Sciencemadness Discussion Board

Anthranilic acid fail

dome13 - 29-6-2019 at 12:26

Hello guys! I have tried to synthetise anthranilic acid two times in a week without any succes. I am really confused, because I think it is an easy synthesis. When I first tried it, I followed this method: https://labmonk.com/synthesis-of-anthranilic-acid-from-phtha...
The second was from erowid: Anthranilic Acid (Method 2) [4]

Prepare a solution of 30 g. of sodium hydroxide in 120 ml. of water in a 360 ml. conical flask and cool to 0° or below in a bath of ice and salt. Add 26.2g (8.4 ml.) of bromine in one portion and shake (or stir) until all the bromine has reacted. The temperature will rise somewhat; cool again to 0° or below. Meanwhile, prepare a solution of 22g. of sodium hydroxide in 80 ml. of water. Add 24 g. of finely- powdered phthalimide in one portion to the cold sodium hypobromite solution; stir vigorously while swirling the contents of the flask and add the prepared sodium hydroxide solution rapidly. The solid will dissolve and the temperature will rise to about 70°. Warm the mixture to 80° for about 2 minutes. Filter, if necessary. Cool in ice and add concentrated hydrochloric acid slowly and with stirring until the solution is just neutral (about 60 ml. are required). [It is recommended that a little of the alkaline solution be set aside in case too much acid is added.] Precipitate the anthranilic acid completely by the gradual addition of glacial acetic acid (20-25 ml) are required): it is advisable to transfer the mixture to a 1 litre beaker as some foaming occurs. Filter off the acid at the pump and wash with a little cold water. Recrystallise from hot water with the addition of a little decolourising carbon; collect the acid on a Buchner funnel and dry at 100°. The yield of pure anthranilic acid, m.p. 145°, is 14 g.

The experiment went right, until I added the acid. First time, I added only glacial acetic acid and it released a lot CO2. When I added more, there was no more CO2 evolution, the pH was around 5-6.
On the second run I started adding conc. HCL droppwise, while white smoke produced, thats OK. When I felt to close to equivalent point (but it was still alkaline) I added glacial acetic acid. There was again a lot of CO2 and bubbles. But like first time, I added more and nothing precipitated.

Honestly I have only one idea: add the acedic acid after the pH stabilized with HCL to under 6. Guys, do you have any suggestion? (sorry if my english was bad)

Schleimsäure - 29-6-2019 at 14:56

Here is almost the same, probably the original route:

Alternatively, 40 g of sodium hydroxide are dissolved in 200 ml of water and to this solution 32 g of bromine are added under cooling. The obtained NaBrO is cooled to -10° C and with proper stirring a mixture of 29.6 g phthalimide and 20 mL of cold water are introduced. After the addition of phthalimide the reaction mixture is warmed to 10° C, the solution becomes completely transparent. Then 24 g of finely divided sodium hydroxide are added, the temperature reaches 40 ° C and the reaction mixture is heated to 80° C. Then the cooled reaction mixture is treated with 64 ml of conc. hydrochloric acid and 30 ml of glacial acetic acid and left to stand overnight. The precipitated anthranilic acid is filtered, washed with a small amount of ice water and crystallized from boiling water in the presence of active carbon.

Л.М. Кульберг, Синтезы органических реактивов для неорганического анализа, 16, (1947) (LM Koulberg, Synthesis of organic reagents for inorganic analysis, 16, (1947))

Note, that a lot more bromine is used. Also left to stand overbight after adding the acids.

Schleimsäure - 29-6-2019 at 15:06

I wonder if using KBrO3 directly is also a possibility. Well, in situ NaBrO3 is more cost effective, if you have lots of bromine.

