Sciencemadness Discussion Board

P-nitrobenzoic acid reduction pH level question

aromaticfanatic - 2-12-2020 at 12:58

I failed my last go at making benzocaine from toluene (isomer separation failed from lack of vacuum equipment) but I am dabbling with the idea of trying again.

The reduction step utilizes the usual HCl/Sn reduction bath and I don't have any questions about that.

The issue I am having is that at the end, they use concentrated ammonia solution as well as glacial acetic acid to adjust pH. I can definitely make both reagents just fine, however, time is tight and I'd like to not waste it making a reagent that could be substituted.

I wonder if I would be able to play with pH by just using the usual mineral acids and something like sodium hydroxide or carbonate. The margin for error would be much tighter but I don't mind going slow. I can also dilute said solutions.

The problem is that I just want to be sure that the acetic acid and ammonia solution aren't vital for this. I can see no reason but then again chemistry has some hidden traps.

Here is what the paper says to do. I cut out the initial reaction as it's pretty straightforward and standard:

After 30-45 min, when most of the Sn has dissolved, cool and decant the liquid into a 250
ml beaker.

Wash the flask to transfer any remaining organics into the beaker. Discard any remaining Sn metal.

Add conc. NH4OH to the beaker until pH~8 with pH paper. (I needed ~30 ml).

Filter white pasty hydrated tin oxide via 7 cm Buchner/500 ml filter flask. Rinse ppt with water. Discard ppt.

Transfer clear filtrate to 250 ml beaker with stir bar and evaporate to ~40 ml, or until the mixture becomes cloudy on hotplate set at 330C.

Cool beaker contents, add glacial acetic acid to pH 3-4 via pH paper. (I needed 1-2 ml).

Cool beaker contents to RT, then in ice.

Filter yellow product on 7 cm Buchner/500 ml filter flask.

Wash ppt with cold water. Transfer ppt to tared weigh boat to dry. Discard filtrate.


The initial basifying is done with ammonia to precipitate the tin. I think NaOH solution can do just that as well so I don't see a real need for the ammonia.

The acidification should be able to be done with some conc. H2SO4 right? They use glacial acetic acid to keep water content down so I assume 98% H2SO4 will do just that. The only issue I can see is that the H2SO4 is too good at dropping the pH. I would be running it at a slightly larger scale so I have a tiny bit more margin of error but I don't see harm in using some. Or are they using an organic weak acid instead of a strong mineral acid for a reason unbeknownst to me?

[Edited on 2-12-2020 by aromaticfanatic]

UC235 - 2-12-2020 at 20:17

Hydrous tin oxide is soluble in excess NaOH forming sodium stannate.

Acetic acid is used to reduce the pH because it forms a buffer solution whose pH is much more stable and slow to change than if, say HCl was used. p-aminobenzoic acid is a bit tricky to get to crystallize. It is a zwitterion with both + and - charges present due to the functional groups.

In alkaline solution the aniline group is not protonated, but it forms a carboxylate salt which is highly soluble. In an acidic solution, the carboxylic acid group is protonated, but the aniline group is also protonated forming a soluble anilinium salt. The solubility is lowest right in the middle of these two possibilities at the isoelectric point (the average of the two groups' pKas) which is pH 3.7. Yield will be lower at any other pH.

You can probably get away with using other acids that buffer effectively at around pH 3.7. Citric acid is the only readily available one that comes to mind.


[Edited on 3-12-2020 by UC235]

aromaticfanatic - 2-12-2020 at 20:29

Quote: Originally posted by UC235  
Hydrous tin oxide is soluble in excess NaOH forming sodium stannate.

Acetic acid is used to reduce the pH because it forms a buffer solution whose pH is much more stable and slow to change than if, say HCl was used. p-aminobenzoic acid is a bit tricky to get to crystallize. It is a zwitterion with both + and - charges present due to the functional groups.

In alkaline solution the aniline group is not protonated, but it forms a carboxylate salt which is highly soluble. In an acidic solution, the carboxylic acid group is protonated, but the aniline group is also protonated forming a soluble anilinium salt. The solubility is lowest right in the middle of these two possibilities at the isoelectric point (the average of the two groups' pKas) which is pH 3.7. Yield will be lower at any other pH.


Thank you! This is very informative. I had known about the zwitterion but I did not know it was the reason for tricky crystallization and need for good pH control. That's very interesting actually.

