Sciencemadness Discussion Board - Preparation of anthranilic acid

Preparation of anthranilic acid

Sauron - 30-3-2008 at 02:25

This compound, o-aminobenzoic acid, or aniline-2-carboxylic acid, is usually prepared, per Vogel, from phthalimide by treatment with hypobromite or hypochlorite.

I made some phthalimide a couple years ago by three different methods.

Phthalic anhydride and ammonium hydroxide
Pthalic anhydride and urea
Phthalic anhydride and ammonium carbonate

All of which worked nicely.

But I never managed to get Vogel's procedure to work, for making anthranilic acid.

I have now compared Vogel with the procedure in "Fundamental Processes of Dye Chemistry", q.v. and find Vogel somewhat wanting.

However, even the more detailed and somewhat larger scale procedure in this book, is rather fiddly and at some junctures vague.

I therefore propose an entirely different route.

2-nitrobenzoic acid is a commercial product and not too costly.

Org.Syn. has a procedure for the low pressure (3 atm.) hydrogenation of p-aminobenzoic acid ethyl ester with PtO2 catalyst. Sounds like a job for my Parr 3911 shaker type hydrogenator. This apparatus was designed by Roger Adams for use with his catalyst, and he happens to be author of this procedure. The yield is quantitative.

Other reductions are referenced.

[Edited on 30-3-2008 by Sauron]

Attachment: CV1P0240.pdf (129kB)
This file has been downloaded 2855 times

Klute - 30-3-2008 at 02:37

I had the idea of reducing ethyl p-nitrobenzoate to benzocaine, using Zn/CaCl2. The yields are said to be moderate, but i think a CTH (not having a Parr shaker myself

)could work nicely here, though i haven't got any specific reference to back this up.
I was thinking of chlorinating p-TsOH, oxydizing the chloro-sulfonic acid at the methyl group with dilute HNO3, removing the sulfonic group as in the o-chlorobenzoic thread, then esterification and reduction.

The only difference i though was keeping the sulfonic group on during the methyl oxydation, as i would be using dilute HNO3. I don't think any di-nitration could accur with those conditions, but i prefer staying on the safe side. I guess some displacement of the sulfonic group can happen in those conditions though.
Just an foolish idea, but if after nitrating TsOH with mixed acids, could diluting the mixture directly oxidize the methyl group (dilute HNO3) AND remove the sulfonic group (dilute H2SO4)? That a long shot, and could give rise to isomer formation, but it would save alot of time! one-pot-3-reactions..

Sorry if i'm drifting away.

Sauron - 30-3-2008 at 02:59

There's two Model 3911 sets, at least one of them including the 4833 controller, on offer right now. The asking price is $400 each. I know how to get 25% off that, but I already have this model and its controller. These are in Michigan and are US voltage.

This is the 500 ml model. I paid that much for the 3911, and bouth the controller seperately for almost as much again. So whoever gets these will be getting a great deal. Almost makes me wish I needed more than one of them. But, I don't, although I do lust after the two liter version.

Anyone interested PM me for details.

benzylchloride1 - 31-3-2008 at 19:36

I produced about 50 grams of phthalimide several weeks ago for this synthesis. I planning on try it a soon as a get some more MnO2 for making the required bromine. Did you use glacial acetic acid in the end to precipitate it? Tomorrow, in ochem lab, I am going to synthesize P-aminobenzoic acid from p-acetotoluidide. This uses the same precipitation procedure. I want to make some anthranillic acid to produce benzyne which will be used in some Diels-Alder reactions.

Aubrey - 26-3-2009 at 00:11

I have also attempted the Dye chemistry synth and am planning to try again this weekend. Has anyone had success using the method described in dye chemistry?
The part that seems vague to me is the quantity of hypochlorite solution to use. It mentions using 2 mol of NaOH for each mole of NaOCl, but it also mentions putting in 20cc of NaOH immediately beore use. Do I calculate the quantity of NaOCl based on this total figure? How important is the accuracy of the amount of NaOCl? For the quantities described I calculaed the amount of NaOCl to use at about 1300ml but i dont feel confident in my calculations.

I'm also interested to know how important it is to neutralise it at 80 since other sources mention cooling it before neutralising.

It also mentioned (elsewherre) not to use too much HCl because it could destroy the anthranilic, but then it says the filtrate is acidifed with 40cc of HCl and 12cc glacial.

I'd be very interested to hear if anyone could shed light the above points. If only I had a Par apparatus.

benzylchloride1 - 28-3-2009 at 21:26

O-aminobenzoic acid is a amino acid and has a very narrow isoelectric point. This means that it only precipitates at a certain pH in high yeild. I was able to obtain about 15g of crude anthranilic acid from this reaction. The neutralization is the tricky part. The reaction mixture is first brought to about pH 7 with hydrochloric acid and glacial acetic acid is slowly added with stirring until precipitation of the dark colored product is nearly complete; easy to over shoot. The product is then filtered off and recrystallized from water. The product is tan in color and melts sevral degrees below the theoretical value.

PHILOU Zrealone - 31-3-2009 at 01:46

One could consider starting from easily available ortho-nitrotoluene.

