DDNP & related compounds: The über thread!

Quote: Originally posted by nitro-genes

Thanks Philou, it seems the melting point is in range of what I found for the chlorinated benzoxazolone. Read something about it's use as muscle relaxant, though I'm not inclined to eat anything I produce in my shed. Ok, here is something strange... This is based on 1 rushed experiment, it would need further examination, but I'll add it here since I don't store these things on my computer. I was bored and had a lot of purified isopicramic acid left and decided to look at what exactly forms during nitration using 97% SA and a nitrate salt. To a 20 ml beaker, 10 grams of 97% SA was added and 0.5 grams purified isopicramic acid added and stirred at 20 deg C until everything dissolved, producing an almost red/black but transparent solution. Next, it was added to an icebath and at 0 deg C, 0.25 grams of KNO3 (~1 mol eqvt) was added. The solution was allowed to stir and no gas evolution was seen. Gradually, the solution took on a much lighter red transparent colour, indicating something was happening. A small sample withdrawn at this stage dissolved in icecold water completely, not a single bubble of gas was produced. Since this could be exlained by the H2SO4 adduct with isopicramic itself, I decided to add another mole equivalent of KNO3 and very soon, gas formation became evident. It was allowed to stir overnight in the icebath, going from 0-10 degrees, with steady evolution of gas. When water was added in the morning, copious amounts of NOx were liberated and a transpararent yellow/orange solution was left. No precipitate occured when kept at 4 deg C overnight. About 3 ml's of ethyl acetate were added and briefly stirred, upon which almost all of the colour transfered to the organic phase. This was siphoned off and allowed to evaporate. A small amount of a yellow/orange crystalline precipitate formed, that burned very characteristic for a diazonium compound, very vigorous (more than p-DDNP) and with yellow flash. It dissolved very easily in water again, but adding a saturated KNO3 solution and chilling produced no precipitate. It further seems to attack metals like crazy, although this could also be due to some extracted acid by the ethylacetate. So, what formed here? Since it is water soluble, even after extraction, it is not p-DDNP iself, since the diazonium sulfate salt dissociates very quickly upon dilution. It still contains a diazonium group, but since it does not produce a precipitate with KNO3, it likely also isn;t DDNR, since the K-salt is reported to be very insoluble (DDNR istelf as well). One of the options is that a 1,2 quinone 3,6 dinitro 4-diazo is formed due to hydrolysis of the 2-nitro of isopicramic acid, though I'm not sure this would explain the reactivity towards metals. Another option is the 2,3,6 trinitro 4-diazo phenol descirbed before, though this would be strange considering the described deactivation of the amine group in 97% SA. Other IMO, less likely options would be some sulfonic acid replacement of the nitro, or more likely maybe some azoxy compound from coupling reactions. Any guesses, anyone? [Edited on 17-12-2016 by nitro-genes]

Answer is hard to give as usual

You have initially H2N-C6H2(NO2)2-OH probably as a sulfate.

Adding KNO3 results into liberation of some HNO3.
-->Part of the HNO3 may oxydise the para-amino-phenol into a para-quinon (mono-imine)...
-->Another part may nitrate further the dinitro compound into a trinitro one...or doubtfully to a tetranitro one
I don't think so, but, if it does, then you will have two NO2 on neightbourgs C (ortho vs each other and ortho vs the NH2 for one and vs the OH for the other)...and in such conformation; one of the two neightbourgs NO2 may be expandable/lost by hydrolysis afterwards and lead to even more oxydable molecules (dihydroxyaminobenzene and by further hydrolysis to trihydroxybenzenes (or trihydroxyaminobenzene to tetrahydroxybenzene).

Oxydation of the substrate by HNO3 is concomittant with reduction of HNO3 into NxOy fumes in what nitrosonium is present as NO(+)...a powerful nitrosating agent.
I have seen into a book about diazocompounds a process to make diazoniums starting from an aniline and plain HNO3 only; thus without any nitrite...about half of the HNO3 serves to make the oxydation and NO(+); the other half is used to make the salt diazonium nitrate.

Usually para- and ortho-aminophenol when diazotized makes an unsoluble internal diazonium-oxyde (diazo-phenate); but here maybe the H2SO4 keeps it solubilized? And then it reacts further with unoxydized p-amino-phenol molecule to make a triazene...or a quinone based Schiff's base.

The reaction scheme here-below is simplified since all ortho- and para-quinon (and imines) derived from transcient trinitro and tetranitro derivatives by hydrolysis/oxydation may also react
--> with the NO(+) to make diazoniums
--> or with the free p-amino-phenol molecule to make triazenes or quinon based Schiff's bases.



[Edited on 18-12-2016 by PHILOU Zrealone]

Best wishes for 2017 everyone! Thanks for that scheme Philou, there is probably a lot happening as indeed is usually the case.

So here are my further incoherent ramblings regarding the subject :

When the product after KN/SA nitration was extracted with ethylacetate, redissolved in a minimum of 80 deg C water, pH adjusted to 6 and KNO3 added, a light yellow-orange coloured compound precipitated, contaminated by a brownish-orange compound, which is most likely pDDNP. The light yellow-orange coloured material is incredibly explosive though, the tiniest specks detonate with considerable strength, much resembling a heavy metal azide. It concurs with desricptions of the potassium salt of DDNR, although a potassium salt of a 1,2 quinone diazide is also possible. Total yield was only about 100 mg or so, so for curiosity purpose only. Most of the ethylacetate extracted stuff is lost when redissolved, so I guess that the 1,2 and 1,4 quinones and accompanying diazo derivatives therefrom are likely far more water soluble than pDDNP and DDNR. The compound could also be a potassium salt of the triazene as indicated in your scheme, though I don't think this is likely under these extremely acidic conditions. Likely, the nitramine is formed first after which several things could happen. Also don't think direct NOx mediated oxidation to the quinone imine is likely under these acidic conditions. When I dissolved purified isopicramic acid in boiling (!) 30% sulfuric acid once, and added potassium nitrite in small portions, almost exclusively pure, crystalline pDDNP was produced as large coarse crystals. The reaction was much slower in this case, likely due to the fact that the anilinium ion is unreactive towards NOx, only the small amount of dissociated and free acid reacting. Then again, these temperatures would also lead to fast decomposition of the HNO2 and N2O3 produced, which would be the main diazotizing species in this case.

Still puzzled why heating 2,3,6-trinitro 4-aminophenol with 65% nitric produces DDNR selectively and in such good yields. Maybe this is a tradeoff which allows hydrolysis of the 3 nitro, enough water to allow diazotization by the NOx produced (or nitramine formation) while relatively disfavouring further nitration of the ring? It is strange though, both DDNR and 2,3,6-trinitro 4-aminophenol are reported to be reasonably stable in cold concentrated sulfuric acid (no hydrolysis there), so yields must be bad for other reasons, maybe using sulfuric acid changes the way the nitramine rearranges or reacts, though I would have guessed ring nitration (if possible) would be favored this way. I was unable to find how a diazonium salt or diazophenol behaves in concentrated sulfuric acid but maybe some side reactions occur here as well. Also curious whether using 100% SA or even oleum as opposed to 97% SA would have made any difference, or 2 mole equivalents of 100% HNO3 in an inert solvent or something. Or similar nitration using 1 mole eqvt of nitrate salt.