Metacelsus - 29-6-2019 at 22:31

Why not NaOCl? I've made anthranilic acid using it. Yield isn't great (around 40%) but it's generally reliable.

dome13 - 30-6-2019 at 09:57

Update: yesterday I let it stand. Today (after about 14 hours later) I checked my flask and I noticed some precipitate. I vacuum filtered and it gives 50% yield for crude product (it is as brown as cacao). After recrystallization with active carbon it has got still a slightly brown colour. All in all, it seems that I must have let it stand after acidification.
Thanks for your replies! :)

Pumukli - 30-6-2019 at 11:58

Did it stand in the fridge or at room temperature?

Chemi Pharma - 30-6-2019 at 15:52

You can try another route, starting with salicylic acid and converting it to anthranilic acid in one pot with chloroacetamide, K2CO3, KI and DMF. Look at here: http://www.sciencemadness.org/talk/viewthread.php?tid=82095&...

dome13 - 2-7-2019 at 10:09

Pumikli: it standed in room temperature.

Chemi Pharma: thanks for your information :)

chemplayer... - 2-7-2019 at 23:06

First of all have a look at how we do it using bleach: https://www.bitchute.com/video/lt07v5I2lSEp/

It did take quite a few goes to get this right and we'd tried with bromine and got sub-standard results (as well as the hassle of using bromine), but this is now quite a consistent and repeatable process, and works with readily available 5.25% bleach.

Remember that anthranilic acid will dissolve in either acid or alkaline solution (as it's an amino-acid and create soluble salts with either). So it's important to add the HCl and acetic acid carefully with the right pH control.

Boffis - 4-7-2019 at 08:55

@Schleimsaure, do you have an electronic copy of this book in German or English?
Quote: Originally posted by Schleimsäure  
Л.М. Кульберг, Синтезы органических реактивов для неорганического анализа, 16, (1947) (LM Koulberg, Synthesis of organic reagents for inorganic analysis, 16, (1947))


I did a search on various library sites such as libgen etc but couldn't find a copy.

S.C. Wack - 4-7-2019 at 12:04

Chemistry manuals translated from Russian are a lot rarer than those to Russian. The entry appears to be an uncredited translation lifted from the 1927 edition of Henle's Anleitung für das organisch-chemische Praktikum; at 2x scale, according to Weygand. I can only wonder how the translation from Russian to posting at prepchem came about.

Schleimsäure - 5-7-2019 at 10:00

Quote: Originally posted by Boffis  
@Schleimsaure, do you have an electronic copy of this book in German or English?

Quote: Originally posted by Schleimsäure  
Л.М. Кульберг, Синтезы органических реактивов для неорганического анализа, 16, (1947) (LM Koulberg, Synthesis of organic reagents for inorganic analysis, 16, (1947))


I did a search on various library sites such as libgen etc but couldn't find a copy.


As S.C. Wack correctly noted, this is from prepchem.

@S.C. Wack Interesting. Seems like some Soviet chemists copied the German approach from 1927 shortly after the war. Without credits I don't know, since I haven't seen the original paper, or book that is.
And some guy is in possession of that book who posted some of it on prepchem. There are more synthesis cited from that book on prepchem. Probably most of the synthesis in that book is more or less copied, probably without credits, that's why you won't find it in chem libraries. Just guessing though.


[Edited on 5-7-2019 by Schleimsäure]

S.C. Wack - 5-7-2019 at 12:47

Ironically the English translation of Weygand's rendition of Henle is itself stolen (legally by the US govt)(as if being killed by the US army wasn't enough). The Russian book can be found at libgen; everything except the 1927 book is somewhere...it was scanned by google; hathi has it, isn't going to let anyone see it. Weygand 1938/45:

...wofür die Vorschrift von Hoogewerff und van Dorp [Recueil Trav. chim. Pays-Bas 10, 6 (1891)] zur Darstellung der Anthranilsäure aus Phthalimid nach Henle (Organ. Chem. Praktikum, Leipzig 1927, S. 91) als Beispiel dienen mag:

