Possible route to resorcinol from Tylenol / Paracetamol?

The only step I'm unsure about is how good the reductive demamination with EtOH is.



[Edited on 12/29/2007 by guy]

Edited title. Chemoleo

[Edited on 30-12-2007 by chemoleo]

Edit: No "e" at the end of Paracetamol

[Edited on 12/30/2007 by guy]

So thanks to not_important, I revised the 1st step:




I trust this article more because they actually conducted 2 melting point tests and NMR to confirm the product.

[Edited on 12/30/2007 by guy]






I have a question: Will the nitro group ortho to the amino group make the reduction of the diazonium favorable even though there is an -OH group para to it? I looked up the activating effects of substituents and the nitro group seems to be a stronger inducing group than the -OH group is activating.

[Edited on 12/30/2007 by guy]

I would rather use HCl like it's done in several of the above examples. I managed to find a literature example that is nearest to what you are up to (attached):

2-Methyl-5-hydroxybnzimidazole
S Gershon, G Webster
J. Am. Chem. Soc., 63 (1941) 2853-2853.

They cyclisize 4-acetyloxy-2-aminoacetanilide to 2-methyl-5-hydroxybnzimidazole with conc. HCl in ethanol which is just fine since your 4-hydroxy-2-aminoacetanilide is the first intermediate (phenyl acetates hydrolyze very rapidly).

Keep also in mind that 2-Methyl-5-hydroxybenzimidazole, like benzimidazoles in general, is a weak base and can form stable salts with strong acids (pKa of protonated 2-methylbenzimidazole form is 6.3). The hydrochloride (mp 256°C), picrate (mp 217°C) and nitrate are described in the literature. Due to the phenolic group it is also a weak acid (pKa of phenols is about 10, depending on the substituents).

Quote: Originally posted by Nicodem

Even if the nitrosation/oxidation gives you the correct regioisomere as the above reference clams (I have doubt's but could be), you will still have to deaminate by nitrosating a p-hydroxyaniline which I do not think it is possible without ending with the corresponding p-benzoquinone (I could be wrong though, but not motivated enough to check the literature). PS: Resorcinol is terribly cheap and easy to buy. You might have at least chosen a better target. For example, if you manage to prepare 3-nitro-4-acetamido-phenol you could make 2-methyl-5-hydroxybenzimidazole from it... or anything at least remotely more interesting than resorcinol.

Although this post is currently more relevant to the DDNP thread in the energetic forum I am posting here for two reasons, firstly I think this post it nolonger quite relevant to the current state of the DDNP topic and secondly it was first described in this thread.

I recently started some experiment with the aim of producing benzene-ring substituted benzotriazoles. in this experiment the starting material was paracetamol and the intermediate target was the 3-nitro-paracetamol via nitrosoation as described by Matsuno et al., Chem. Pharm. Bull. v37(5), 1422-1423; 1989.

The original reference by Matsuno et al states that paracetamol can be smoothly nitrated under weakly acid to neutral conditions by a three to five fold excess of nitrous acid (27% yield) or by a mixture of nitrous acid and hydrogen peroxide (96%) but experimental details are at best sketchy.

Since the use of hydrogen peroxide presumably acts as an oxidizing agent there should be a lesser requirement for the nitrous acid and so in the initial experiment described below I only used a threefold excess of nitrous acid. However, due to the presence of much tarry product and the continued evolution of nitrogen oxides for many hours I feel that with hydrogen peroxide only a small excess, perhaps as little as 1.1 to 1.2 moles of nitrous acid per mole of paracetamol, are required particularly given the mechanism proposed in Matsuno’s paper.



If the hydrogen peroxide brings about the initial oxidation to the quinimine then only one molar equivalence of nitrous acid is require. That said when the hydrogen peroxide was added to the paracetamol no visible reaction was seen (I would expect the quinimine to be coloured). However, on the addition of the nitrite an immediate orange colour developed, the reaction is slightly exothermic and appears very rapid.

Experimental.
3.006g of recrystallized (from tablets) paracetamol were dissolved in 20ml of warm 80% acetic acid and cooled to 5° C, 11.5 ml of 6% hydrogen peroxide (1 mol. equiv.) were added and the mixture cooled in the ice chest to 0° C. 4.131g (3 mol. equiv.) of sodium nitrite were dissolved in 10ml of water, diluted to 18ml and cooled to 0° C in the ice chest.