Below is 1 mg of the putative potassium salt of DDNR on aluminium foil (Dornier style). Pretty hot stuff, destroyed the rest...

Attachment: 1 mg Potassium salt of DiazoDiNitroResorcinol vid - Copy.avi (2.8MB)
This file has been downloaded 930 times

[Edited on 3-1-2017 by nitro-genes]

Quote: Originally posted by nitro-genes

Best wishes for 2017 everyone! Thanks for that scheme Philou, there is probably a lot happening as indeed is usually the case. So here are my further incoherent ramblings regarding the subject : When the product after KN/SA nitration was extracted with ethylacetate, redissolved in a minimum of 80 deg C water, pH adjusted to 6 and KNO3 added, a light yellow-orange coloured compound precipitated, contaminated by a brownish-orange compound, which is most likely pDDNP. The light yellow-orange coloured material is incredibly explosive though, the tiniest specks detonate with considerable strength, much resembling a heavy metal azide. It concurs with desricptions of the potassium salt of DDNR, although a potassium salt of a 1,2 quinone diazide is also possible. Total yield was only about 100 mg or so, so for curiosity purpose only. Most of the ethylacetate extracted stuff is lost when redissolved, so I guess that the 1,2 and 1,4 quinones and accompanying diazo derivatives therefrom are likely far more water soluble than pDDNP and DDNR. The compound could also be a potassium salt of the triazene as indicated in your scheme, though I don't think this is likely under these extremely acidic conditions. Likely, the nitramine is formed first after which several things could happen. Also don't think direct NOx mediated oxidation to the quinone imine is likely under these acidic conditions. When I dissolved purified isopicramic acid in boiling (!) 30% sulfuric acid once, and added potassium nitrite in small portions, almost exclusively pure, crystalline pDDNP was produced as large coarse crystals. The reaction was much slower in this case, likely due to the fact that the anilinium ion is unreactive towards NOx, only the small amount of dissociated and free acid reacting. Then again, these temperatures would also lead to fast decomposition of the HNO2 and N2O3 produced, which would be the main diazotizing species in this case. Still puzzled why heating 2,3,6-trinitro 4-aminophenol with 65% nitric produces DDNR selectively and in such good yields. Maybe this is a tradeoff which allows hydrolysis of the 3 nitro, enough water to allow diazotization by the NOx produced (or nitramine formation) while relatively disfavouring further nitration of the ring? It is strange though, both DDNR and 2,3,6-trinitro 4-aminophenol are reported to be reasonably stable in cold concentrated sulfuric acid (no hydrolysis there), so yields must be bad for other reasons, maybe using sulfuric acid changes the way the nitramine rearranges or reacts, though I would have guessed ring nitration (if possible) would be favored this way. I was unable to find how a diazonium salt or diazophenol behaves in concentrated sulfuric acid but maybe some side reactions occur here as well. Also curious whether using 100% SA or even oleum as opposed to 97% SA would have made any difference, or 2 mole equivalents of 100% HNO3 in an inert solvent or something. Or similar nitration using 1 mole eqvt of nitrate salt. Below is 1 mg of the putative potassium salt of DDNR on aluminium foil (Dornier style). Pretty hot stuff, destroyed the rest... [Edited on 3-1-2017 by nitro-genes]

@ Nitro-genes ... Beste wensen! Vroolijke nieuwe jaar!
@ The rest of the world ... May all your wishes come true! Happy new year!

Nice video-tje!

Still puzzled why heating 2,3,6-trinitro 4-aminophenol with 65% nitric produces DDNR selectively and in such good yields. Maybe this is a tradeoff which allows hydrolysis of the 3 nitro, enough water to allow diazotization by the NOx produced (or nitramine formation) while relatively disfavouring further nitration of the ring?
Probably by the reason I exposed about the making of diazonium nitrate from anilin and HNO3...but with one more subtility
The 2,3,6-trinitro-4-amino-phenol is first oxydised partially into 2,3,6-trinitro-4-imino-benzoquinone by HNO3 and generating subsequently HNO2.
HNO2 turns the rest of the 2,3,6-trinitro-4-amino-phenol into 2,3,6-trinitro-4-diazo-phenate.
The later 2,3,6-trinitro-4-diazo-phenate display synergetic effect of the diazonium (meta director) in position 4 and of the two nitro (meta director) from position 2 and 6...all working against the NO2 in position 3 (a bit the same as having 2,3,4,6-tetranitrophenol)...which rearranges into nitrite what is readilly hydrolyzed by the water arround and the boiling providing a -OH group and fresh N=O(+) as (HONO).

Ar-NO2 <--==> Ar-O-N=O (forced by ortho and para EWG groups)
Ar-O-N=O + H-OH <--==>Ar-O-H + HO-N=O
Thus the subtility here is that only a tiny amount of nitrite is needed to persue the full diazotation because it is regenerated by the hydrolysis step...and acts a bit like a catalyst.

For the rest you get more questions/interogations than answers...vive la science

Let's hope you "destroyed the rest" with the due timing arround midnight amongst the explosions of the new year fireworks .

[Edited on 3-1-2017 by PHILOU Zrealone]

Thanks Philou, that would indeed be as described by Meldola and reverdin and in later publication by Atkins and Wilson. That procedure you mention about making diazophenols from aniline would be interesting to read btw!

Out of curiosity I set up 2 new experiments involving the nitration of isopicramic acid both using 97% SA and KNO3:

To 2 separate 50 ml beakers were added 15 grams of 97% sulfuric acid and cooled to -20. To the first beaker was added 0.55 grams of KNO3 (1.1 mole eqvt) and to the other 1.1 grams of KNO3 (2.2 mole eqvt). Both were chilled to -20 deg C after which 1 gram of purified isopicramic acid was added with good stirring. No ice cooling was used, the reactions were allowed to stir at ambient temperature (~10 deg C.) for 12 hours in a water bath, after which about 15 m's of water were added. When water was added to the beaker containing 1.1 mole eqvt of KNO3, some NOx escaped and after a short while, crystalline pDDNP started to crystallize, with bright yellow supernatant.

When water was added to the beaker containing 2.2 mole eqvt of KNO3, lots of foaming occured, leaving a dark orange liquid, with no precipitate. The liquid was extracted using about 10 ml's of ethylacetate and added to a 100 ml beaker. The beaker containing the ethyl acetate extract was heated to 50 deg C. and allowed to evaporate nearly completely. During evaporation, a crystalline, bright yellow compound started to precipitate. Only when nearly all of the ethyl acetate was evaporated, an additional dark orange compound also seemed to precipitate. When all ethyl acetate was evaporated, 10 ml of cold water was added and heated with stirring to 80 deg C. When about 40-50 deg C was reached, the solution started to fizzle, NOx was released and a yellow-orange precipitate occurred, with dark orange supernatant. The contents were filtered and heated to 80 deg C. in 10 ml of fresh dH2O, 1 gram KNO3 was added and allowed to cool to 4 deg C. About 300 mg of a yellow-orange crystalline precipitate occured, behaving much like a heavy metal azide. After second thought, based on the colour and water solubility, I would say the compound is most likely the potassium salt of 2-oxide 3,6 dinitro 4-diazophenol.