CH3-C6H4-NO2 -KMnO4/H(+)-> HO2C-C6H4-NO2
HO2C-C6H4-NO2 -Fe/HCl-> HO2C-C6H4-NH2

Ortho aminobenzoic acid can be diazotised diluted in an inert solvant and lead by gentle warming to a strange coupling reaction by concomitant decarboxylation and denitrogenation forming (C6H4)°° diradical species...
If nothing else is present in the media to react, two of those gather to form a linear tricyclic compound, two benzene ring joined by a cyclobutane ring (Ar=Ar)

Attachment: C6H4=C6H4.pdf (2kB)
This file has been downloaded 937 times

S.C. Wack - 31-3-2009 at 03:05

It has been made from o-nitrotoluene 3 times in the literature, but with NaOH or KOH, and the yield is only 13-15%.

PHILOU Zrealone - 31-3-2009 at 04:02

Quote: Originally posted by S.C. Wack

It has been made from o-nitrotoluene 3 times in the literature, but with NaOH or KOH, and the yield is only 13-15%.

Strange, the oxydation of alkyl part of aromatic ring, especialy with a NO2 protective group must be very high even in pushed conditions...conversion of toluene to benzoic acid is nearly 100%...

S.C. Wack - 31-3-2009 at 14:49

I was wrong on the number of references; I forgot about the patent, and the Russian article.

The original 1899 BASF patent from o-nitrotoluene and NaOH:
GB189918319
DE114839

The latest (1972) article that I am aware of is this one.

All other references to this that I am aware of are given there. The local library has closed journal access for some time now, anyone interested in the references in the article will have to rely on Russian good will and Wiley archive access.

This route may be worthwhile for some despite the low yield.

S.C. Wack - 5-7-2009 at 11:15

Well I see that espacenet only supports full (very long) urls now. Nice of them to change that. Do the manual number search if you want to see the patents.

So after two years I finally have full journal access again...Z. angew. 385 (1900) [1 part o-nitrotoluene, 1 part KOH, 1/2 part water, 5 hr reflux] if anyone cares: http://ifile.it/iywpnc3

[Edited on 5-7-2009 by S.C. Wack]

crazyboy - 5-7-2009 at 14:16

I have been thinking about synthesizing anthranilic acid, what about o-xylene to phthalic acid via potassium permananate then phthalic acid to phthalic anhydride by heating and then any of the methods mentioned by Sauron to anthranilic acid?

Alternatively "oxidation of naphthalene tetrachloride (prepared from naphthalene, potassium chlorate and hydrochloric acid) with nitric acid" can supposedly yield phthalic acid but I haven't seen any experimental procedures for that either.

[Edited on 6-7-2009 by crazyboy]

DJF90 - 6-7-2009 at 07:59

Here is the synthesis from "Experimental organic chemistry - principles and practice"

Quote:

Procedure
Dissolve 8.0g of sodium hydroxide in 30ml of distilled water in a 100ml erlenmeyer flask containing a magnetic stirrer bar and cool the solution with stirring in an ice bath. Add the bromine [2.1ml, 6.5g, 41mmol] (CARE!) in one portion and stir the mixture vigorously until all of the bromine has reacted [look for the disappearance of the brown colouration] and the mixture has cooled to ca. 0*C. Continue vigorous stirring and add all of the finely powdered phthalimide [5.9g, 40mmol] to the solution, followed by a solution of a further 5.5g of sodium hydroxide in 20ml of water. Remove the ice bath, allow the temperature of the mixture to rise spontaneously to ca. 70*C and continue stirring for a further 10 min. Cool the clear solution in an ice bath with stirring (if the mixture is cloudy, filter under gravity before cooling), and add concentrated hydrochloric acid dropwise with a pipette until the solution is just neutral when a drop is spotted onto universal indicator paper (ca. 15ml should be necessary). If too much acid is added, the mixture may be brought back to neutrality by adding further quantities of sodium hydroxide solution, but it is better to avoid this by careful addition of acid in the first instance. Transfer the mixture to a 500ml beaker (foaming occurs in the next stage) and precipitate the 2-aminobenzoic acid by addition of glacial acetic acid (ca. 5ml). Filter off the precipitate with suction, wash the residue with 10ml of cold water and dissolve it in the minimum quantity of boiling water containing a little activated charcoal. Filter the hot solution to remove the charcoal and cool the filtrate with ice. Filter off the pure acid with suction, dry the residue with suction on the filter for 5 min and complete the drying to constant weight in an oven at 100-120*C.

[Edited on 6-7-2009 by DJF90]

S.C. Wack - 12-7-2009 at 14:28

Quote: Originally posted by crazyboy

what about o-xylene to phthalic acid

If you happen to have o-xylene, maybe you have some sodium azide as well. In Angew. Chem. 45, 536 (1932) a bunch of amines are made using the Schmidt reaction: aniline, benzylamine, phenethylamine, etc... and the title cpd., from o-phthalic acid. They say that with 4.15 g. of the acid with 10 ml. conc. H2SO4 and 40 ml. CHCl3 and a temperature maintained at 45-50C by the slow addition of 4.9 g. NaN3, they obtained 2.7 g. anthranilic acid (yield 80%) and 0.1 g. o-phenylenediamine. No workup details; see volume 3 of Organic Reactions.