When removed from the ice chest after about 40 minutes the solution was at -7° C. The sodium nitrite solution was added to the paracetamol solution 2 ml at a time, after 4 ml had been added the reaction mixture had gone from colourless to dark orange brown and the temperature had risen to +7° C. The reaction mixture was returned to the ice chest for about 25 minutes to drop the temperature to -5° C and the addition continued. Only a very slight temperature rise now occurred and it is noticeable that almost immediately that a total of 6ml (1 mol. equiv.) of nitrite had been added the solution began to effervesce giving off nitrogen oxide (brownish).

The reaction mixture was allowed to stand overnight to complete the reaction and warm up to room temperature, when a brown precipitate formed. 10ml of 40% NaOH was added to neutralise part of the acetic acid and the slurry and diluted to twice its volume with water and chilled in the fridge for 1 hour and then filtered. The cake was washed with a little (2-3ml) cold water and dried, the dried product weight 1.833g and was clearly a mixture of two substances, one deep yellow spherical aggregates in a very dark brown amorphous material. A further 3ml of 40% NaOH solution were added to the filtrate and the mixture extracted twice with 30ml quantities of ether. The ether was evaporated to leave a further 0.636g of orange coloured crystals mixed with a reddish brown tarry material. The combine precipitate and extracts were mixed and recrystallized from 40ml of 75% aqueous methanol with the addition of 0.2g of decolourising carbon. It was filtered through a preheated 42.5mm Buchner funnel and cooled to 5° C and the brownish felted crystals recovered by filtration. Much material appears to remain in the filtrate and work on a better recovery method is ongoing. The filter cake was dried overnight at 30° C and weighed 0.567g of orange brown prisms.

Discussion
Even with a threefold excess of sodium nitrite only about 4.3ml of 80% acetic acid are actually require by the reaction the remainder is merely solvent. It may therefore be possible to reduce the overall volume and acetic acid concentration by adding a little methanol as a co-solvent and reducing the acetic acid greatly to say 5ml or less. The dilution used above was used because only a very slight precipitate formed spontaneously, probably because strong acetic acid is a good solvent for polar organic compounds. If this results in complete solution it would be advantageous to use less and stronger hydrogen peroxide to minimise dilution of the reaction mixture. Furthermore the effervescence that occurs after the addition of just 1 mol. Equiv. of sodium nitrite solution suggest to me that with hydrogen peroxide as an oxidant only one molar equiv. of sodium nitrite should be required.

I plan to run several further experiments using only 5ml of 80% acetic acid 6ml or so of methanol and an appropriate volume of 12% hydrogen peroxide. To these mixture I plan to add only about 1 to 2 molar equivalences of sodium nitrite as a near saturated aqueous solution.

A further issue that requires investigation is whether or not the use of a threefold excess of nitrite in the presence of hydrogen peroxide can give rise to a dinitration product or just tarry products. From an examination of the various solid product it is clear that they are heterogeneous.

Work is also in progress on the best work-up protocol.

[Edited on 1-7-2015 by Boffis]

Maybe? you are assuming hydrogen peroxide is necessary for formation of the N-acetyl-p-benzoquinone imine intermediate, but that could come about by oxidation by the nitrite. I think it is the excess of nitrite that drives the reaction to completion forming the 3-nitro-4-acetylaminophenol as the ultimate main product, blowing right past the intermediates produced at lesser concentrations of nitrite as would exist under in vivo reaction conditions of the stomach where the reaction proceeds only part way to completion, producing the intermediate benzoquinone instead.

Aqueous acetic acid at pH 4 is 0.00066 molar which is some really dilute vinegar. so waving a pickle over the beaker while breathing across it gently might be enough to do it.

Seriously what may work fine is try an equimolar amount of acetic acid equivalent to the total NaNO2 plus maybe 2% in excess of theory needed to convert all the NaNO2 to NaOAc and call that good.

I am guessing that is the adjusted pH for the reaction system with the sodium nitrite already added, since the sodium nitrite dissolves and forms a distinctly basic solution, there must be a titration with acetic acid to lower the pH back to neutral and beyond that to a slight acidity at pH 4. It would have to work that way I think and a sodium acetate / acetic acid buffer is being created in situ by that titration. They couldn't make it easy and just state the quantities and volumes used. The maximum activity for the nitrosation is reported by others for similar reactions for the pH range about 5 and I think pH rising to around 5.5 the activity quenches entirely, so the pH ranging 4 to 5 seems about right. It can maybe be a pH target range for the pH maintained by buffering as is suggested for the use of a phosphate buffer as the alternative reaction system. I believe the lower pH system at about pH 4 to maximum pH 5 would be a faster completing reaction with the higher yield.