Very curious what exactly the bright yellow compound after ethyl acetate evaporation is. It does seem to react with water releasing NOx and is highly soluble in ethyl acetate. The beaker containing 1.1 mole eqvt of KNO3 produced mostly pDDNP, this would be consistent with simple dissociation of a diazonium sulfate or hydrolysis of a nitramine , and not a trinitro compound though. This makes an interesting question why 2.2 mole eqvts of KNO3 produces a different product though. Either p-DDNP is nitrated further, or the primary nitramine is nitrated further, perhaps by NyOx present, perhpaps more like a radical nitration. Or the compound is a trinitro nitramine, which hydrolyses expelling one nitro group and leaving a diazo as has been shown before (nitration of 3-nitro aniline to tetranitronitranline --> O-DDNP), this would make the compound more likely to be DDNR though. If the compound would be a trinitro nitramine, it would be interesting to repeat this experiment using 100% sulfuric acid maybe or even oleum to prevent premature diazo formation and oxidation side reactions, very similar to why only oleum can be used to produce tetranitro aniline. If it is more like a radical nitration, less product should be obtained, or even only pDDNP maybe?

[Edited on 5-1-2017 by nitro-genes]

Very exiting, just had to make a post!

Assuming that the intermediate upon nitration of isopicramic is indeed a trinitrodiazoxy, and some intermediates may not be stable for prolonged time in sulfuric acid I changed the procedure somewhat:

10 grams of 97% sulfuric acid was cooled and 1.05 grams of KNO3 was added and stirred to dissolve. This was cooled to -20 and 1 gram of purified isopicramic acid was added at once. This was allowed to stir in an icebath for 4 hours, yielding a clear orange red solution, only slight gas formation occured. After 4 hours, about 0.5 ml of water was added while still in the icebath, NOx escaped and this mixture was stirred for another 30 minutes on ice to allow diazotization, untill a sample withdrawn produced a light yellow precipitate. Slowly, not letting temperature rise, more water was added and 3 ml ethylacetate to extract the dark orange stuff. To my delight, the bottom of the 20 ml beaker is now covered with a good layer of light yellow, almost transparent glittering crystals. Hopefully the ethylacetate didn't precipitate KHSO4 or something, or this is some differently coloured pDDNP crystal variant or pDDNP sulfate.

Calling Dr. Klapotke ...structural theory / resonance hybrid / NMR anomaly ahead....
time to get your freak on

Maybe we should drill down a bit further and see what we find huh



[Edited on 1/7/2017 by Rosco Bodine]

It'aint rocket science, but with the extended laws on prohibition of HE related chemicals over here (AN, conc. nitric, (per)chlorates, nitromethane etc), dumping some isopicramic acid in rootcleaner with KNO3 is at least still somewhat legal over here, besides I wanted to get rid of the isopicramic acid left as well. These are just not the times anymore for amateur chemistry, especially HE related... Maybe sometime again...thank you guys for all the help!

Last overview of the experiments done last couple of weeks:

0.5g pDDNP --> suspended/dissolved in 5 grams 97% SA, dark red black colour --> 4h 10 deg C. (no gas formation, no change in colour) --> Adding ice --> no gas formation, immediate dissociation (or precipiation) --> crystaline pDDNP

0.5g Isopicramic --> dissolved in 5 g 97% SA, transparent dark red colour --> 4h 10 deg C. (very little gas formation, no change in colour) --> Adding ice --> very little gas formation, no smell of NOx, yields transparent red solution with some minor dark red flocculent material.
Adding solution of nitrite yields crystaline pDDNP, with minor amounts of dark brown material

1 g Isopicramic --> dissolved in 10 g 97% SA + 2 mole eqvt of KNO3 --> 4h 10 deg C. (some gas formation, smell of NOx, colour changes from dark transparent red, to very dark red, to light orange red) --> adding water --> lots of gas formation, smell of NOx, upon cooling precipiation of ~0.3 grams of light yellow needle shaped crystaline material + small amount of dark orange/red ethyl acetate soluble stuff. The light yellow material is almost water insoluble and acidic. Adding bicarbonate and heating until near neutral dissolves most of the compound with dark yellow colour, seemingly precipitating some pDDNP similar material, over neutralizing leads to dark brown solution and no precipitation.

1g pDDNP --> suspended in 97% SA + 1 mole eqvt of KNO3 --> 4h 10 deg C. (some gas formation, colour changes from dark black-red to transparent red. --> adding water --> Some gas formation, smell of little NOx -> precipiation of large amount of unchanged pDDNP, dark yellow orange supernatant. Decanting and heating solution causes more NOx formation and precipitation of small amount of light yellow material. Light yellow compound is acidic in water and burns characteristic for diazo compound, leaving significantly more black residue though.

0.5g p-DDNP --> dissolved/suspended in 10g 97% SA + 2 mole eqvt of KNO3 --> heating (40-50 deg C) untill all dissolves --> very fast change to orange-yellow colour, very little gas formation, very faint smell of NOx --> 1h 40 deg C. no further colour change--> adding water --> very slight gas formation --> upon cooling, large amount of the light yellow crystalline precipitate.

Baffled regarding the identity of the light yellow compound. Difficult to tell, but during the reaction of pDDNP with KNO3, the reaction happens too fast and there seems too little N2 escaping to explain the formation of a triazene. The burn characteristics and extra amount of residue produced on burning would however be consistent with the triazene from Philou's scheme. Other options are a very differently coloured crystal variant of pDDNP (that can be reversed by boiling in water), or some really weird double salt /adduct/complex of p-DDNP with something. Either way, it seems that the small amount of the extremely explosive diazo compound obtained earlier occurs through direct ring nitration of isopicramic, which yield is probably only slightly increased when performing the reaction in 100% SA.

Attachment: Light yellow mystery compound - Copy.avi (1.3MB)
This file has been downloaded 896 times

[Edited on 8-1-2017 by nitro-genes]

I forgot one point that puzzle me about diazoniums...and that I should try one day.

Picric acid, trinitroanilin and related compounds (2,4-dinitrophenol pKa= 4,09) are acidic...so there is also a chance to make a diazonium picrate.

If one puts for example water solutions of picric acid, NaNO2 and dinitroanilin into a single recipient...one should see...
sodium picrate form but also a putative dinitrodiazonium picrate...
NaNO2 + HO-C6H2(NO2)3 --> NaO-C6H2(NO2)3 + HNO2
HNO2 + H2N-C6H3(NO2)2 --> HO-N=N-C6H3(NO2)2
(O2N)3C6H2-OH + HO-N=N-C6H3(NO2)2 --> (O2N)3C6H2-O-N=N-C6H3(NO2)2 + H2O
There is also a chance some extra picric acid stack onto it via complexation.

I haven't taken trinitroanilin into my example because it is following pKa tables more acidic than TNP.
But thinking about it, it would be worth to allow trinitroanilin to react with 1/2 equivalent of NaNO2...
H2N-C6H2(NO2)3 + NaNO2 --> NaHN-C6H2(NO2)3 (*) + HNO2 --> Na-O-N=N-C6H2(NO2)3 (**) + H2O
(O2N)3C6H2-NH2 + Na-O-N=N-C6H2(NO2)3 --> (O2N)3C6H2-NNa-N=N-C6H2(NO2)3 (***) + H2O
I wonder about the primary abilities and explosive properties of trinitroanilinates (*), trinitrodiazotates (**) or bis-trinitrophenyl-triazenates (***)...