I suppose I'll go through the trouble of making the ammonia and glacial acetic acid. Would you happen to know if the ammonia solution needs to be concentrated? It seems they only use concentrated as to reduce the amount of water to boil off but I could be wrong.

The glacial acetic acid is straightforward to make so I'm not worried about that, I'll just go ahead and make it.

Thanks again, I learned a lot.

UC235 - 2-12-2020 at 20:44

I'm sure you can use more dilute ammonia for the initial neutralization. Another benefit to using ammonia is that NH4Cl is extremely soluble vs NaCl which would probably crash out when you boiled the solution down. A saturated solution of citric acid which is almost 60 w/w% would probably be a good OTC stand-in for the glacial acetic acid. It buffers effectively in the same range.

aromaticfanatic - 2-12-2020 at 21:31

Quote: Originally posted by UC235  
I'm sure you can use more dilute ammonia for the initial neutralization. Another benefit to using ammonia is that NH4Cl is extremely soluble vs NaCl which would probably crash out when you boiled the solution down. A saturated solution of citric acid which is almost 60 w/w% would probably be a good OTC stand-in for the glacial acetic acid. It buffers effectively in the same range.


Thank you. I was actually talking to my friends about this and citric acid buffer came up. I was about to ask you what you thought of that. The buffer can range down to 3 and I can do some calculations to try and get to the 3.7 mark.

I've decided to split the product solution into two batches. One will try ammonia or possibly carbonate for the base to precipitate the tin and then subsequent citric acid acidification.

The other solution will be used as a back up in the event of failure. I'm making all of the precursors from scratch and they aren't exactly quick reactions. Time is tight so I'd rather play it safe.

I'll make some ammonia solution hopefully soon and then I'll try out the citric acid solution. I have a large amount leftover from ages ago so I'm sure glad to get rid of some.

Thanks again for the help. I really appreciate it.

Opylation - 3-12-2020 at 00:16

It may be easier to get PABA online. I have purchased it from vitamin retailers to use for aniline production. It is used by people with vitamin deficiencies.

aromaticfanatic - 3-12-2020 at 15:08

Quote: Originally posted by Opylation  
It may be easier to get PABA online. I have purchased it from vitamin retailers to use for aniline production. It is used by people with vitamin deficiencies.


Yea it certainly is but the point of the project is to make benzocaine from toluene which was extracted from carb cleaner. I got to the nitrotoluene step last time but I failed isomer separation from lack of equipment. I might try again and I've got a better plan. I mean I got some separation but the yield of para was probably less than half a gram from a theoretical of at least 10 grams.

DraconicAcid - 3-12-2020 at 21:41

Actually, I'm following this, because I've been trying to make some m-aminobenzoic acid, aka benzamic acid.

aromaticfanatic - 4-12-2020 at 08:48

Quote: Originally posted by DraconicAcid  
Actually, I'm following this, because I've been trying to make some m-aminobenzoic acid, aka benzamic acid.


Nice! In the case you are starting from toluene, oxidize it into benzoic acid first because I think the carboxylic acid group is meta directing in the nitration. I have to nitrate toluene first before oxidizing it for that very reason.

Good luck!

Fery - 4-12-2020 at 13:51

This is just my idea, I will certainly try it sometimes in the future...
When obtaining mixture of 2- and 4-nitrotoluene by nitration of toluene, it is necessary to freeze out substantial part of the 4-isomer at T -10 C when 4-nitrotoluene solid (m.p. 51,6 C) but 2-nitrotoluene still liquid (m.p. -10,4 C) which is not too much difficult, almost everyone has a freezer with -18 C, and then vacuum distill the remaining mixture using powerful column (Hempel packed with Rashig rings) and distillation head for taking off reflux at some ratio like 1:10... I have all the necessary glass and equipment... no problemo for me, but thinking about easier way suitable for everyone... the final step is oxidizing -CH3 to -COOH, maybe oxidizing both isomers together (without necessity of separation of 2 and 4 nitrotoluenes) and only then separate acids by steam distillation? I expect 2-nitrobenzoic acid could be more volatile by steam than 4-nitrobenzoic acid due to intramolecular hydrogen bond (I separated that way 2-nitrophenol from 4-nitrophenol). Could someone skilled correct me if I'm wrong? A real experiment will definitely tell the truth but I do not yet know when I perform it.