[Edited on 12-7-2009 by S.C. Wack]

crazyboy - 12-7-2009 at 17:48

No I don't have sodium azide. I plan on separating o-xylene from meta and para xylenes using the procedure detailed in COPAE:

"When the mixed xylenes are treated with about their own weight of 93 per cent sulfuric acid for 5 hours at 50°, the o-xylene (b.p. 144°) and the m-xylene (b.p. 138.8°) are converted into water-soluble sulfonic acids, while the p-xylene (b.p. 138.5°) is unaffected. The aqueous phase is removed, diluted with water to about 52 per cent acidity calculated as sulfuric acid, and then heated in an autoclave at 130° for 4 hours. The m-xylene sulfonic acid is converted to m-xylene, which is removed. The o-xylene sulfonic acid, which remains in solution, may be converted into o-xylene by autoclaving
at a higher temperature."

manimal - 13-7-2009 at 18:46

Aniline used to be mfgd. by heating phenol in a sealed tube with ammonia. If that reaction is applied to salicylic acid, would anthranilic acid be the result? I suppose there may be steric hindrance.

stygian - 13-7-2009 at 19:09

In the scimad library, in one of the dye-chemistry books, they refer to producing (di)methylaniline by autoclaving aniline with methanol. This + Hoffman-Martius rearrangement + some oxidation should yield anthranilic.

not_important - 13-7-2009 at 20:51

Those reactions require temperatures in the range of 200 to 250 C, and 50 bar or higer pressures.

Salicylic acid would likely decarboxylate, likely you'd have to use ammonium carbamate ( "ammonium carbonate" ) as a source of ammonia and CO2 to prevent that. At the temperatures required I suspect the para isomer would predominate.

As for the methylaniline route, you left out protecting the amino group and aromatic ring against oxidation, which will add at least two more steps.

atomicfire - 18-4-2011 at 11:33

What is the reaction between phthalic anhydride and urea? Under what conditions does this occur?

byko3y - 18-4-2015 at 15:39

Some time ago I posted some short report about my failure to perform Hofmann rearrangmenet of phthalimide to anthranilic acid and about running qualitative analysis on the pthalimide http://www.sciencemadness.org/talk/viewthread.php?tid=19662&... Finally I managed to get a successfull reaction with a decent yield.
A nice book named "Organic Chemistry" by Fieser, Fieser (1956) helped me to rule out what is actually going on in the hofmann rearrangement of phthalimide. The truth is - you can't do Hofmann rearrangement on phthalimide. Phthalimides, like succinimides, can reversibly attract halogen atom, and they remain phthalimides. Vogel also mentions the hydrolysis, but there are no instructions about it in the experimental part, they just mix everything together and let the reaction heat itself.
The actual compound undergoing the rearrangement is a base hydrolysis product - sodium phthalamate or sodium 2-carbamoylbenzoate. If you fail to hydrolyse the phthalimide properly - you gonna get really poor yields of anthranilic acid, you gonna get a clear amber solution with no precipitate, which is exactly what happened to me.
You should not add more hypochlorite then stochiometric amount, because on heating the unreacted hypochlorite will convert to a more stable mixture of chloride and chlorate. The latter is a really strong oxidant that will oxidize and chlorinate the benzene ring, lowering the yields. This is why it's important to titrate the hypochlorite (e.g. iodine-thiosulfate).

US patent 2653971 has some nice instructions on the reaction conditions. I came to the same reaction sequence as the patent author, while instructions from the rhodium's page look insufficient. So briefly what I did was 14 g Pthalimide + 10 g NaOH + 80 ml water + cooling ... + 150 ml NaOCl + cooling ... heating... + HCl... + Acetic acid -> collect precipitate.