On another hand 2,4,6-trinitroanilin when diazotized is apparently (from what I have understood in one of my readings) subject to hydrolysis of one of its ortho NO2 via nitro-nitrite rearrangement (thus substituted by a OH) and leads to another (iso-)o-DDNP variant (2-diazo-3,5-dinitro-phenol) to compare with conventional o-DDNP (2-diazo-4,6-dinitro-phenol)...so maybe no trinitroanilinates (*), trinitrodiazotates (**) or bis-trinitrophenyl-triazenates (***)...and only iso-ortho-DDNP.

So the DDNP tread is far from dead...stil a lot of research to do for all SM members

A diazonium picrate/2,6 dinitrophenate would certainly fit the bill, being titratable with weak base with recovering of some pDDNP. I must say, I've never seen any precipitation of picric using long nitration times of isopicramic though, puzzling...Would be easy enough to weigh the yellow product and pDDNP recovered and see what is present in the supernatant after dissociation. Picric acid, 2,6 dinitrophenol or perhaps KHSO4. Usually, more forcing conditions are needed to remove the diazo group IIRC, especially expect so for pDDNP, but maybe I'm wrong there and part of the diazonium is replaced by a nitro or eliminated. Such a diazonium salt would in any case be remarkable considering the presence of SA, which is far more acidic than even picric, so maybe it would be largely solubility driven or more like a complex?

Where did you read about the hydrolysis of the 2-nitro of picramide btw? Nitration of picramide using NA/SA yields nothing but starting material IIRC, and only under special conditions (Conc SA or PPA, or something) a diazonium salt can be obtained. Boiling in alcohol of the latter yields mostly trinitrobenzene from what I've read, reaction with water picric.

[Edited on 8-1-2017 by nitro-genes]

Quote: Originally posted by nitro-genes

Where did you read about the hydrolysis of the 2-nitro of picramide btw? Nitration of picramide using NA/SA yields nothing but starting material IIRC, and only under special conditions (Conc SA or PPA, or something) a diazonium salt can be obtained. Boiling in alcohol of the latter yields mostly trinitrobenzene from what I've read, reaction with water picric.

Into the following book:
Traité de chimie organique-Tome XV - Masson & Cie Editors
by G.Dupont, R.Locquin
Under the direction of V.Grinard and as secretary Paul Baud
Tittle:
Diazoïques et azoïques - triazènes, tetrazènes et pentazdiènes - azoxydérivés - hydrazines - hydroxylamines - oximes - amidoximes - azides
815 pages of pure delightfull informations all in french

Nice, sounds like an interesting book to read (a translation that is)! Curious what procedure they used going from TNA, could you give a short outline?

I did a quick search into my book and found the main reference to what I exposed, although it is not strictly speaking 2,4,6-trinitroanilin (picramide) but the reaction must be the same with it because it involves 2,4-dinitroanilin and 2-nitroanilins.Here follow some pictures of pages 129 to 138.

The information is not straight because there are a lot of condensed information into various places of the book and even if I'm french speaking, I have to read it twice or more to fully understand the potential of those condensed and generalised informations (and avoid misunderstandings)...too much scientific condensation may lead to this.
I will try to translate it.

Thanks for that, only scanned through the reaction schemes, translation would be appreciated! Interesting reactions, for the 1-amine-2-chloro-3 amine (bottom 6th document from top), and the topic of halogen migration, some very complex chemistry there. I think the position of the nitrogroups ortho/para repsective of the amino doesn't favor any hydrolysis/substutions and H-bonding of those aminohydrogens to the nitrogroups adds a lot of stability to TNA (and steric hindrence for amino), it's diazonium salt being only obtained with much difficulty and the diazogroup itself most liable to decomposition, being sandwiched between two very strong EWG groups. Would be interesting to hear your ideas though.

I was wrong about tetranitroaniline to o-DDNP conversion, it was 2,3,5 trinitroaniline shown to behave like that because of liable 2 nitro and (presumably, maybe?) absence of additional 6 nitro.

[Edited on 9-1-2017 by nitro-genes]

Quote: Originally posted by nitro-genes

Thanks for that, only scanned through the reaction schemes, translation would be appreciated! Interesting reactions, for the 1-amine-2-chloro-3 amine (bottom 6th document from top), and the topic of halogen migration, some very complex chemistry there. I think the position of the nitrogroups ortho/para repsective of the amino doesn't favor any hydrolysis/substutions and H-bonding of those aminohydrogens to the nitrogroups adds a lot of stability to TNA (and steric hindrence for amino), it's diazonium salt being only obtained with much difficulty and the diazogroup itself most liable to decomposition, being sandwiched between two very strong EWG groups. Would be interesting to hear your ideas though. [Edited on 9-1-2017 by nitro-genes]

Point 1 maybe due to the need for strong concentrated acids in order to dissolve all of the TNA as the anilinium ion maybe and keep all of the diazonium formed as an acid salt. Or maybe the diazotizing species is not nitrosylsulfuric, only from 80% sulfuric N2O3 is formed in significant amounts IIRC, and even then this may have a hard time reaching the amino group due to steric hindrance. On the other hand I've also read that this may require long reaction times for diazotization, since the aniliunium ion is unreactive towards diazotization. Don't know which is more correct or maybe both are correct. In any case, this may indeed be more related to the solubility of TNA, diazonium salt, stability and reaction kinietcs.

Point 3 is interesting, maybe the increased reactivity is explained also here by more ready dissociation of the diazonium salt during further reactions, which may not be the case for the solid salt at higher temperatures.

[EDIT] That reaction of the 2,4 dinitrodiazonium with NaOH (first page) is really interesting! Curious what conditions were used, as I would have expected the alkalinity required to hydrolyse one of the o-nitro's would also immediately destroy the formed diazophenol as well. Is that diazophenol much more stable? Did they slowly titrate the diazonium with base. Would sodium bicarbonate also work? The same reaction may be interesting for TNA and tetra nitroaniline as well, though my gut feeling would be no o-nitro hydrolysis there.

Another random thought would be that the obsered light yellow colour for the K-DDNR salt in one of the patents earlier posted might relate to the same light yellow precipitate containing pDDNP. Since they use KNO3-SA mediated diazotization, this would be consistent with an KHSO4 diazonium salt, which upon recrystallization would yield the brown dissociated K-DDNR again. Assuming no decomposition during recrystalization, this would make sense.

[Edited on 10-1-2017 by nitro-genes]

Here comes a fast translation of the previous 10 pages I have posted as images...took me more time because text treatment structure is lost when copy-pasting from word to the forum text protocol.

In red personnal comments or additions to make it more clear.

In green the page cuts.

In blue the provided examples in text instead of structural formulas.



----------------------------------------------------------------------------------
p 129

Elimination, substitution or migration of groups into diazonium compounds

Elimination or substitution

The transormation of the amine function into a diazo one impacts strongly the stability of groups present onto the aromatic ring.
We will see that electro-negative groups provide stability to diazonium salts of strong acids; but it must be noticed that, if those diazonium salts are treated by an alkaline agent or by an alkaline metal salt of a weak acid, then the electro-negative group becomes particularly labile; it is slit off and replaced by a -OH group, what with the diazonium function, will cyclize into a diazo-oxide.