Man löst 20 g Natriumhydroxyd (0,5 Mol) in 100 ccm Wasser, gibt zu der erkalteten Natronlauge 16 g Brom (0,1 Mol), kühlt auf — 10° ab und trägt unter mechanischem Rühren 14,8 g Phthalimid (0,1 Mol) ein, das mit 20 ccm Wasser angerührt ist; mit 10 ccm Wasser spült man die Suspension nach. Dann entfernt man die Kältemischung und läßt unter dauerndem Rühren aufwärmen, bei höchstens 10° soll alles klar gelöst sein. Trägt man nun 12 g gepulvertes Natriumhydroxyd (0,3 Mol) ein, so steigt die Temperatur auf etwa 40°; man steigert sie auf 80°, läßt wieder erkalten, stumpft mit 32 ccm konzentrierter Salzsäure (1,18, etwa 0,4 Mol) ab und säuert mit 15 ccm Eisessig an. Nach längerem Stehen in der Kälte erhält man beim Abfiltrieren, Waschen mit wenig Wasser und Trocknen über Schwefelsäure 12 g Anthranilsäure vom F. 144—145°, entsprechend 87% der Theorie. Das Rohprodukt kann durch Umkristallisieren aus Wasser oder auch aus Ligroin weiter gereinigt werden. Die in der Mutterlauge enthaltene Anthranilsäure kann mit Kupferacetatlösung als anthranilsaures Kupfer gefällt werden.

To a cold solution of 20 g. of sodium hydroxide (0.5 mole) in 100 cc. of water, 16 g. of bromine (0.1 mole) are added. The solution is cooled to -10°C. and a suspension of 14.8 g. of phthalimide (0.1 mole) in 20 cc. of water is added with mechanical stirring. If necessary the last of the suspension is washed into the flask with 10 cc. of water. The freezing mixture is then removed and the mixture allowed to warm up slowly while the stirring is continued. At 10°C. a clear solution should result. Upon addition of 12 g. of pulverized sodium hydroxide (0.3 mole) the temperature rises to about 40°C. It is then increased to 80°C. The solution is then cooled, neutralized with 32 cc. of concentrated hydrochloric acid (D = 1.18), and finally acidified with 15 cc. of glacial acetic acid. After standing for several hours in the cold the anthranilic acid is collected on a filter, washed with a little water, and dried over sulfuric acid. A yield of 12 g. of anthranilic acid, 87% of the theoretical amount, melting at 144° to 145°C., is obtained. The crude product can be purified by recrystallization from water or from ligroin. The anthranilic acid retained in the mother liquor can be precipitated with cupric acetate solution as copper anthranilate.

Schleimsäure - 5-7-2019 at 16:25

Quote: Originally posted by S.C. Wack  
Ironically the English translation of Weygand's rendition of Henle is itself stolen (legally by the US govt)(as if being killed by the US army wasn't enough). The Russian book can be found at libgen; everything except the 1927 book is somewhere...it was scanned by google; hathi has it, isn't going to let anyone see it. Weygand 1938/45:

...wofür die Vorschrift von Hoogewerff und van Dorp [Recueil Trav. chim. Pays-Bas 10, 6 (1891)] zur Darstellung der Anthranilsäure aus Phthalimid nach Henle (Organ. Chem. Praktikum, Leipzig 1927, S. 91) als Beispiel dienen mag:

Man löst 20 g Natriumhydroxyd (0,5 Mol) in 100 ccm Wasser, gibt zu der erkalteten Natronlauge 16 g Brom (0,1 Mol), kühlt auf — 10° ab und trägt unter mechanischem Rühren 14,8 g Phthalimid (0,1 Mol) ein, das mit 20 ccm Wasser angerührt ist; mit 10 ccm Wasser spült man die Suspension nach. Dann entfernt man die Kältemischung und läßt unter dauerndem Rühren aufwärmen, bei höchstens 10° soll alles klar gelöst sein. Trägt man nun 12 g gepulvertes Natriumhydroxyd (0,3 Mol) ein, so steigt die Temperatur auf etwa 40°; man steigert sie auf 80°, läßt wieder erkalten, stumpft mit 32 ccm konzentrierter Salzsäure (1,18, etwa 0,4 Mol) ab und säuert mit 15 ccm Eisessig an. Nach längerem Stehen in der Kälte erhält man beim Abfiltrieren, Waschen mit wenig Wasser und Trocknen über Schwefelsäure 12 g Anthranilsäure vom F. 144—145°, entsprechend 87% der Theorie. Das Rohprodukt kann durch Umkristallisieren aus Wasser oder auch aus Ligroin weiter gereinigt werden. Die in der Mutterlauge enthaltene Anthranilsäure kann mit Kupferacetatlösung als anthranilsaures Kupfer gefällt werden.