First I dissolved the phthalimide in the 80 ml solution of 10 g NaOH. This gives a pale yellow solution (first flask in the picture). It's important to make sure the solution temperature is always lower then 20°C, otherwise the phthalimide will be hydrolizing into phthalic acid rapidly (the patent says the rapid hydrolysis starts at 35°C). This is why you probably want to dissolve it in an ice cold NaOH solution, add phthalimide in portions, and keep the reaction flask in a fridge. For some reason I could smell a definite odor of ammonia coming off the solution - probably because of some inorganic ammonium salts like NH4Cl.
After few hours a conversion of phthalimide to sodium phthalamate is complete. Next you should add not more than a stochiometric amount of a cold hypochlorite solution trying to keep the mixture below 40°C, otherwise the anthranilic acid will start forming, further reacting with hypochlorite, thus forming chlorinated byproducts. Note that the color of the mixture didn't change. At least I didn't noticed nothing.
Also, patent recommends to ensure the hypochlorite solution was never heated higher then 15°C during it's preparation and storage, but I have no way to comply the requirement because my hypochlorite solutions all were stored at 20-25°C. So this one probably led to some byproducts and excess coloration of my final solution.
It's important to keep the alkali always in excess, otherwise the chloramine will react with chloride ions generating a free chlorine, which is useless for our reaction.
When everything is mixed and exothermic reaction is no longer observed, you can continue to a heating of the mixture. It's possible to use the heat of chlorination for the subsequent rearrangement, but I'm not that good at this reaction.
Patent tells 60°C is enough, and I have a big research of the Hofmann rearrangement that says only a brief heating is needed for the rearrangement to complete, although there's no big harm in reaching 75-80°C for a 1-2 minutes to ensure the rearrangement is complete. Thus the reaction mixture turns pale amber as the patent says (2nd flask in the picture).
I think it's important to pay attention at the odor of the reaction mixture above 40°C. If you have an odor of ammonia - most likely some wrong reactions are going on. In my last attempt using a freshly regenerated phthlimide I could not notice even a slight odor of ammonia at this step.
Then, after cooling, I carefully adjusted the ph to 7-8 and got almost no precipitate of unreacted something (less than 1 g), because I used at least stochiometric amount of the hypochlorite. There's a question about what compound it is and should any preciitate be formed at all. I think the precipitate from rhodium's procedure is unreacted phthalimide, but we ensured it is completely hydrolized.
You should use an excess of phthalimide if you want a reasonably pure product, but I have no idea about how to separate the anthranilic acid and phthalamic acid (2-carbamoylbenzoic), and nor patent, neither vogel or rhodium says about any recovery instructions.
So the last step is a carefull acidification with acetic acid accompanied by a lot of bubbling of the CO2 coming out of the solution. The solution turns maroon (third flask in the picture).
If you screwed something and your yield is really poor or not enough acetic acid is added yet, your solution will be clear. But when you have a lot of free anthranillic acid, it will start precipitating as it does at the fourth flask in the picture.
So, after filtering, washing with water and drying the precipitate I've got 6 g of crude anthranilic acid, corresponding to 45% yield plus copper salt might give aprox 2 g of more anthranilic acid. Which is almost the same as the yield in the vogel's procedure via hypocbromite and rhodium's procedure which is an adoptation of vogel's procedure using hypochlorite, though for some reason I could not reproduce the results using hypochlorite. Notice that vogel's procedure uses freshly prepared hypobromite.

Obtained crystals of anthranilic acid have peru color and seem to have at least few percents of impurities. I'm pretty sure I've got as high as 5-10% of byproducts, while a 99% compound shall be white colored. Anthranilic acid can be purified by recrystallization from water-detergent mixture. See "Recrystallization of Organic Compounds From Detergent-Water Systems" Sugihara, Newman, J. Org. Chem. 21, 1445 (1956). DOI: 10.1021/jo01118a032
Or you can just use hot water as Vogel says.

There's a nice qualitative test for anthranilic acid - a solution of Cu(II) (e.g. CuSO4) in dillute acetic acid.

A usual benzoic acid just floats on top as white flakes, phthalic acid doesn't change the solution color (remains blue, not shown on the picture), anilines give soluble salts too, but anthranilic acid (o-aminobenzoic) gives a green precipitate and m-aminobenzoic acid just gives a blue coloration (not tested by me). This test also probably will give a green precipitate for chlorinated derivatives of anthranilic acid, like 3-chloro-2-aminobenzoic acid and 5-chloro-2-aminobenzoic acid.

Also, after you filtered off the precipitate of anthranilic acid, you will get a clear maroon solution with a tiny amount of anrthranilic acid in it. You can precitate it with a Cu(II) salt (CuCl2 or CuSO4). Actually, when I added a CuSO4 solution, a lot of bubles started to come out and I've got two layers of precipitate (I used a CuSO4 saturated solution): lower gray heavy layer of unknown substance and middle green layer of copper anthranilate which is much lighter and is really hard to comletely precipitate, so I just filtered it. At first I thought it's a good idea to filter the solution with a green anthranillate precipitate in it, avoiding the gray bottom layer which is relatively hard to perturb. Latter I found that the green layer has almost no anthranilic acid. Also most likely I didn't add enough acetic acid and the carbamate didn't hydrolize completely (maybe even as high as 20% of yield), this is why more CO2 came out on this step and reacted with coper. Thus there are a mixture of some copper salts in the precipitate.
Next I hydrolized the solid with a NaOH solution with ph >> 12. Solution color changed to turbid green, so I'd say that the very first copper salt precipitate content was copper bicarbonate, copper acetate and copper anthranilate. After filtering the solid I've got a familiar clear maroon solution of sodium anthranilate, which I overheated and discarded, so I don't know exactly how much of the product was there, but judging by a solid left I can say there was up to 4 g (30% yield) of anthranilic acid there.

Another possible reason why I got a poor yield in my first attempts is a dirty phthalimide. Melting point of mine was like 224°C, but after melting the crystallization point was like 232°C.
The good news are: phthalimide is the most stable of all the phthalic coumpounds in non-alkali high temperature conditions. So you can just add ammonia solution to anything you've got, except alkaline salts and esters, and after boiling off the water and reaching the temperature above melting points of all possible phthalic compounds ( 300°C ) you will get a reasonably pure phthalimide. Basically, you can follow the procedure Org. Synth. 1922, 2, 75. DOI: 10.15227/orgsyn.002.0075 http://www.orgsyn.org/demo.aspx?prep=CV1P0457

Sources: some internet article and "Concise Encyclopedia Chemistry" By Mary Eagleson (1994), p. 832-833 (you can read at google books)
Aforementioned orgsyn article says you can make phthalimide form ammonium phthalate, but the yield is in question. This can be a nice OTC way to get the phthalimide derectly from the phthalic acid, either purchased or obtained via hydrolysis of different phthalates.
Amonium carbonate also will work, but I don't know whether urea is a good idea, because the reaction can become violent and the urea can't withdraw the heat from the mixture unlike ammonium carbonate and water solution of ammonia.