If I understand the meaning correctly 7 moles of NaNO2 were used per 1 mole of paracetamol to obtain the 3-nitro-4-acetylaminophenol. 1 mole HNO2 is needed to produce the N-acetyl-p-benzoquinone imine plus 1 more mole to convert the N-acetyl-p-benzoquinone imine to the 3-nitro-4-acetylaminophenol, plus 5 moles NaNO2 excess was used, or 350% amount of theory of the 2 moles NaNO2 per 1 mole of acetaminophen. That 7 to 1 molar ratio works out as 3.2 grams NaNO2 per each 1 gram paracetamol for the nitration. 102% of the theoretical amount of acetic acid for neutralization of that Na value would be 2.84 grams acetic acid per 3.2 grams NaNO2. The buffer calculations for the resulting byproduct sodium acetate solution will need to be done to calculate the needed additional acetic acid for buffering the spent nitration mixture at the pH 4 or in the pH 4 to pH 5 range. That should put things in the ball park for what reaction conditions the Japanese chemists neglected to provide details. Need to look at the solubilities and see what seems reasonable for a reaction volume also.

The only way I can see the described reaction condition happening is to have a stirred slurry of paracetamol in an acetic acid / sodium acetate buffer at pH 4, in an ice bath, to which is added in separate streams, equal drip rates of NaNO2 solution as reactant [A] and a calculated concentration and equal volume of acetic acid as reactant [B] This scheme could maintain the pH 4 for the reaction as paracetamol in solution reacted with the 2 incoming reactant streams. As dissolved paracetamol is nitrated and drops out of solution, fresh paracetamol from the slurry will enter solution and react, until the reaction is complete.

Aqueous acetic acid at pH 4 or a phosphate buffer solution at pH 7 is a kind of indefinite description of the reaction volume and reactant concentration which would require some reverse engineering to solve that cryptic description.

Here is the prior article which led to the article you reference and is reference 4

Formation of N-acetyl-p-benzoquinone imine, the well-known toxic metabolite of acetaminophen, by the reaction of acetaminophen with nitrite under model stomach conditions.

Takafumi Ohta, Hiroaki Oribe, Manabu Ide, Shoji Takitani

https://www.jstage.jst.go.jp/article/cpb1958/36/11/36_11_463...

Abstract:

The main reaction product of acetaminophen (4mM) with nitrite (1mM) under model stomach conditions was identified as p-benzoquinone. HPLC analysis of the reaction mixture with an electrochemical detector revealed the presence of N-acetyl-p-benzoquinone imine, the toxic metabolite of acetaminophen, as an intermediate in p-benzoquinone formation. About 14% of acetaminophen was estimated to be transformed into this intermediate when the reaction was carried out at pH 3 for 1 h at 37°C.

and here is the Journal page

https://www.jstage.jst.go.jp/browse/cpb/36/11/_contents?from...

Attachment: Chemical and Pharmaceutical Buletin Vol36 1988.pdf (501kB)
This file has been downloaded 650 times

What is going to be the potential "real bite" here is if the Japanese have misidentified a nitroso for a nitro.
3-nitroso instead of 3-nitro
would not make me a happy camper

If indeed it does turn out to be a 3-nitroso then it may possibly oxidize to the nitro by warming with dilute HNO3

[Edited on 7/2/2015 by Rosco Bodine]

I added this as an edit to my earlier post before I realized you had replied here on a new page so I'll say what I was thinking again here.

The only way I can see the described reaction condition happening is to have a stirred slurry of paracetamol in an acetic acid / sodium acetate buffer at pH 4, in an ice bath, to which is added in separate streams, equal drip rates of NaNO2 solution as reactant [A] and a calculated concentration and equal volume of acetic acid as reactant [B] This scheme could maintain the pH 4 for the reaction as paracetamol in solution reacted with the 2 incoming reactant streams. As dissolved paracetamol is nitrated and drops out of solution, fresh paracetamol from the slurry will enter solution and react, until the reaction is complete.

Calculating the solution strengths to have the reaction begin and end buffered at pH 4 is the math that needs to be done.
Actually I think the reaction may even run faster at pH 4.5 to 4.75

If it works the way I am thinking it does, I think I know how to do this buffer scheme. You make up your buffer solution of sodium acetate and acetic acid for slurrying the paracetamol.
It should be a fairly strong buffer solution at maybe 90 to 95% saturation at 0C.