To a cold solution of 20 g. of sodium hydroxide (0.5 mole) in 100 cc. of water, 16 g. of bromine (0.1 mole) are added. The solution is cooled to -10°C. and a suspension of 14.8 g. of phthalimide (0.1 mole) in 20 cc. of water is added with mechanical stirring. If necessary the last of the suspension is washed into the flask with 10 cc. of water. The freezing mixture is then removed and the mixture allowed to warm up slowly while the stirring is continued. At 10°C. a clear solution should result. Upon addition of 12 g. of pulverized sodium hydroxide (0.3 mole) the temperature rises to about 40°C. It is then increased to 80°C. The solution is then cooled, neutralized with 32 cc. of concentrated hydrochloric acid (D = 1.18), and finally acidified with 15 cc. of glacial acetic acid. After standing for several hours in the cold the anthranilic acid is collected on a filter, washed with a little water, and dried over sulfuric acid. A yield of 12 g. of anthranilic acid, 87% of the theoretical amount, melting at 144° to 145°C., is obtained. The crude product can be purified by recrystallization from water or from ligroin. The anthranilic acid retained in the mother liquor can be precipitated with cupric acetate solution as copper anthranilate.


You wouldn't have to translate the synthesis for me, since I'm native German. Just kidding. Very detailed and 87% of theory is very good.
From the looks of it, it seems like Henle did an improvement of the original 1891 synthesis of anthanilic acid by the Dutch chemists in 1927.
And everything afterwards was just "stolen".

Pumukli - 5-7-2019 at 21:42

Nice find and translation, Mr. Wack!

The yield is especially remarkable, taking into account that the best we (sciencemadness members) could achieve was some 45% reported as I remember. (Byko3y, 2015.)
Relevant thread: http://www.sciencemadness.org/talk/viewthread.php?tid=10272#...

Btw it seems that Weygand in this synth did not hydrolised phthalimide to phthalamic acid at first unlike as byko3y suggested. Instead he made N-bromo-phthalimide at first (-10 C to 10 C), then hydrolised it to N-bromo-phthalamic acid (up to 40 C or so), then rearranged this haloamine as "usual" (heating up to 80 C). At least this is how I interpret the process. :)

If my understanding is correct then this method may be superior to byko3y's suggestion. He reported he could smell ammonia. This is only possible if phthalimide is "overhydrolised" to phthalic acid and ammonia instead of phthalamic acid. (Or the used phthalimide was contaminated with an inorganic ammonium salt.)

Moreover if my thinking is correct then one may omit the fiddling with hypobromite/hypochlorite altogether. Instead of these questionable quality reagents it may be possible to use TCCA to chlorinate the phthalimide first! TCCA reportedly could chlorinate e.g. benzamide in acetone/chloroform with 97% yield! (De Luca, L.; Giacomelli, G.; Nieddu, G.; Synlett, 2005, 223.)
Other approach might be to use calcium hypochlorite. It may be more stable than the liquids and surely more concentrated source of hypochlorite. It works great in the oxidation of secondary alcohols. Maybe a solventless "green" reaction in a mortar with a pestle and a few drops of water, grinding for 15 mins the paste would be possible...