[Edited on 18-4-2015 by byko3y]

[Edited on 18-4-2015 by byko3y]

S.C. Wack - 19-4-2015 at 08:50

BTW the preparations I've come across that use hypochlorite (obviously one might not want to trust the manufacturer's % statement):

Systematic Organic Chemistry:

40 g of finely powdered phthalimide and 80 g of NaOH are dissolved together in 280 ml of water, the solution being cooled during the operation. The solution is agitated, and 400 g of a 5% solution of sodium hypochlorite run in. When all is added, the solution is warmed for a few minutes at 80° to complete the reaction; it is then cooled and neutralised exactly with hydrochloric or sulphuric acid. An excess of strong acetic acid is added to precipitate the anthranilic acid, which is filtered off and washed with water. Any anthranilic acid remaining in the filtrate is precipitated as copper anthranilate by the addition of a saturated solution of copper acetate. After standing for some time the precipitate is filtered off and suspended in a small quantity of warm water, while a current of hydrogen sulfide is passed into the suspension. The copper sulphide formed is filtered off, and anthranilic acid recovered from the filtrate by concentration on a water bath. It may be recrystallised from hot water.

Total Yield.-85% theoretical (31.5 g). mp 145°.

+

JCE 54, 643 (1977):

Dissolve 16.8 g of NaOH in 50 ml of water in a 500-ml Erlenmeyer flask and cool in an ice-salt bath until the temperature is about 10°C. Add 250 ml of Chlorox (sodium hypochlorite) and cool to below 5°C. Prepare a solution of 22 g of NaOH in 80 ml of water and cool to below 20'C. Add 24 g of finely powdered phthalimide in one portion to the cold, alkaline Chlorox solution, and swirl vigorously. Add the second NaOH solution to the reaction mixture, swirl vigorously, and place on the bench top. Place a thermometer in the flask to monitor temperature. The solid should dissolve as the temperature rises slowly to about 25°C; the temperature then should rise rapidly to about 50°C. Heat the reaction to 80°C on a steam bath and maintain that temperature for ~3-5 min. If any undissolved material is present in the reaction mixture, filter at this point. Cool the reaction mixture in an ice bath, then add concd HCl slowly until the reaction mixture is just slightly basic (wide range pH paper or a pH meter) (about 60 ml of acid is required). Precipitate the anthranilic acid by slowly adding 25 ml of glacial acetic acid (Caution: reaction mixture tends to foam at this point). Collect the product on a Buchner funnel and wash with small portions of very cold water until the odor of acetic acid is no longer detectable. Air dry the solid until the next laboratory period. The yield of anthranilic acid should he 12-17 g; mp 142-144°C (lit. 144-145°C)). If desired; the product can be decolorized (NORIT) and recrystallized from water...

byko3y - 19-4-2015 at 10:16

Where have you been all this time? Basicly the first procedure is the same as I posted, and they also completely developed the recovery of copper anthranilate.
So where did you find it? I don't understand or can't find the source.
The second preparation is the same as at rhodium's page and it didn't work for me.

S.C. Wack - 19-4-2015 at 12:53

Quote: Originally posted by byko3y

I don't understand or can't find the source.

It's one of my books that I scanned. Their references are a 1901 lecture involving indigo from naphthalene:
https://books.google.com/books?id=voU3AQAAMAAJ&jtp=139
and the 1890 BASF patent DE55988 which they pretty much translate. The JCE authors adapted Vogel's hypobromite.

Templar - 21-4-2015 at 07:36

Anthranilic acid is good, but if we're going to bother going to quinazolinones, we might as well look at processes where chemicals are less watched.

Phthalic acid is a good starting point for getting either anthranilic acid or isatoic anhydride.

Anthranilic acid still requires the moderately hard to get acetic anhydride, and if a good yield is desired POCl3 or PCl3/PCl5.

I have been investigating synthesis of isatoic anhydride, I posted this here http://www.sciencemadness.org/talk/viewthread.php?tid=33344 in regards to that. However, I didnt realize at the time that the precursor, phthalimide, had a melting point within 5 degrees of isatoic anhydride, so I wasnt really able to tell what kind of purity I had.

If one could get decent isatoic anhydride, several journal entries and two synthesis reports confirm that ring opening with an aromatic amine followed by ring closure with an acetylating agent (acetylacetone) would work to produce any range of quinazolinones.

Anthranilic acid is a bit of a strange beast, I didnt realize how annoying it can be dealing with zwitterions until I met this one. The basic form of anthranilic acid seems to be a brown oil (confirmed by another patent as well) which can solvate in non polar solvents. The amine salts of anthranilic acid also appear to have a decent solubility in water.

I came across a bird repellent that reportedly claimed 40% methyl anthranilate with 60% veggie oil. This was mixed with 20% HCl soln, resulting in a suspension of what I believe is the amino HCl salt of anthranilic acid, the oil, and water.