That will give you the model composition which will be the same ratio of sodium value and acetic acid value as will be needed to formulate the equal volumes of NaNO2 solution [A] and its simultaneously added reactant [B] which is acetic acid equivalent to neutralize the sodium value of [A] plus or minus the amount of acetic acid required to maintain the composition of the buffer solution throughout the reaction. When all the nitrous acid has dissipated and all the reactions are complete the pH should be sitting at pH 4 at endpoint right where it started for the slurry before additions of reactants began.

It is a constant reaction condition scheme, a continuous reaction scheme.

I would make the buffer solution and reactant [A] & [B] buffer precursor solutions almost as strong as possible like about 90 to 95% of saturation at 0C for the "model" solution and its maintained constant "byproduct" composition of increasing volume, but kept at a constant near saturation concentration by the incoming reactants [A] and [B].

The common ion effect is going to come into play limiting the concentration of an acetate buffer at 0C due to the lowered solubility of the sodium value. And I'm not sure I am reading these ancient charts correctly.

But it looks like a usable buffer "model" pH 4 buffer composition could be 20.4% acetic acid / 8.2% sodium acetate trihydrate / 71.4% H2O

Then basically you calculate the equal reactant volumes for [A] and for [B] which add to form an ultimate reaction product matching that composition, so that when all the reactions are completed the supernatant liquid has the same composition.

Bromocresol green indicator is probably about right, keeping toward the yellow range during the reaction.

I will complete the calculations for the reactants and post the values later.

The following is a kind of educated guess and reverse engineering attempt for the abbreviated and summary reaction description provided by the Chem. Pharm. Bull. v37(5), 1422-1423; 1989 article attached on the preceding page and is the same article as was a screen shot of the reaction posted by Boffis. I believe it is likely the Japanese chemists used a reaction scheme closely analogous to the following:

Okay I have some preliminary calculations done well enough to see that for 10 grams of paracetamol to be nitrated is going to have a finish reaction volume of about 800 ml and given the likelihood of foaming this size batch should be run in a two liter beaker. It may be possible to compress this reaction volume using stronger solutions. I tried to stay well away from the saturation point on the reacting solutions so byproduct salts would not precipitate and make any foaming more complicated to manage.

For 10 grams of paracetamol to be nitrated

Here is the reactant specification for [A] 32 grams NaNO2 + 367.7 ml H2O = calculated total volume 382.5 ml

Here is the reactant specification for [B] 184.8 grams acetic acid + 206.5 ml H2O = calculated total volume 382.5 ml

The volumes should be equal for [A] and [B] or very close. Any possible small effect of solution compression or density differences between 20C and 0C was not factored. If the specification calculation based on density is slightly off then it can be adjusted with H2O to make the volumes exactly equal.

The initial slurry of paracetamol should be made up by adding just enough of the "model" pH 4 buffer solution made separately, which is a composition of 20.4% acetic acid / 8.2% sodium acetate trihydrate / 71.4% H2O. Add just enough of this solution to make the paracetamol a thin stirrable paste or thin slurry suspension.

At 0C with stirring begin simultaneous additions at equal rates [A] and [B] and the reaction should proceed buffered at pH 4.

Bromocresol green should work as an indicator to monitor pH, being kept towards yellow for the reaction to proceed. The end product is IIRC yellow and the expected quench pH possibly about pH 5.5 due to sudden reduction of the active evolution of HNO2. The mix of color in the optimum range is likely then to be a yellowish green with the quench arising in the really solid green that would come with rising pH in the likely pH 5.1 to pH 5.5 range with the peak HNO2 activity expected at about pH 4.8 to pH 5. Anyway, the Japanese chemists specified pH 4 so pH 4 is what I modeled, just in case they are right on the bullseye. It could be interesting to try an alternate model with the buffer at pH 4.8 to see if yield goes up or down.



There should be adequate solubility for the paracetamol slurried in the 20.4% acetic acid / 8.2% sodium acetate trihydrate "model" buffer reaction medium for it to very gradually dissolve enough to react incrementally in the contemplated reaction scheme.

How cleanly will be the separation of the end product is unknown, and is dependent upon the solubility in the spent reaction mixture. Neutralzing and salting out may be required, and / or evaporation possibly vacuum evaporation at reduced temperature to facilitate recovery of the end product. Or an extraction using a solvent, possibly methylene chloride or toluene may be required.