I think it's time to start a scimad challenge: let's improve the synthesis of anthranilic acid from phthalimide! The first who reports 80%+ yield wins the challenge. ;)

Metacelsus - 5-7-2019 at 22:37

Quote: Originally posted by Pumukli  
I think it's time to start a scimad challenge: let's improve the synthesis of anthranilic acid from phthalimide! The first who reports 80%+ yield wins the challenge. ;)


I think the yield (and purity of product) should probably also be verified by another member. That should make sure the procedure is replicable. This is what OrgSyn does for their published methods.

Of course, the member who replicates it should also get some of the credit.

I think the initial N-chlorination or N-bromination step is a good idea. But isn't that supposedly what already happens the the anthranilic acid synthesis most of us have done already? The first step is a cold NaClO addition. I would focus more on optimizing workup instead, since it seems that's where most of the problems are with low recovery of product.

Pumukli - 5-7-2019 at 23:39

The actual mechanism is in question. (At least not clear 100% for me.)

Byko3y emphasized the importance of the initial hydrolysis, he said that the mechanism was something like this:

phthalimide -> phthalamic acid -> N-halo-phthalamic acid ->>> product

He explicitely starts with the addition of hydroxyde and only later introduces the hypochlorite.

Another possibility is (Weygand?):

phthalimide -> N-halo-phthalimide -> N-halo-phthalamic acid ->>> product.

These different starting steps may seem irrelevant but there's a published method (actually I heard of two) for phthalamic acid. One starts from phthalimide, the other one from phthalic anhydride (or acid).

Someone should try to start from phthalamic acid, because the mentioned overhydrolysis of phthalimide surely decreases the yield. Question is how severe this loss is.

Someone should also try to first halogenate phthalimide and then hydrolyse it to N-halo-phthalamic acid. Unfortunately it may well be possible that N-halo-phthalimide lose some percent halogene during the hydrolysis process and it would mean losses of product too. Question is how severe the N-dehalogenation during the hydrolysis.

Question is which of the above mentioned two losses is the more severe one. :)

And of course, the workup is also very important. Who cares if one starts from this or that and looses a few percent during the first two steps when half of the otherwise successfully prepared acid remains in solution and goes down the drain -literally- in the end?! :)

In the Weygand article I don't see anything fancy regarding workup though. And still, the author states 87% crude, dried product with 144-145 C melting point. (Hah, that would be fine for me!) Or was he infected with the "anthranilic acid bug" too? (= copy/paste work of others, blindly copy the yield too, never actually doing the synthesis.)

The product identification, replication of the method is also important. But at first someone should actually do the work and see how the different routes work (or not). Then he can still screw up the workup and come back with more questions than solved problems. :D
I have some starting materials but lack free laboratory time in these days so I'm mostly just an armchair hobbyst right now. ;)

Pumukli - 8-7-2019 at 04:37

Well, now I'm over three different batches of anthranilic acid preparations.
But I feel dumber than before! :o

In a nutshell I tried the following:

1. phthalimide + calcium hypochlorite powder (you know, the OTC disinfectant), both used as powders

2. phthalimide powder + calcium hypochlorite filtered solution

3. phthalimide dissolved in NaOH solution + calcium hypochlorite filtered solution

The hypochlorite content was first determined by thiosulphate titration (iodometry), the powder form and the solution too.
I used almost molar equivalents of the reagents, the imide was always in a slight (say 2%) excess. First run was 100 mmol scale, the later ones 10 mmols.

First run, hypochlorite powder (finely ground in a mortar) was added to water, mixed for a while, suspension formed, added to this suspension the mortar powdered phthalimide. Temp rose to about 40 Celsius, mixture became ugly blackish brown, like very dark cocoa powder mixed in water. Filtered the cooled solution, before the acidification, yield a lot of (cc. 50% by weight of the starting materials) ugly, MnO2-like fine powder, was a nightmare to filter. (As if it was real MnO2!) The filtrate was a clear, very dark maroon solution. Upon acidification it bubbled and slowly deposited some crystals. Then in the fridge it gave another crop. Then upon warming to room temp it gave another. :) Then in the fridge - it gave another crop... Seems I discovered the "everyielding crystal source".

