ester, except under neutral and mild conditions and where product purity is not an issue.
Using NaOH up to a pH where methyl and ethyl esters rapidly hydrolyze certainly does no good either. Why didn't you use NaHCO3?

Archy12345 - 4-3-2016 at 16:18

Quote: Originally posted by Nicodem  
Quote: Originally posted by Archy12345  
If you read the paper I attached a link for in my original post, there has been good success with using ethanol and sonication for this sort of reduction.

I can't see methyl m-nitrobenzoate among the checked substrates. You must have misread something.
I seriously doubt that any serious chemist would use ethanol as a solvent for a methyl ester, except under neutral and mild conditions and where product purity is not an issue.
Using NaOH up to a pH where methyl and ethyl esters rapidly hydrolyze certainly does no good either. Why didn't you use NaHCO3?



First, please cut me some slack. I am still an undergrad, and this is my first time trying to do a synthesis that I can't find much on. I am here to learn.

Second, you're right. Methyl 3-nitrobenzoate is not among the substrates they checked. I said "this sort of reduction," not "this reduction." Among their substrates they did have an ester compound which gave good reduction yields using their general procedure. I hadn't realized that my compound being a methyl ester made that big of a difference, but now that I think about it I can definitely see how it would make a big difference. This is why I didn't think of using sodium bicarbonate.

Also, just how basic does normally does the aqueous phase have to be to effectively extract an amine out?

Third, is the reason ethanol is a bad solvent in these conditions because of the likelihood of transesterification? Which solvent would you suggest switching it out with?


DraconicAcid - 4-3-2016 at 17:21

Quote: Originally posted by Archy12345  
Also, just how basic does normally does the aqueous phase have to be to effectively extract an amine out?

Third, is the reason ethanol is a bad solvent in these conditions because of the likelihood of transesterification? Which solvent would you suggest switching it out with?


If it's a methyl ester, use methanol.

The pKa of regular ammonium salts tends to be around 9-10, so a bit more basic than that. Aniline and its derivatives tend to be a lot less basic- the pKb of aniline is 9.1, so it could be extracted above a pH of 5. An ester of aminobenzoic acid should be similar.

Archy12345 - 4-3-2016 at 17:41

Quote: Originally posted by DraconicAcid  
Quote: Originally posted by Archy12345  
Also, just how basic does normally does the aqueous phase have to be to effectively extract an amine out?

Third, is the reason ethanol is a bad solvent in these conditions because of the likelihood of transesterification? Which solvent would you suggest switching it out with?


If it's a methyl ester, use methanol.

The pKa of regular ammonium salts tends to be around 9-10, so a bit more basic than that. Aniline and its derivatives tend to be a lot less basic- the pKb of aniline is 9.1, so it could be extracted above a pH of 5. An ester of aminobenzoic acid should be similar.




Well, I guess the methanol answer should have been right in my face.

I completely forgot that it's possible to infer a good pH for extraction based on basicity constants. Thank you for reminding me.
And if I'm thinking through this correctly, the electron withdrawing nature of the ester group should actually even raise the pkb of the amine a little bit, right?

Perhaps if I just bring the solution to neutral pH would get good results. At 7 I am really slowing down hydrolysis kinetics, I have the vast majority of iron out of solution. And I should be able to extract the amine.

Thank you everyone for all of the input! I will take all of these things into consideration tonight when I run this reaction yet again.

chemplayer.. - 4-3-2016 at 17:44

We did a trial of interesting nitro-reduction methods in a video a while back. The one which worked best was sodium dithionite using sodium carbonate as a base (made fresh from sodium bisulfite and zinc powder).

We actually used this process in a follow up to produce methyl 4-aminobenzoate (benzocaine) so it seems to be relatively ester-friendly, but the yields weren't spectacular, so there was probably some degree of hydrolysis occurring.

DraconicAcid - 4-3-2016 at 17:57

I wonder....if the ester was hydrolyzed, would the iron salt be insoluble? I know a lot of anthranilates are, but iron(III) doesn't form a simple anthranilate.

Archy12345 - 4-3-2016 at 18:22

Quote: Originally posted by DraconicAcid  
I wonder....if the ester was hydrolyzed, would the iron salt be insoluble? I know a lot of anthranilates are, but iron(III) doesn't form a simple anthranilate.



I'm not sure. I imagine that most of the iron would be precipitating out as a hydroxide though because the concentration of hydroxide is so much higher than the amine.

The other thing i was just thinking about is I might decide to not add water into my reaction mixture. That just seems like a bad idea with the methyl ester. Perhaps instead of using 2/2/1 Methanol/Acetic Acid/Water I should use 2/3 Methanol Acetic Acid. I guess the presence of water gives the iron ions somewhere to go, but will removing water hinder the reaction? I imagine methanol absorbs enough water out of the air (East Coast) to solvate whatever needs solvating.

[Edited on 5-3-2016 by Archy12345]

chemplayer.. - 4-3-2016 at 21:01

Would be worth trying with copper (II). Adding copper sulfate solution to the filtrate after doing an anthranilic acid synth via Hoffman rearrangement gets you a fair bit of copper anthranilate precipitate. The only trouble then is how to convert to the acid without something like H2S (highly toxic).

DraconicAcid - 16-8-2019 at 15:44

Anyone know if this turned out? I'm working on exactly the same reaction....

Metacelsus - 17-8-2019 at 04:38

When I was making p-aminobenzoic acid, I had much better results using tin granules instead of iron (both with HCl). That might be worth a try.