Attachment: Sodium Acetate Solubility Seidell.pdf (111kB)
This file has been downloaded 658 times

[Edited on 7/2/2015 by Rosco Bodine]

Quote: Originally posted by Nicodem

, you will still have to deaminate by nitrosating a p-hydroxyaniline which I do not think it is possible without ending with the corresponding p-benzoquinone (I could be wrong though, but not motivated enough to check the literature).

Quote: Originally posted by benzylchloride1

Direct diazotization of p-aminophenol leads to a mess of quinone

That's an interesting route. I never considered that, so I can't really comment.

I used to be interested in resorcinol many years ago. I forget exactly why, but nevermind that. The best route I could come up with while be relatively possible for an amateur was not a clean one.



Starts with aniline then
1) EAS with Br₂. No clue how well the selectivity for the di-ortho (relative the amine) would be. Pretty sure aniline is so activated here that this would go without catalysis. If Para was preferred then huge blunder.
Edit-
Come to think of it this would likely fail to give any selectivity for the di-product especially not the right regioisomers unless the para position was blocked. Woops.

2) Then the classy sandmeyer elimination type reaction which has been discussed here several times in several ways. Again, messy, but doable.

3) Dihydroxylation reaction not sure how clean that'd go either. I know some users have reported hardship with similar reactions.

[Edited on 5-7-2015 by smaerd]

Maybe there's a selective way to nitrate an acetanilide amide for the meta regioisomer. Or go for dinitrobenzene...

The real difficulty is just meta positioning.

edit again - just checked the wiki. It says melts of sulphonic acids are typically used. That's definitely approachable.

[Edited on 5-7-2015 by smaerd]

@Gdflp - I didn't know that. I always assumed it would nitrate ortho/para but now that you explained it that makes sense. Hmmm..


Here's a scheme that should actually be valid. Only missing one reference.



1. Start with benzene and sulfonate as usual with H2SO4.

2. Create the Barium salt with BaOH or similar.

3. Reflux with Conc Sulfuric acid at 209*C ~4Hr's to create the 1-3 disulphonate.
Benzenedisulfonic Acid from Benzenemonosulfonic. C. E. Senseman. Ind. Eng. Chem., 1921, 13 (12), pp 1124–1126. DOI: 10.1021/ie50144a012 (http://pubs.acs.org/doi/abs/10.1021/ie50144a012?journalCode=...)

4. Melt in potassium carbonate (missing ref?) Meyer, J (1897). Ber 30: 2569

Probably not cheap but it seems like there would be appreciable control in the intermediates synthesis (no regioisomer hell). Pretty simple chemistry. Not sure what temp the melt is performed at though.

[Edited on 5-7-2015 by smaerd]

@Rosco and Boffis,
The nitrosation of paracetamol (para-acetaminophenol) is a stange story!
In books I see that:
-sometimes nitrous acid provoques the deacetylation and diazotation of the amino group
-sometimes there is reference of acetanilide being nitrosated as N-nitroso-acetanilide (isolable) wich upon basification (KOH) turns into K acetate and K benzodiazotate.
-phenols tends to nitrosate in para to the hydroxy group, but if not accessible then it goes to ortho position to the hydroxy group...why then the N=O in position 3 instead of 2...does it come from a transcient N-Nitroso-p-acetamino-phenol and vicinal deplacement at the closest position?
-nitrosoaromatic may oxydise with HNO2 into nitroaromatics

About bromocresol green...is it stable towards nitrous acid?
The colour shift is interesting but if you form a yellow compound in the blue shift region the total solution will look green giving misleading info about pH.
If you form a brown compound the colour will also be hard to figure...

Last but not least...paracetamol will also make a buffer arround pH 9.3 +/-1 and the resulting 3-nitroparacetamol will add complexity with a third buffer arround pH 8.6 +/-1...
So wich one will win? Acetic acid/acetate, p-acetaminophenol/p-acetaminophenate or 3-nitro-p-acetaminophenol/3-nitro-p-acetaminophenate?

I think you are only at the start of a very hard work

This may help as it is related (see S11 to S20)

Attachment: Functionalized alkoxy arene diazonium salts from Paracetamol.pdf (5.3MB)
This file has been downloaded 778 times

[Edited on 5-7-2015 by PHILOU Zrealone]

@Rosco. I’ve re-read your post above and thought about this for some time. I was not really happy with my own initial attempt, however, a few points are worth making before I discuss your proposals. In the original Japanese paper they stated a 3 to 5-fold excess of nitrous acid between pH 3 and 7. I made no attempt to buffer the reaction mixture, the large amount of acetic acid was merely to act as a solvent for the paracetamol but the pH would still have been >3 and so within the lower limit, admittedly only just. I agree that a buffer to give a pH closer to the mid-range would have been better but this was intended to be the first of a series of experiments. Unfortunately I have been sent overseas again and will not be home for another 6 weeks.