Those substitutions can occure during the diazotation process itself when a weak acid is present, and in that case, one gets not the expected diazonium salt but the same diazo-oxide as one would get from performing diazotation onto an amino-phenol.

example:
1-diazonium 2,4-dinitro-benzen salt + NaOH provides the same as 2-amino-4-nitro-phenol + HNO2 = 2-diazo-4-nitro-phenoxide

Those group replacement phenomenons have been wel studied by Meldora and his students into the case of polynitroanisidins; we will spend a little time on those works to allow for a better understanding of the large number of cases that we will further examine.

Study of the polynitro-anisidins.

If we take a look at the different dinitro-anisidins, we notice that when diazotized in the presence of sulfuric acid, they yield the normal diazonium sulfate; but that when diazotized in the presence of chlorhydric acid, one of the NO2 groups in ortho vs the NH2 one is replaced by Cl; and finally when a weak acid is present, like acetic

----------------------------------------------------------------------------------
p 130-131

Table: Diazotation of some polynitro-anisidins in presence of a weak acid

para-anisidins and one p-amino-phenol (missing data 4-amino-3,5-dinitro-anisole)
Line 1: 4-amino-2,6-dinitro-anisole eliminates the methyl to yield para-DDNP (2,6-dinitro-4-diazo-phenoxide)
Line 2: 4-amino-2,5-dinitro-anisole eliminates the methyl to yield another para-DDNP (2,5-dinitro-4-diazo-phenoxide)
Line 5: 4-amino-2,3-dinitro-anisole eliminates the 3-NO2 to yield 6-diazo-2-nitro-3-methoxy-phenoxide
Line 7: 4-amino-2,3-dinitro-phenol makes another para-DDNP (2,3-dinitro-4-diazo-phenoxide)
Conclusion from line 5-7 -OH is leading over -OCH3

ortho-anisidins (missing data 5,6-dinitro; 3,5-dinitro; 4,6-dinitro and 3,6-dinitro)
Line 3: 2-amino-4,5-dinitro-anisole eliminates the 5-NO2 to yield 4-diazo-2-nitro-5-methoxy-phenoxide
Line 4: 2-amino-3,4-dinitro-anisole eliminates the 5-NO2 to yield 4-diazo-2-nitro-5-methoxy-phenoxide

meta-anisidin (missing data 2,4-dinitro; 2,6-dinitro; 2,5-dinitro; 4,5-dinitro and 5,6-dinitro)
Line 6: 3-amino-4,6-dinitro-anisole forms a normal diazonium…no elimination stange because 2,4-DNA does

----------------------------------------------------------------------------------
p 132

acid, then a methoxy group or a nitro group is replaced by -OH and subsequent formation of a diazo-oxyde ⁽¹⁾.

Meldora has established that to be labile, the group must be placed in ortho or para position of the diazotizable amine function but also that it must be kind of activated by the presence of a NO2 group in ortho position next to it ⁽²⁾. So a methoxy group -OCH3 in meta position vs the NH2 is not exchangeable even if next to a NO2 group.

In the case where two groups -OCH3 and -NO2 fulfill simultaneously those conditions; then it is the later that is favored for the splitting away.

Those reactions does not occure when the ring is holding a free -OH group in ortho or para of the function -NH2 ⁽³⁾; in that case indeed, the stable diazo-oxide form is generated without any elimination.

Those rules are demonstrated by a few examples gathered into table at page 130 (and 131).

Noticing that when the -NO2 group is split away it turns into nitrous acid form, Meldora was able to persue diazotation of those nitro-anisidins with about 1/4th of the theorical quantity of sodium nitrite normaly required ⁽⁴⁾; indeed this quantity starts the diazotation and the acid HNO2 generated by elimination of the -NO2 group, is enough to allow it to go on.

Elimination of the methoxy group.

Other cases than those of the polynitro-anisidins are known, in particular those of the nitro-anisic sulfonic acids.

Here also the -OCH3 group must be placed in ortho or para of the amine function and be activated by an ortho -NO2 group. It is the case to get to the following diazo-oxides ⁽⁵⁾:

2-amino-6-nitro-4-sulfonic-anisole turns into 2-diazo-6-nitro-4-sulfo-phenoxide
4-amino-6-nitro-2-sulfonic-anisole turns into 4-diazo-6-nitro-2-sulfo-phenoxide

Elimination of the nitro group.

In a more general way than for the polynitro-anisidins, amines that are nitrated ortho vs the NH2 give rise to elimination of the -NO2 group during diazotation if the para position is occupied by an electro-negative group.

Into the benzen family, 1-amino-2,4-dinitrobenzen ⁽⁶⁾, 1-amino

----------------------------------------------------------------------------------
p 133

-2-nitro-4-sulfo-benzen, 1-amino-2-nitro-4-arseno-benzen ⁽⁷⁾ , 1-amino-2-nitro-4-chloro-benzen ⁽⁸⁾, 1-amino-2-nitro-4-chloro-5sulfo-benzen ⁽⁹⁾, all lose their -NO2 group placed in ortho vs the -NH2 and lead to the formation of the corresponding diazo-oxide by diazotation in acidic media. (weak or strong acid?)

1-amino-2-nitro-4-chloro-benzen turns into 2-diazo-5-chloro-phenoxide

Same kind of reaction process is observed into the naphtalen family. So 2-amino-1,6-dinitro-naphtalen diazotized in presence of concentrated sulfuric acid and then dilluted by water immediately delivers the diazo-oxide ⁽¹⁰⁾:

2-amino-1,6-dinitro-naphtalen tuns into 2-diazo-6-nitro-naphtyl-1-oxide

The same happens to 2-amino-1,8-dinitro-naphtalen and to 1-amino-2,4-dinitro-naphtalen ⁽¹¹⁾.

Halogen elimination.

Meldora observed that not only -NO2 and -OCH3 groups but also halogen ring substituants become labile by treatment of diazoniums salts with alkaline agents, or during the diazotation itself when weak acids are present.

So 2-amino-1,4-dibromo-naphtalen when diazotized in presence of acetic acid and then brought to boiling leave a diazo-oxide after subsitution of a Br atom by a -OH group ⁽¹²⁾:

2-amino-1,4-dibromo-naphtalen turns into 2-diazo-4-bromo-naphtyl-1-oxide

The same occurs with 2-amino-1-chloro-4-bromo-naphtalen. Bamberger ⁽¹³⁾ observed an identical reaction when submitting 2,4,6-tribromo-benzen-diazonium chloride to the action of an alkali, thus leading to the dibromo-diazo-oxide:

2,4,6-tribromo-benzen-diazonium chloride turns into 2-diazo-3,5-dibromo-phenoxide

----------------------------------------------------------------------------------
p 134

Orton has studied that reaction by making a comparison of various aromatic halogenoamines on their diazo-oxide speed of formation after halogen elimination ⁽¹⁴⁾.

He put those into the following order, after determination of the transformation coefficient in 24 hours at ambiant temperature.

1-amino-2,4-dichlorobenzen - only traces
1-amino-2,6-dibromobenzen - 24% conversion
1-amino-2,4,6-trichlorobenzen - 63%
1-amino-2,4,6-tribromobenzen - 64%
1-amino-2,3,4,6-tetrabromobenzen - 74%
2-amino-1-chloronaphtalen - 76%
1-amino-2,4-diibromo-5-nitrobenzen - 80%
1-amino-2,4,6-tribromo-3-nitrobenzen - 93%
1-amino-2,4-dibromonaphtalen - 97%

The elimination of the halogen and the formation of the diazo-oxide happen when the pH of the medium has a value such that both hydroxide and diazonium ions are present ⁽¹⁵⁾.