Next experiment, I filtered the hypochlorite solution, titrated (1.39 M), mixed in some imide into the cold solution, warmed up to 80 C, acidified, no crystals. Not even after fridge treatment. (OK, it deposited a few milligrams, say 10-20 mg.)

Third experiment: made 1M NaOH solution, cooled in the fridge, in the cold solution dissolved the proper ammount of imide with stirring. Gave a nice, almost colourless solution. Then with the help of a Pasteur pipette I added the equivalent ammount of hypochlorite solution to this imide solution, drop by drop, during say 25 mins, vigorously stirred. At around 15-20% of the pipetting the solution started to become turbid and the colour turned to yellowish. Whitish precipitate appeared. At one point the precipitate failed to sink from the surface of the sulution for a few seconds and a drop of hypochlorite landed on it. It quickly turned brown! So it seems that the brown colour is from some sort of oxydation. Around 50% of the pipetting the reaction mixture was already "cocoa powder in water". The temp rose to 38 C in the end. Then heated up to 80 C, cooled down, acidified with first 30% HCl, then 96% acetic acid, filtered. On the filter very small ammount of black powder remained (a few tens of mg), filtrate is dark maroon clear liquid, started to become turbid (crystals formed), so I put it in the fridge.
After overnight fridgeing filtered the deposited crystals: big dissapointment, only cc 100 mg deposited. (Started from 10 mmol imide, 1.5 g!) Decided to concentrate the filtrate, cc 45 mls. By the time I got around 5 ml solution evaporated, lots of pinkish/minced meat coloured crystals deposited. I collected them on the filter, dried on a hotplate and was VERY HAPPY AND SATISFIED IN ADVANCE of the nice crop of my crude anthranilic acid.

Well, you can guess what happened. :D
I took a sample and tried to determine its melting point!
Epic failure!!! The annoying powder failed to melt even at 295 Celsius, the end range of my mp equipment!!!

What in the ... did I prepared???

I'm open to questions, suggestions, anything that can help me go further!

Pumukli - 8-7-2019 at 23:49

I was wondering that maybe what I got is the calcium salt of AA as a precipitate. (Because I used Ca-hypochlorite instead of the standard Na-hypochlorite.)
Could someone provide melting point data of this compound? I could not find any on the Net.

I plan to burn a sample of the substance. My understanding is if it leaves relatively big pile of ash then it contains inorganic stuff too.

Maybe a classical Ca2+-determination is coming! :) I think I have EBT and EDTA and such somewhere...



Any thoughts? Easier (more amateur friendly) method to determine Calcium - in case I don't feel inclined to run a fullfledged complexometric titration?

Pumukli - 9-7-2019 at 06:53

To continue my monologue I burned a sample of the powder at the tip of the blade of a thin knife. Yes, there remained a big pile of ash. The ash fizzed and pretty much dissolved when was contacted by a drop of 10% hydrochloric acid.

Tried to get more exact answers so I burned a weighed sample and determined the ratio of powder/ash. In case of Ca(anthranilate)2 the ratio is around 3.12 (anthranilate/carbonate ratio)

In my case I got only 2.74 but it was not a 100% burning, the ash was a bit blackish, so the real ratio is closer to the theory somewhat. And the sample was not a pure one either, it may very well contained a few percent inorganic salts (NaCl, CaCl2).

By the way the putative Ca-anthranilate is not soluble in 10% hydrochloric acid - it surprised me.

Pumukli - 13-7-2019 at 01:22

The "everyielding crystal source" deposited ridiculous ammounts of tan/light brown crystalls during the past few days, at room temp, without any cooling.
Essentially I filter the mass, get a clear, deep orangish red mother liquor. This liquid turns opaque in a few hours and after a day or two it becomes a fairly dense slurry. Rinse and repeat.