Thinking about things again the large excess on nitrous acid make no sense. My experience of phenolic nitrosation reactions is that they are very rapid and the subsequent oxidation is slow. So with the addition of a more potent oxidizing agent the addition of more than one equivalence of sodium nitrite is un-necessary. In fact my experience is that excess nitrous acid almost always results in the generation of much tarry material in any reaction and my experience, as posted above, with this reaction seems to bear that out.

Looking at your proposals I could carry out these reaction as prescribed but I have a few issues:
1) Even without the addition of hydrogen peroxide your sodium nitrite addition seems vastly excessive. For 10g of paracetamol a threefold excess would be only 13.7g and a fivefold excess 22.8g or were you working on the 7-fold equivalence (which I don’t understand) you mentioned about 3 post back? In the original paper they state that the yield of 22% was obtained with the excess of nitrous acid and with hydrogen peroxide the yield was 94% but now they don’t state that they used an excess of nitrous acid at all and I believe they didn’t and were being deliberately obscure. The authors merely state that the nitration is “greatly facilitated by treatment with equimolecular amounts of hydrogen peroxide…”.

2) Given that any concentration of acetic acid will give you pH of 3-7 I think we are going overboard on pH buffering. I think that to use a suspension of finely ground paracetamol as you suggest or with a small volume of co-solvent such as methanol or DMSO it will be sufficient to add the sodium nitrite and acetic acid with a slight excess of the latter. A small quantity of acetic acid may be added to the paracetamol solution/suspension just to ensure a starting pH of <7. I am always wary of swamping the reaction with dissolved salts unnecessarily.

My plan is to start out with a 1:1 ration of paracetamol and nitrous acid at pH >4 and see if I can isolate the 3-nitroso compound and then see if I can oxidise this with hydrogen peroxide to the nitro compound. I will then repeat the experiment changing the various parameters until I get products that appear reasonable. As I stated in my initial experimental details the initial precipitate contained both a yellow and a brown product while the subsequent ether extraction was crystalline and orange with a little tarry material so it looks like I got a mixture of products but the nitro compound is most likely the yellow one as the nitroso compounds are usually brown.

By the way I found a paper that describes the assay of paracetamol in blood serum using the nitration with nitrate and they claim that the product is the 2-nitro compound but they carry out the nitration in trichloroacetic acid-hydrochloric acid so probably lower pH but the important thing is that the reaction appears almost instantaneous.

The next question is how do we know what we have got? My plan now is to try the reaction above with only one equivalence of nitrous acid and then try reducing the nitroso compound to and amine. Then without de-acetylation attempt to diazotize the product with one equivalence of nitrous acid to see if I can produce 1-acetyl-5-hydroxybenzotriazole directly. I have a paper (somewhere) that gives the properties of the 5-hydroxybenzotriazole.

I also intend to try the di-acetylation with acetic anhydride and then nitrate to provide material for comparison.

I also intend to test the various products with either 4-amino-dimethylaniline or sodium sulphanilate. Many aromatic nitroso compounds condense with these amines to give azo dyes while nitro compounds do not normally react. Possible removal of excess nitrous acid may be necessary first with urea or similar. I am going to investigate the product from the initial experiment with chromatography (TLC on silica 60 plates any ideas about solvents?) to see how many compounds are present. Another test might be ferric chloride which gives a deep green colour with most 2-nitrosophenols so this test would be a negative test in that the 3-nitroso compound should not.

@smaerd, I don't think that brominating aniline would work like that. I have a reference somewhere to bromination aniline with N-bromosuccamide to avoid oxidation and side reaction IIRC.

@Philou, Nitrosation of paracetamol strange! that's an understatement. Part of the problem though is some of the papers had only a passing interest in the outcome so didn't persue the product while others are just vague.

Yes I think the colour of the reaction mixture will mask the colour of bromocresol green if used in the reaction mixture (and survives) I would use the dye to prepare paper strips. Actually all of the reactive (ie phenolic) o and p sites in bromocresol green have already been blocked so the compound may survive under the mildly acid conditions, it depends on its ability to resist oxidation.

And, as for all those buffers; my brain hurts. I think I stick with my system given above because the addition of sodium nitrite will have its own buffering effect too.