Many other cases have been observed and lead to patents. Like for example the diazonium derivatives of the following halogenoamines:

?-amino-2,3-dichloro-4-sulfobenzen ⁽¹⁶⁾,
1-amino-2,3-dichloro-4-sulfobenzen ⁽¹⁷⁾,
1-amino-2-chloro-3-nitro-5-sulfobenzen ⁽¹⁸⁾,
1-amino-2,3,6-trichloro-4-sulfobenzen ⁽¹⁹⁾,
2-amino-1-chloronaphtalen ⁽²⁰⁾,
?-amino-2,4-dichloro-6-sulfonaphtalen ⁽²¹⁾.

Another interesting case is the transformation of the 1,5-bis-diazonium-2,6-dibromoanthraquinon bis-sulfate into a bis-diazo-oxide by treatment with diluted sulfuric acid:

1,5-bis-diazonium-2,6-dibromoanthraquinon bis-sulfate turns into 1,5-bis-diazo-anthraquinon-2,6-dioxide

With a few halogeno-diamines was noticed that diazotation occured normaly on one of the -NH2 group and that the other one in ortho of the halogen provided after elimination of the later a diazo-oxide. So ⁽²²⁾:

2,6-diamino-1-chloro-4-sulfobenzen turns into 6-diazonium-2-diazo-4-sulfo-phenoxide salt

----------------------------------------------------------------------------------
p 135

2,4-diamino-1-chloro-6-sulfobenzen turns into 4-diazonium-2-diazo-6-sulfo-phenoxide salt

Elimination of the sulfo group.

Just like the groups -NO2, -OCH3 and halogens, the sulfo group in ortho position vs a -NH2 can be eliminated and replaced by -OH, with subsequent formation of a diazo-oxide when submitting the diazonium salt to an alkaline agent.

This phenomenon is illustrated by the examples of diazonium salts of the sulfonic acids 2-amino-1,5,7-trisulfonaphtalen, 1-amino-2,4-disulfonaphtalen, 1-amino-1,5-disulfonaphtalen (there must be a mistake since position 1 can't hold two groups at the same time) that decompose with formation of diazo-oxides ⁽²³⁾:

8-amino-1,5-disulfonaphtalen turns into 8-diazo-5-sulfo-naphtyl-1-oxide

Into the case of diamines, we get simultaneously a diazo-oxide and a diazoic compound (diazonium).

1,3-diamino-4,6-disulfobenzen turns into 2-diazo-4-diazonium-5-sulfo-phenoxide salt

It is noteworthy that the second sulfo group is not eliminated, like the first, despite it is also positioned in ortho of a -NH2 group.(and thus that a bis-diazo-phen-dioxide is not obtained…but this may be a consequence from the impossibility to make a meta-quinonic structure)

Migration

A good deal of cases are known where the anion of a diazonium switch its place with one of the ring groups.

Para-chlorobenzen-diazonium thiocyanate obtained by double decomposition between chlorobenzen-diazonium chloride and potassium thiocyanate, turns into the chloride of p-thio

----------------------------------------------------------------------------------
p 136

cyanobenzen-diazonium when exposed to a mildly acidified alcoholic solution ⁽²⁴⁾:

1-diazonium-4-chloro-benzen thiocyanate turns into 1-diazonium-4-thiocyano-benzen chloride

There is also a similar case of switching of Cl and Br atoms; for example, 1-diazonium-2,4,6-tribromo-benzen chloride turns into 1-diazonium-4-chloro-2,6-dibromo-benzen bromide ⁽²⁵⁾:

1-diazonium-2,4,6-tribromo-benzen chloride turns into 1-diazonium-4-chloro-2,6-dibromo-benzen bromide

That transposition occurs with difficulty into an aqueous solution, but happens with more ease into an alcoholic solution, and if a brom atom is positionned in ortho or para of the diazoic group; it must be noticed that the transposition is easier with an increasing number of bromine atoms onto the ring; so 1-diazonium-2,4-dibromobenzen chloride salt is transformed while into alcoholic solution, but is stable when in the dry state while for the 1-diazonium-2,4,6-tribromobenzen chloride transformation occurs even when dry.

Anion exchange does'nt happen when it goes about iodine atoms from the ring (like for 1-diazonium-2,4,6-triiodobenzen chloride), or when it goes about fluoride anion (like for 1-diazonium-2,4,6-tribromobenzen fluoride) ⁽²⁶⁾.

Once was also mention of an anion exchange involving a -NO2 group. So 2-diazonium-1-nitronaphtalen chloride turns, when in anhydrous acetic acid media, into 2-diazonium-1-chloronaphtalen nitrite ⁽²⁷⁾:

2-diazonium-1-nitronaphtalen chloride turns into 2-diazonium-1-chloronaphtalen nitrite

----------------------------------------------------------------------------------
p 137

Elimination, substitution or migration of groups into diazonium compounds. - Bibliography*.

References ¹⁾ to ²⁷⁾ see picture of page 137.

Stability of diazonium salts

Stability into the solid state

Into the solid and dry state, diazonium salts are very unstable and decompose explosively by shock ^(1,2).

The anion has a determining effect on the unstability of the salt that increases in the following order: SO₄^2-, < Cl^-, < NO₃^-, ClO₄^-.The groups onto the ring have also a marked effect; while halogens increase stability, -NO2 groups decrease it and, as a result, increase the tendancy to explode ^(3,4).

Caro and Griess did patent the use of benzen diazonium chromate as an explosive ⁽⁵⁾, but it seems that it didn't got widespread use. On the other hand, 2-diazo-4,6-dinitrophenoxide, obtained by diazotation of picramic acid, revealed to be a powerful explosive worth getting attention ^(6,7,8), and abbreviated into the United States by the letters D.D.N.P.:

picramic acid (2-amino-4,6-dinitrophenol) turns into 2-diazo-4,6-dinitrophenoxide

*See note on foot page 126 (abbreviations).

----------------------------------------------------------------------------------
p 138

That compound takes the form of a yellow crystalline powder with a true density at 25°C of 1,63. The apparent density under a pressure of 239 kg-m² is 0,86. It is soluble into nitrobenzen from what it can be recrystallized, unsoluble into petroleum ether and into carbon tetrachloride, weakly soluble into chloroform, into benzen, into alcohols and into ethers. Clark ⁽⁹⁾ has made a comparison of its explosive properties vs those of mercury fulminate and of lead azide. It is less sensitive to shock as can be concluded from the following comparison:

Minimum falling height of a 500 g weight to generate an explosion

• D.D.N.P
  22,5 cm
  
  
• Lead azide
  17,5 cm
  
  
• Mercury fulminate
  15 cm
  
  

Its explosive power is about three times stronger than that of lead azide. The "Sand Test" examination yielded the following results:

Charge 0,60

• D.D.N.P
  54
  
  
• Mercury fulminate
  27,5
  
  
• Lead azide
  21
  
  

Charge 1,00

• D.D.N.P
  90,6
  
  
• Mercury fulminate
  48,4
  
  
• Lead azide
  36
  
  

As initiating material, it is very close to lead azide and more potent than the fulminate.