After filtering, this was dissolved in a NaOH solution of pH 11.6 and a milky opaque solution with red/brown oil droplets appeared. I thought washing with hexane would remove residue veggie oil; turns out it also removed anthranilic acid.

Re acidifcation of the aqueous gave no ppt, but re-combinging with the hexane/oil mix on high stir produced fluffy white crystals that were filtered, dried and melted at 168C.

I would have taken a water insoluble compound like isatoic anhydride over this anyday, but I must still figure out how to test the purity of it.

Hm..

Metacelsus - 21-4-2015 at 11:56

The isatoic anhydride synthesis you posted seems a lot like the synthesis of anthranilic acid (and you did observe some anthranilic acid produced, you say). How is it different, in terms of conditions and mechanism?

byko3y - 21-4-2015 at 15:01

The yields of isatoic anhydride are essencialy the same as for anthranilic acid procedure, if the latter is done correct. For some reason there was a 1890 with a perfect procedure but nobody remembered it, so I needed to reinvent it once again.
And it's not easier to prepare an isatoic anhydride first, because you will need to separate the anrthranilic acid somehow, and this way the procedure becomes even thicker to perform than the straight one.
As mr wack and me already wrote, you need to precipitate copper salt - it's the best way to separate anthranilic acid from anything.
Gonna write an answer to the isatoic thread describing the reaction mechanism according to patents US 3,867,951 and US 4,328,339.

Templar - 21-4-2015 at 22:45

Quote: Originally posted by Cheddite Cheese

The isatoic anhydride synthesis you posted seems a lot like the synthesis of anthranilic acid (and you did observe some anthranilic acid produced, you say). How is it different, in terms of conditions and mechanism?

This is a good question. I was wondering this too, but I think the differences are in the details. The temperature ranges are more extreme.

In a 250ml Erlenmeyer mix 80ml H2O and 8.0g NaOH, cool to about +5°C and don't mess with it until later. To another 500ml Erlenmeyer add 210ml Alclorite (or other ~3.5% NaOCl solution, 215ml is 0.1mol) and cool this to +5°C or cooler. Then add 8.0g (0.2mol) NaOH to the now cold Alclorite solution, by precooling it we prevent the temperature from getting too high. Cool the NaOH/Alclorite solution to -10°C in the freezer (If you check the pH of the Alclorite you will find it to be 12-13, but this doesn't mean that I contains much NaOH from the beginning, a 0.05% solution has pH 12).
Grind 15.0g (0.1mol) phthalimide to a very, very fine powder. Now add this to the cold (-10°C) Alclorite solution, plug the flask and mix with magstirrer on fastest setting (or shake). The phthalimide will dissolve within 5 min and the solution will not be more colored than before. Place in the freezer again for at least 30 minutes.
Now add the cold NaOH solution prepared above to the cold (not over +5°C) Alclorite solution, nothing happens. Let stand in room temperature until it reaches 20°C. Now heat to 80-85°C over 10 minutes and hold there for at least two minutes but not more then five. The solution will be a little less achromatic now, but only slightly. Cool in water to room temperature or preferable below so the awaiting hydrochloric acid addition will not heat it up to much.
Transfer the solution to a 1000ml beaker, keep some (~20-30ml) in the Erlenmeyer flask. Stir at medium speed on a magstirrer. Now begin to add hydrochloric acid, the ~30% OTC version will do more then fine. You can probably add the first 25ml rather careless. But now you must measure pH, by adding acid drop by drop and using the saved solution (the 20-30ml above) as backup you should eventually have a nice pH 7.0 solution (if you mess this up, a pH above 7.0 is MUCH better then getting it below 7.0). The solution will become a little more colored when you add the hydrochloride acid.
At this point your solution has probably become cloudy because of a precipitate, this must be unreacted phthalimide (about 2g) so filter (suction!) and save it. Now you will have a clear solution again.
Again add your solution to a 1000ml beaker, make sure the temperature is about +20°C or below. Strirr at medium speed and begin to slowly add glacial acetic acid. If foaming has not begun after the two first milliliters, you probably should have neutralized better above, but just keep adding acid. A lot of CO2 is formed so add it slowly. 15ml will probably be enough, but some extra milliliters will do no harm. The antharanilic acid will precipitate, as you add the GAA.
Filter (Buchner, plastic cover to such away solution) and dry the crystals in an oven at +50°C (higher temperature might darken the crystals) for several hours. The resulting powder should be slightly yellow/brown but NOT much. The yield at this point would probably be over 9g. Recrystallisation from water with some decolorizing carbon present will give nice clean crystals.
After some runs you can grind up the saved phthalimide and reflux it in water (10-15ml/g, it will not dissolve) for half an hour, then cool and filter to get pure phthalimide ready to be reused.

Contrast between this and the isatoic anhydride synthesis. The isatoic anhydride synthesis lacks the same extremes in temperature, going from about 5-10C to 25-30C. The differences are subtle but the reaction products are very clearly different.

byko3y - 21-4-2015 at 23:08

The procedure you posted is not working well. This is a copy paste of the procedure from rhodium's page you can find by googling, and the link is already posted here, and it is in fact an adotation of vogel's procedure, which, in turn, can't be reproduced by some people. So don't mislead people - you don't know what you talk about.
And actually I answered the question about the difference of the procedures in this thread and the one dedicated to isatoic anhydride ( http://www.sciencemadness.org/talk/viewthread.php?tid=33344 )
Also some of the patents mentioned in the thread contains research about the yield of byproduct - which is anthranilic acid, while the desired product is isatoic anhydride. And it's all about hydrolysis, as I said.