Meanwhile I prepared 0.01M EDTA solution and located my Eriochrome Black T indicator. Acquired some alt. purity CaCO3 too. Seems a proper Ca2+ determination is in progress. :)

Pumukli - 14-7-2019 at 11:54

If there was an "Analytical Chemistry" subsection on this board then I were posting this there:

I failed the standardization of my 0.01M EDTA with CaCO3 dissolved in dil. HCl and buffered to pH 10 by NH3/NH4Cl buffer.
Reason: I simply could not detect a sharp endpoint.
I will redo the whole procedure, from step 0, (Wash your equipment and rinse with real distilled water.)

-----------------

On the other hand it seems that starting from N-chloro-phthalimide (NCP) may have merit. I prepared some such compound (mp: 171-174 C, lit.: 183-198 C), obviously not too pure but run a synth with it. Something precipitated but still in the mother liquor, I'll do the workup later.

To do list: prepare better quality NCP next time. Dissolve the NCP in cold hydroxide solution (instead of room temp). Cool when hydrolysing. Catch CO2 to evaluate transformation and get clues on the best workup procedure.

Amos - 14-7-2019 at 16:13

I didn't honestly read too much of the lengthy comments above, but if anyone has doubts as to the purity of a fine, free-flowing cocoa-brown product, I've put some from the hypochlorite reaction with that exact description on a GC before and only got one peak. It's very easily discolored by extremely minute amounts of impurity that seem more or less unavoidable.

Metacelsus - 14-7-2019 at 22:36

Quote: Originally posted by Amos  
I didn't honestly read too much of the lengthy comments above, but if anyone has doubts as to the purity of a fine, free-flowing cocoa-brown product, I've put some from the hypochlorite reaction with that exact description on a GC before and only got one peak. It's very easily discolored by extremely minute amounts of impurity that seem more or less unavoidable.


That may be so, but GC won't detect nonvolatile impurities.

Amos - 16-7-2019 at 07:57

Quote: Originally posted by Metacelsus  

That may be so, but GC won't detect nonvolatile impurities.


Do you have any suggestions for what nonvolatile impurities might exist there? Anything heavy enough to not volatilize or at least decompose would VERY likely be be insoluble in water, which would allow for its easy removal by a filtration step which is already part of most procedures for producing anthranilic acid that involve unwanted precipitation prior to acidification. I imagine most byproducts would at the very least degrade to volatile compounds given that the GC oven reaches 280 C and the sample I used was in fact very saturated. To me it is like the same issue people commonly encounter with homemade or commercial samples of aniline. You can re-distill an amber product 15 times for a recovery of 10% out of frustration if you really want to, but you won't actually improve the quality of your reactant by any measurable amount.

[Edited on 7-16-2019 by Amos]

Metacelsus - 16-7-2019 at 09:40

Salts, for example. Also oligomerized oxidation products (similar to aniline).

I agree that it probably isn't worth trying to purify the brown powder to get rid of the color. It could be 99.9% pure and still be brown. The brown stuff is probably good enough for most purposes.

Pumukli - 17-7-2019 at 11:37

Thanks Amos, good to know that "cocoa powder" from this reaction is still as good as the original thing! :)

If you have some pure anthranilic acid - and some time - then you may conduct a precipitation experiment with it. I'd like to hear about the results of this. :)
Basically, dissolve some AA in dilute acetic acid and mix this solution with the solution of some water soluble Ca2+-salt. (Chloride, acetate, what you have at hand). Then wait and report what happens! I think you will got precipitation after a few hours (days), and that precipitate won't melt below 300 C.

Pumukli - 21-7-2019 at 09:50

I slowly explore parts of this particular Hofmann rearrangement and I found it more and more interesting from an amateur's point of view.
A seemingly simple reaction but there are several stages and hidden steps involved and it is a challenge to try to trace them.

My biggest achievement is that I think I solved the "black tar" problem and now I know how to avoid it altogether!
I also prepared a few hundred milligrams pretty good AA at last (mp: 146-147 C). Yield is not great yet but hopefully I could improve it too.