Minimum charge to initiate explosion (detonation) of explosives

Picric acid

• Mercury fulminate
  0,225
  
  
• D.D.N.P
  0,115
  
  
• Lead azide
  0,12
  
  

Trinitrotoluen

• Mercury fulminate
  0,240
  
  
• D.D.N.P
  0,163
  
  
• Lead azide
  0,16
  
  

By the specific example of 2-diazo-4,6-dinitrophenoxide, one can see that the diazo group increases the explosive properties of dinitrophenol.

For the large amount of uses resulting from the growing use of diazonium salts as intermediary agents into the synthesis of dyes, the unstability was a major problem that pushed to research stable or stabilized salts on what we will come back later and appart because of their importance (see p. 215).

Stability into aqueous solutions

Into solution, the decomposition reaction of diazonium salts is much slower than in the solid state.

As a result, solutions and in particular aqueous solutions, were used to study the decomposition mechanism and to compare the stability of various diazonium salts with each other.



[Edited on 11-1-2017 by PHILOU Zrealone]

Yes took me some time.
Sadly this forum text protocol doesn't allow for the use of alignement via tabulation so I had to find another way to put tables (via insertion of list and centering of the text). Of course I could have compacted all the text into a single block text, but the aim here was to keep the original structure sothat one reading the initial document finds its marks directly with the translation in parallel.

If we take a deeper look at the following sentences:
------------------------------------------------------------------
"In a more general way than for the polynitro-anisidins, amines that are nitrated ortho vs the NH2 give rise to elimination of the -NO2 group during diazotation if the para position is occupied by an electro-negative group."

"Into the benzen family, 1-amino-2,4-dinitrobenzen, 1-amino-2-nitro-4-sulfo-benzen, 1-amino-2-nitro-4-arseno-benzen, 1-amino-2-nitro-4-chloro-benzen, 1-amino-2-nitro-4-chloro-5-sulfo-benzen, all lose their -NO2 group placed in ortho vs the -NH2 and lead to the formation of the corresponding diazo-oxide by diazotation in acidic media."

"Same kind of reaction process is observed into the naphtalen family. So 2-amino-1,6-dinitro-naphtalen diazotized in presence of concentrated sulfuric acid and then dilluted by water immediately delivers the diazo-oxide:

2-amino-1,6-dinitro-naphtalen tuns into 2-diazo-6-nitro-naphtyl-1-oxide

The same happens to 2-amino-1,8-dinitro-naphtalen and to 1-amino-2,4-dinitro-naphtalen."
------------------------------------------------------------------
We may conclude that:
1°) the presence of an electron withdrawing group or atom in para of the NH2 and the presence of a NO2 in ortho allows for elimination of the ortho NO2. But this electron withdrawing group can be a simple chloride atom (like for the example nitrochloroanilin); or an aromatic ring activated by a -NO2 like for the last 2-aminonaphtalen example.

2°) typical examples of dinitroanilin and dinitronaphtylamin, but also of 2,4,6-tribromoanilin allowed me to extrapolate for the case of trinitroanilin (picramide) what would generate as said another ortho-DDNP variant (the same as the normal one but with the diazo and OH switched) thus the one that would result from diazotation of 2-amino-3,5-dinitrophenol instead of conventional picramic acid.

3°) It seems that for NO2 elimination, basic treatment or weak acid treatment are not mandatory and that simple dillution of the sulfuric acid media is sufficient.

4°) It seems that into concentrated H2SO4 the diazonium salts is stable and dissolved while upon dillution the diazo-oxide precipitates...and this may be paralleled with last Nitro-genes experiments.

About other thoughts: (see picture below)
1°) It would be interesting to test diazotation for the following compounds and investigating new powerful diazo-oxide primaries
-EWG trinitromethyl in para position
--> 2-nitro-4-trinitromethylanilin
--> 2,6-dinitro-4-trinitromethylanilin
--> 2-nitro-4,6-bis-trinitromethylanilin
-EWG nitro in para position
--> 2,4-dinitro-6-trinitromethylanilin

2°) Inspired by the bis-diazo-oxide-anthraquinon case, the design of a bis-diazo-oxide compound seems very interesting because it would mean higher density and more power... what can be interesting and desirable for initiation purpose as a primary or simply as a HE explosive:
-for benzen nucleus (with two nitro groups or one or two trinitromethyl groups)
-for naphtalen nucleus (the diazo-oxide can involve positions on the same ring side of the molecule thus a 5-membered cyclus or may involve the two rings (like positions 1,8 and 4,5) via a 6-membered cyclus)
-for biphenyl nucleus (with two nitro groups or one or two trinitromethyl (nitroformyl) groups on each phenyl ring.
-In the case of benzen because of the unpossibility to get a bis-diazo-oxide involving a discrete meta-quinonic structure, one would have to hypothesize that an ortho-quinonic or para-quinonic structure may hold the line.
As such one would have to investigate pyro-catechol (o-benzendiol) and hydroquinon (p-benzendiol) derivatives like:

1. 1,2-dihydroxy-3,6-diaminobenzen (with explosophoric groups into position 4 and 4,5)
   
2. 1,4-dihydroxy-2,5-diaminobenzen (with explosophoric groups into position 3 and 3,6)
   
3. 1,4-dihydroxy-2,3-diaminobenzen (with explosophoric groups into position 5 and 5,6)
   

In this last case one may get also a competitive reaction vs the diazo-oxide formation because two NH2 are vicinals and usually this may lead to benzo-triazole ring...what may also be interesting.



I keep for me the drawings for naphtalen and biphenyl since it involes a lot more of isomerism possible cases.

[Edited on 13-1-2017 by PHILOU Zrealone]

Thanks Philou, interesting reading material! Would have liked to try the TNA diazonium + base reaction, though I don't have anything HE related anymore. Lots of interesting possibilities for further study, although some of the precursors to the diazophenols you describe in your last schemes may be very difficult to obtain. I also wonder how stable any of the bis-diazophenols would be. Adding the trinitromethyl moiety is an interesting thought, just like diazonium salts and diazophenols from coupled molecules, biphenyls, stilbenes, maybe even some more OTC options like xanthones etc., like nitration off 2,7 diaminoxanthone or bis(4-aminodiphenylether) For the most part, obtaining the required nitro and amino derivatives selectively from nitration and nitro reduction will the challenging here. Maybe some other aromatic carbon rings or even heterocycles might also be able to form stable diazonium phenols so lots of possibilities for sure!

After trying the nitration of some left over pDDNP again (Cautiously adding few mg's at a time to large amount of-20 SA), the light yellow compound is probably the triazene from your scheme. After boiling in potassium acetate until colour change to the colour of pDDNP, the supernatant was treated with CaCl, no precipitation, so no SO4 present. Adding extra KNO3, results in no precipiation of K-picrate. When the brown salt is reaciidified, the light yellow colour returns. The monopotassium salt of the triazene is probably somewhat soluble, while the di-potassium salt upon basifying further is very soluble or leads to destruction. It is interesting that nitroamidoresorcinol can actually be nitrated at low temp using KNO3/SA, still wonder what happens here and in what order, in either case that extra hydroxyl is probably vital.

I wonder if the schffbases from your scheme may form from oxidation by SA itself, stupid... never occured to me that hot SA may be able to oxidize the isopicramic to the quinone imine, could swear smelling some SO2, although impurities in the isopic could also be the cause here. It is very difficult to get completely pure. Takes several crystallizations to get those beautiful golden hair-like crystals, and not all was purified to this extend.