[Edited on 22-4-2015 by byko3y]

Templar - 22-4-2015 at 01:28

Quote: Originally posted by byko3y

The procedure you posted is not working well. This is a copy paste of the procedure from rhodium's page you can find by googling, and the link is already posted here, and it is in fact an adotation of vogel's procedure, which, in turn, can't be reproduced by some people. So don't mislead people - you don't know what you talk about.
And actually I answered the question about the difference of the procedures in this thread and the one dedicated to isatoic anhydride ( http://www.sciencemadness.org/talk/viewthread.php?tid=33344 )
Also some of the patents mentioned in the thread contains research about the yield of byproduct - which is anthranilic acid, while the desired product is isatoic anhydride. And it's all about hydrolysis, as I said.

[Edited on 22-4-2015 by byko3y]

I havent had time to read fully through the patents, but if anthranilic acid is the byproduct and not phthalic acid or phthalimide salts, then a melting point test should readily determine whether or not you have a product of mainly isatoic anhydride.

Anthranilic acid HCl melts at 168C or so, and zwitterion anthranilic acid at 146C or so. These would produce significant depressions in purity if present at levels more than 10% I would say.

Recrystallization report

byko3y - 4-5-2015 at 08:05

Finally performed purification of the anthranilic acid I prepared in my previous report.
Attempt to sublime the crude product at atmpspheric pressure led to melting into tan oil and distilling an aniline from it (aniline detected by odor and by Na2Cr2O7+H2SO4 oxidation into aniline black). Maybe the pure anthranilic acid was dissolved in the aniline, but their separation gonna be a pain.
You might have already known that the anthranilic acid is susceptible to decarboxylation on heating above melting point - this is why I've got the aniline.
Rate of decarboxylation of the anthranilic acid after heating for 1 hour (decomposition rate is almost constant for the first hour), data from "The Decarboxylation of Anthranilic Acid" - Stevens, Pepper, Lounsbury, which in its turn quotes data from Pawlewsbkri., Ber. 37: 592. 1904 :
150°C - 24%
160°C - 48%
190°C - 88%
200°C - 96%
So, heating at 200°C for 5 minutes leads to ~10% decomposition.

However, there's a much, much easier way to purify the anthranilic acid, which is crystallization from water. I was not able to obtain its solubility at 100°C, but I could say by my observations it's somewhere at 30...100 g/L.
I suspended 1.78 g in 30 ml of water. Something like half of my crude product refused to dissolve even after boiling, so I needed to filter it (fast filtration through a preheated funnel-paper filter). In the end I've got a clear maroon-colored solution (at almost 100°C).
While cooling to 5°C, nice long needles of anthranilic acid apeared all over the solution and the solution lost half of it's color.

After filtering and drying, 0.73 g of tan-colored crystal were obtained. As you can see, those are most likely 99% pure.

Less than half of the crude product made it through the crystallization, so I'd say the original product had 50% content of anthranilic acid. However, I lost a decent amount of anthranilic acid due to insufficient amount of acetic acid (see my previuos report for details) and failure to process a copper anthranilate precipitate. The former could lead to isatoic anhydride formation.

byko3y - 6-5-2015 at 08:35

Small update:
Totally (after few trials) I've got 2.8 g of a pure anthranilic acid from 4.5 g of crude product, and the yield could be even higher (up to 80-90%). First time I did some mistakes, such as not heating the solution strong enough (80°C instead of at least 95).
You can detect the undissolved anthranilic acid by its color - it is tan colored, while everything else is either black or white.
After I have filtered off the hot solution, I got only black tar and almost invisible white crystals left on the filtering paper. This is what it should look like.

If the residue contains peru-tan crystals - you either used not enough water, or your solution is too cold.
After cooling and filtering off precipitated crystals, mother liquor containg a decent amount of anthranilic acid (you can judge by it's maroon color), so I carefully evaporated it until 1/4 of the original volume left, and then precipitated the crystals again (gives amount like 1/4-1/2 of the first crop).
Also, I'd like to mention that the solubility of anthranilic acid rises sharply near the water boiling point, but it's really hard to filter the solution and keep it 100°C so there's no strick proportions of crude product to water. I have a data about 450g/L solubility at 100°C, but I failed to reach these numbers.
The data all over the internet says the anthranilic acid has white color - I doubt it's true. From some specific point of view some crystalls might look white, but mostly they have a definite tan color.

[Edited on 7-5-2015 by byko3y]

DraconicAcid - 6-5-2015 at 08:43

My anthranilic acid is beige-yellow, without a trace of maroon.

byko3y - 6-5-2015 at 08:50

Edited the previuos post - mine were not maroon either, it was less reddish, peru-tan describes it better. But they were not beige-yellow for sure.
UPD once again: do you have crystals or a fine powder?