Quote: Originally posted by Boffis

@Rosco. I’ve re-read your post above and thought about this for some time. I was not really happy with my own initial attempt, however, a few points are worth making before I discuss your proposals. In the original Japanese paper they stated a 3 to 5-fold excess of nitrous acid between pH 3 and 7.

See page 10 of the pdf attached, page 442 of the journal article. The melting point of 218C is different from the 139C reported by Matsuno et al so this is more evidence they have the 3-nitroso rather than the 3-nitro compound.

If the F. 218 identified by Reverdin is fahrenheit, there is still a discrepancy because that is 103 C which is not the 139 C given by Matsuno. There is a problem there with the mp either way.

You and I have found errors before in the literature, haven't we? Keeping all those phd's honest is becoming tedious work. Maybe they suffer from confirmation bias writ large.

Quote:

I also intend to test the various products with either 4-amino-dimethylaniline or sodium sulphanilate. Many aromatic nitroso compounds condense with these amines to give azo dyes while nitro compounds do not normally react. Possible removal of excess nitrous acid may be necessary first with urea or similar. I am going to investigate the product from the initial experiment with chromatography (TLC on silica 60 plates any ideas about solvents?) to see how many compounds are present. Another test might be ferric chloride which gives a deep green colour with most 2-nitrosophenols so this test would be a negative test in that the 3-nitroso compound should not. Yes I think the colour of the reaction mixture will mask the colour of bromocresol green if used in the reaction mixture (and survives) I would use the dye to prepare paper strips. Actually all of the reactive (ie phenolic) o and p sites in bromocresol green have already been blocked so the compound may survive under the mildly acid conditions, it depends on its ability to resist oxidation. And, as for all those buffers; my brain hurts. I think I stick with my system given above because the addition of sodium nitrite will have its own buffering effect too.

Hey the buffers are easy at the pKa they all work the same

And a java applet can do the rest.
http://clymer.altervista.org/buffers/acetic.html

Here's another article attached that shows H2O2 used to oxidize dinitrosoresorcinol probaly the 2,4-dinitroso to dinitroresorcinol See page 8 of the pdf

There is a patent for this too I think. The reference for the H2O2 use is not stated or cited, so I'll have to look for it.

You don't want to use methyl alcohol for a cosolvent because of easy formation of methyl nitrite which is like ether or butane in volatility and will escape with loss of nitrogen from the reaction system. If methanol was used as a cosolvent by Matsuno that could easily explain the need for a 5 molar excess because most of the needed nitrogen in the reaction mixture just like Elvis had left the building.

Solubility for APAP at OC is .72 grams in 100 ml H2O so really a cosolvent is probably not needed, and even if DMSO was added it wouldn't take very much maybe 2% DMSO to triple the water solubility and that solubility could be the reaction rate governor. However there is a possibility that changing the solvent system could also change the product. For example when resorcinol is treated with nitrite in aqueous medium a nitroso derivative results. But when treated in solution in ether a diazo derivative results. A similar difference could occur by changing the solvent system with APAP.

Matsuno cited these articles (reference 7) with regards to the imine being isolated and reacted separately

http://pubs.acs.org/doi/abs/10.1021/jm00350a001

http://www.sciencedirect.com/science/article/pii/S0040403900...

Here is another article that could be interesting

http://pubs.acs.org/doi/abs/10.1021/ja01659a030

Attachment: Reverdin and Dresel, Archives des sciences physiques et naturelles, 1904, 18, 442.pdf (392kB)
This file has been downloaded 633 times

Attachment: AD0399653.pdf (1.8MB)
This file has been downloaded 672 times

WRT the bromocresol green if it does survive the reaction system then yes I knew the green range would likely be obscured, but thought the yellow would probably work as the marker. What will be the color development for the reaction mixture is an unknown that could defy any use of any indicator dye and it could require monitoring with a meter. Also possible is that any buffer system would be overloaded but that would tend to not make sense for reason that use of a buffered reaction system was specified by Matsuno. To specify a buffered system if a buffer wasn't even used would be misleading more than simply being deliberately obscure and would be unethical and dishonest.

As a note on the subject of pH indicator dyes, I ran across this interesting bit about lacmoid or lackmoid (resorcinol blue) pH indicator



Attachment: lackmoid pH indicator Bulletin_of_Pharmacy Vol2 1888 pg286.pdf (278kB)
This file has been downloaded 594 times

[Edited on 7/6/2015 by Rosco Bodine]

Quote: Originally posted by PHILOU Zrealone

The total reduction of the DNB into 1,3-diaminobenzene is possible and bis-diazotation also following certain rules to avoid copulation of the diazocompound with the diamine to form Bismark Brown.