DraconicAcid - 4-12-2020 at 13:57

Quote: Originally posted by aromaticfanatic  
Quote: Originally posted by DraconicAcid  
Actually, I'm following this, because I've been trying to make some m-aminobenzoic acid, aka benzamic acid.


Nice! In the case you are starting from toluene, oxidize it into benzoic acid first because I think the carboxylic acid group is meta directing in the nitration. I have to nitrate toluene first before oxidizing it for that very reason.

Good luck!

Actually, I'm starting from methyl benzoate. The nitration works quite well, but the reduction was giving me some problems.

Fery - 4-12-2020 at 13:59

https://sci-hub.st/10.1002/14356007.a03_555
4-Nitrobenzoic Acid is produced commercially by the oxidation of 4-nitrotoluene.
An interesting method involves the nitration and subsequent oxidation of polystyrene [28].
This method uses the steric hindrance of the polymer chain to improve the para to ortho ratio of the product.
[28] C. S. Rondestvedt, Jr., J. R. Jeffrey, J. E. Miller, Ind. Eng. Chem. Prod. Res. Dev. 16 (1977) no. 4, 309.

DraconicAcid - 4-12-2020 at 17:27

Quote: Originally posted by Fery  
https://sci-hub.st/10.1002/14356007.a03_555
4-Nitrobenzoic Acid is produced commercially by the oxidation of 4-nitrotoluene.
An interesting method involves the nitration and subsequent oxidation of polystyrene [28].
This method uses the steric hindrance of the polymer chain to improve the para to ortho ratio of the product.
[28] C. S. Rondestvedt, Jr., J. R. Jeffrey, J. E. Miller, Ind. Eng. Chem. Prod. Res. Dev. 16 (1977) no. 4, 309.


Anyone got a copy of this paper?

aromaticfanatic - 7-12-2020 at 15:34

Quote: Originally posted by Fery  
This is just my idea, I will certainly try it sometimes in the future...
When obtaining mixture of 2- and 4-nitrotoluene by nitration of toluene, it is necessary to freeze out substantial part of the 4-isomer at T -10 C when 4-nitrotoluene solid (m.p. 51,6 C) but 2-nitrotoluene still liquid (m.p. -10,4 C) which is not too much difficult, almost everyone has a freezer with -18 C, and then vacuum distill the remaining mixture using powerful column (Hempel packed with Rashig rings) and distillation head for taking off reflux at some ratio like 1:10... I have all the necessary glass and equipment... no problemo for me, but thinking about easier way suitable for everyone... the final step is oxidizing -CH3 to -COOH, maybe oxidizing both isomers together (without necessity of separation of 2 and 4 nitrotoluenes) and only then separate acids by steam distillation? I expect 2-nitrobenzoic acid could be more volatile by steam than 4-nitrobenzoic acid due to intramolecular hydrogen bond (I separated that way 2-nitrophenol from 4-nitrophenol). Could someone skilled correct me if I'm wrong? A real experiment will definitely tell the truth but I do not yet know when I perform it.


Interesting. I later found out that I could've just used all of the isomers throughout the entire process and then since the ethyl-anthranilate was liquid at room temp and the benzocaine was not, I could've likely tried freezing the benzocaine out of the isomer mix.

However, I am trying a different route now. I couldn't oxidize the nitrotoluenes in the beginning because that would produce the meta nitrobenzoic acid as the major product. I found out that the structural isomer of benzocaine, ethyl-3-aminobenzoate, also has anesthetic effects.

I'm going to go ahead and make the ethyl-3-aminobenzoate. It should be higher yielding too since I think the meta product forms at a much higher percentage than if I had nitrated toluene with the mixed acids and separated the para product.

I've got it all planned out and once I stop being a lazy fatty I'll clean up the glassware and get started. Oh shoot and I need new keck clips.... I melted the last ones I had left. Remember kids, plastic won't withstand 200C distillations! :)

Fery - 8-12-2020 at 06:05

ethyl-3-aminobenzoate:
yes, tricaine, very poisonous to fish, but fish killed by it are allowed for human consumption

aromaticfanatic - 8-12-2020 at 18:32

Quote: Originally posted by Fery  
ethyl-3-aminobenzoate:
yes, tricaine, very poisonous to fish, but fish killed by it are allowed for human consumption


Correct! I think it is the only anesthetic that is approved for fish that are meant to be consumed by humans.