[Edited on 6-5-2015 by byko3y]

Zephyr - 6-5-2015 at 10:21

I also recently synthesized anthanilic acid, and like byko3y's, it is a light tan powder. I used nilered's procedure and double the amount. The yield was 2.45g but I plan on recrystallizing.

byko3y - 6-5-2015 at 16:02

Nile Red follows rhodium's procedure, and once again - he forgets the importance of complete hydrolysis of phthalimide. Well, I don't think he even knows it.
Anyway, he got the same powder I've got. Light-tan-peru powdery powder.
Pinkhippo11, what was the amount of phthalimide you used, so we can judge about yields.

Zephyr - 6-5-2015 at 16:51

Like I said, twice NileRed's amounts, so 5.16g, which leads to a yield of 51%.
However, I probably could have gotten higher if I had fully acidified it.

byko3y - 6-5-2015 at 17:12

51% is a relatively low yield. If Nile Red got even lower, than we can clearly see, that unmodified RHODIUM'S PROCEDURE just is NOT WORKING (mix NaOCl with NaOH, then add phthalimide, add more NaOH, heat, etc...). I want to emphasize that, because over and over people recommend it.
Forgot to link the post in this thread with a good procedure and yields http://www.sciencemadness.org/talk/viewthread.php?tid=10272&... (first one)

[Edited on 7-5-2015 by byko3y]

byko3y - 10-5-2015 at 05:14

Finally measured the melting point of my product - it is 145+-3°C (at 141°C it is definitely solid, at 148°C it is definitely liquid). Sorry, don't have appropriate devices to measure 0.1°C precise, and this stuff starts to decompose at its melting point to aniline, thus lowering the melting point. At least it looks like anthraniic acid and not some chlorinated deriative.

Templar - 12-5-2015 at 15:41

There is also methyl anthranilate or o toluidine to start the synthesis from.. the route involving o toluidine and permanganate makes an annoying brown manganese sludge, but this is cleanly removed through filtration over celite...

I think the methyl anthranilatr hydrolysis route is very interesting, but i have not managed to precipitate solid anthranilic acid from it in the zwitterionic form, even at the isoelectric point of pH 3.5... The toluidine route may not be so feasible but the ester route should and wont be as watched.

Maybe its even possible to acetylate the ester as is using AA and DMAP if required as someone suggested.

ChemPlayer_ - 22-6-2015 at 03:05

I'm getting a yield of anthranilic acid from the hypochlorite (bleach) Hofmann reaction with phthalimide between 44% (worst case) and 60% (first time was luckiest!). The product is a light tan cream coloured powder which darkens slightly when re-crystallised.

I use the method of dissolving the phthalimide in the NaOH first, then adding the hypochlorite solution (I use 8.25%).

I suspect that the quality of sodium hypochlorite bleach from bottle to bottle (bearing in variations in batch / expiry date storage temperature and ambient conditions in the supply chain) can be quite variable in practise, and this may lead to concentration differences. I haven't yet tried titrating the bleach first to test this but I will start doing this on the next run.

Video of the detailed process: https://www.youtube.com/watch?v=DIadKyVKJAc

byko3y - 22-6-2015 at 04:13

At least 1.5 g of your product is dissolved in the water you've filtered. The color of the water tells about how much anthranilic acid (AA) it contains. AA may be more soluble in an acidic solution than in a neutral one (although I don't know).
Freshly prepared concentrated hypochlorite/hypobromite works the best.
Also, I have no idea about exact impact of excess of hypochlorite on product. AFAIK, aniline on reaction with hypochlorite gives some polymerization product, right? It might give some cyclization product with anthranilic acid.

ChemPlayer_ - 22-6-2015 at 09:18

Thanks - that's very helpful info. I'd been assuming it was the isocyanate intermediate which was potentially prone to being over-oxidised by excess hypochlorite, but thinking harder about it I guess that once this intermediate is formed ultimately the amine is the only likely end product of hydrolysis or anything else that could happen.

It may be that getting the anthranilic acid out of solution could simply be the hardest part of this; it would explain 'beginners luck' because I was certainly over-cautious with neutralisation the first time I did this!

byko3y - 22-6-2015 at 14:49

I have an idea to boil out the excess of water after AA filtration to obtain more AA precipitate without NaCl precipitating (to avoid tricky copper extraction procedure), something like 80 ml of resulting solution per 15 g of total theoretic AA in the reaction. Of course, I'm talking about the case of low concentration NaOCl solution used, which is quiet common because it is mostly sold dilluted, the highest one I was able to obtain is 6%.
Well, actually for most cases you will get more than 80 ml of solution when making 15 g of AA.

CuReUS - 23-6-2015 at 03:13

what about hitting phthallic acid with one mole of HN₃(NaN₃+H₂SO₄).I know azides are dangerous,but they are cheap and the reaction is clean

[Edited on 23-6-2015 by CuReUS]

Recovery of anthranilic acid from copper anthranilate?

Metacelsus - 26-6-2015 at 15:09

I've followed byko3y's adaptation of the US patent procedure 2653971, and got a decent yield of presumed anthranilic acid (tan powder, not fully dry yet so I don't have a mass). However, I was able to precipitate more as copper anthranilate. How do I get it back to anthranilic acid (S.C. Wack's post mentioned hydrogen sulfide, but I don't want to deal with that.)