I later realised that it would be better to do a double diazotisation on 1,3-diaminobenzene instead of doing it one by one because diazo reactions are very messy.Mix the diamino with NaNO₂+HCl in a sep funnel and drip it into a beaker of boiling water with a few drops of acid.
can you tell more about the rules on bismark brown .

Evidence accumulates that the alleged 3-nitro-acetaminophen of Mansunto et al is NOT the 3-nitro but is almost certainly the 3-nitroso-acetaminophen which would have reduced to the same diamino compound which was the basis for identification of the likely precursor. Here is an abstract from the time when different identifications of bona fide 3-nitro-acetaminophen were made by different synthetic methods and the mp identified at 218C



Attachment: Spectrophotometric determination of acetaminophen and salicylamide through nitrosation and subsequent chelation.pdf (167kB)
This file has been downloaded 952 times

See also the attached article published 10 years before the Mansunto et all article which describes a nitrosation of acetaminophen under alkaline conditions as producing a C-nitroso derivative at the 2 position with concurrent dehydration of the phenol to the quinone.

From the references which have been reviewed thus far, it appears to me that the nitrosation or "nitritation" of acetaminophen may proceed different ways, depending upon the pH and probably the temperature also, so that the result may be a conversion of the hydroxyl to the oxime, or an alternate outcome may be a C-nitroso at position 2,(possibly an oxime interactive with the phenol hydroxyl), or a third outcome may be a C-nitroso at position 3,(possibly an oxime interactive with the N-acetyl or with the phenol hydroxyl), and I think a fourth possible outcome may involve a reaction with the 4 position N-acetyl group,(possibly an oxime interactive with the phenol hydroxyl). And while those would seem to be the main 4 possible reactions, there doesn't seem to be any clear mapping of the reaction paths and outcomes ever charted as a kind of summary in any publication describing the synthetic conditions and the various yields and byproducts. In their totality it appears these various reactions are simply not yet charted, thus we are in "uncharted territory" .....like going where none have gone before Who would have guessed

Anyway, sufficient to say that Mansunto et al, have just been sent back to school. And if the structural identifications have been based upon NMR "library" data then obviously the NMR library data is WRONG also.

Nicodem I think recognized and alluded to this possibility when making note earlier that reduction of a nitroso or a nitro could lead to the same amino as the result, so it is not conclusive to identify a nitro as positive identification of what was the actual precursor for the amino. It is a coin toss 50/50 chance it was one or the other, and in this case Mansunto et al, simply guessed WRONG on the coin toss and did not follow up to positively confirm which precursor for the amine had actually been reduced. Evidently it was indeed the 3-nitroso and not the 3-nitro as they presumed.

A classic example would be the case of aminoguandine, where either nitroguanidine or nitrosoguanidine are both reduced by hydrazine to the same aminoguanidine. The nitroso is an actual identified intermediate in the case where the nitro is being reduced to the amino compound. A similar scenario is occurring here with the Mansunto et al article. They presumed the precursor for the amine would have been the nitro being reduced to the amine, but it was actually the nitroso which they reduced to the amine.

This is a discovery I had actually anticipated when I posted my thoughts earlier

It's deja vu all over again I was on the bullseye.

Can't fool old Rosco Trying to pass off a nitroso as a nitro ....shame ....shame

Mansunto et al ........You are so busted!!!

You have the right to remain silent, anything you say can and will be used against you in a court of peer review If you cannot afford a chemist .....one will be appointed for you

https://www.youtube.com/watch?v=ja0_m-4NAec

<iframe sandbox width="640" height="480" src="https://www.youtube.com/embed/ja0_m-4NAec?rel=0" frameborder="0" allowfullscreen></iframe>

[Edited on 7/11/2015 by Rosco Bodine]

Victory lap ^^^^^^^^^^

Quote: Originally posted by Boffis

Hi Rosco, I am going to have another go at this reaction but my paln is to suspend the the paracetamol, sodium nitrite and sodium acetate with and without H2O2 in a mixture of methanol and water and then slowly add slightly more acetic acid than required to liberate the nitrous acid. This should keep the pH fairly high, close to 7 to start with. My theory is that if the reaction is too slow at 7pH then the free acids will cause the pH to fall until reaction occurs. I am working on the quantities at present.