Anyway, most puzzles solved...

[Edited on 14-1-2017 by nitro-genes]

Ok, one more post showing the ignition of both pure crystalline p-DDNP (left) and DDNR (right) Compared to the potassium salt, the free DDNR is relatively tame. Burns somewhat quicker and leaves significantly less residue, which is expected. One final thing that would be interesting maybe is to produce the 4-azido compound of DDNR, not sure how stable it would be. The 4-azido 2,6 dinitrophenol is easily made by reacting 1 mole eqvt of NaN3 in 97% ethanol and leaving it stirring for several hours at room temp until no more evolution of nitrogen is observed, perhaps this would also work for DDNR? It seems that azido phenols can also undergo ring expansion or other internal rearangements somehow, wonder if that is the reason that 2-azido 4,6 dinitrophenol seems more stable than the 4-azido isomer, which cannot be made from aquous solution.

Pure crystalline pDDNP was made dissolving 0.5 grams of purified isopicramic acid in 20 ml 10% w/w SA at 80 deg C. This was cooled to room temperature and filtered to remove minor dark brown impurities. Then 0.2 grams sodium nitrite in 2 ml water was added at once and left stirring for 20 minutes. The pDDNP is obtained as light yellow-orange plates.

DDNR was made by nitration of isopicramic acid: 15 grams 97% SA was added to 20 ml beaker. Then 1.1 grams of KNO3 was added while stirring until all dissolved. This was cooled in an ice bath and 1 gram isopicramic acid added at once. This was left stirring for 16 hours and then slowly diluted with water until total volume of 20 ml. This was extracted with 3X 3 ml ethylacetate at low temperature to extract the 4-diazo 2,3,6 trinitrophenol. All ethyl acetate was left to evaporate. I originally planned to also include the ignition of 4-diazo 2,3,6 trinitrophenol in the video, but it is nearly impossible to get dry due to solvent retaining and reactivity. Finally distilled water added and heated to 80 deg C until no fizzling could be observed. Yield is ~100 mg of light orange-brown crystalline DDNR.

Attachment: Ignition 10 mg pDDNP (left) and DDNR (right).avi (4.6MB)
This file has been downloaded 942 times

[Edited on 1-2-2017 by nitro-genes]

If what you are reporting is true, then this is novel, and goes beyond what Meldola and others had reported about failure of exhaustive attempts to further nitrate isopicramic acid. You should submit this for peer review and publication.

A good way to confirm the DDNR would be to use acetic anhydride to convert some of the paracetamol to the diacetyl prior to nitration and follow the already documented approach to DDNR first reported by Meldola and later confirmed by Klapotke, even though he incorrectly trusted the dubious NMR data that got the structure wrong

Anyway you could then compare the two DDNR samples gotten by different routes and see if they are identical.

The potassium and strontium salts of DDNR are likely both interesting, IIRC and barium and nickel might be also. There could also possibly be a double nickel - potassium salt.

[Edited on 2/1/2017 by Rosco Bodine]

Quote: Originally posted by nitro-genes

IIRC there seemed to be quite some confusion regarding the formation of either the 2,3,5 or 2,3,6 trinitro 4-aminophenol and corresponding products upon diazotization like you mentioned a while back. Not sure if a 2,5 dinitro 4-diazo resorcinol would exist and if this would be most likely to form upon diazotizing the 2,3,5. In this case 2 compounds were wrongly charaterized, which seems very unlikely. If you refer to the compound as DDNR however, then a 4 diazo would be the same as a 6 diazo right? Can imagine how challenging this must have been in a time where no FTIR/NMR etc was available to characterize the compounds produced. In addition, many of these nitrations may produce a mix of isomers. Still wonder if this also is the case for the nitration of paracetamol using conc SA, there is quite some fizzling during the deacetylation, which could originate both from an N-nitroso compound or 2,3 or perhaps even 2,5 dinitroacetaminophenol. Haha, perhaps the DDNR obtained from the nitration of isopicramic is actually from further nitration of one of the former derivatives....not likely, but who knows. Do you have a reference for the nitration of isopicramic? The same reaction at higher temperatures using picramic acid produces mostly DDNP, for isopicramic I would guess the same. Hmm, even 2,6 dinitro 1,4 phenylenediamine produces pDDNP upon nitration using 100% HNO3/AA, so why the sulfuric not? Most likely it prevents premature diazotization as pDDNP doesn't nitrate further. [Edited on 2-2-2017 by nitro-genes]

Actually classical theory approaches and proofs tend to be quite definitive and conclusive, but requires intelligent, logical application of tests and proofs. I don't like it when instruments provide readings that contradict classical theory, and in this case I tend to dismiss the instrument readings and/or the library data or application of such data as suspect. This DDNR compound is obscure enough that it needs good conclusive structure data verification from different methods to have any confidence in what is being reported. It is particularly troublesome to me that the prior art is not properly referenced by the modern authors and Klapotke.

References have already been provided and are in this thread back several pages. I would have to go back and review the entire topic, but I am sure the relevant references are already posted.

Here linked is a post where I was following up on my earlier identification of a probable structure misidentification issue

http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

also some explanation here linked
http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

Classical nomenclature for phenol would go clockwise with position 1 (X) being the OH of the name compound "phenol" (or similarly 1,3-benzenediol for resorcinol). But
position 4 (at 6 O'Clock) would be para, and position 6 (at 10 O'Clock) would be ortho, so no position 4 and position 6 are quite different and there is no "rotation or flipping or reversal or inverting" of the view that accounts for this ring position discrepancy....although there was I think another different compound from DDNR where a view manipulation might accomplish that explanation.



[Edited on 2/3/2017 by Rosco Bodine]

I need to review all of this thread to make certain neither of us have covered this already, but have you tried subjecting the isopicramic acid to the reaction condition described by Benedikt and Hubl or Von Herz?

According to B&H the effect of adding a nitro group simultaneously with diazotization is a heat driven reaction "bonus" observed for the mononitroamidoresorcinol which is IMO an inherently more easily *nitrated* precursor than would be by comparison isopicramic acid. Even so, the scheme of B&H may follow as a general reaction with some variation applied to similarly effect diazotization and nitration upon precursor compounds that are susceptible and isopicramic acid might be susceptible, in cold or hot or sequenced reaction, using a nitrate or nitrite.

The acid concentration conditions and reactants of nitrate or nitrite and sequence and temperature could lead to DDNR or to a different or mixed product result.

Von Herz 1922 patent GB207563 described the boiling condition of SA + NA + an excess of added KNO2 also as a means for pushing the more sluggish and secondary nitration reaction aspect of the two-step reaction sequence that occurred in situ as a "one pot" diazotization/nitration reaction, confirming without attribution what was earlier a similar reaction reported by Benedikt and Hubl.

Attachment: Benedikt and Hubl JCS 1881.pdf (264kB)
This file has been downloaded 499 times




[Edited on 2/3/2017 by Rosco Bodine]

Regarding that diluted H2SO4 described by B&H carelessly, I think they mean to say 26ml H2SO4 + 130ml H2O or perhaps 25ml H2SO4 + 125ml H2O. The other possibility is 155ml H2SO4 + 775 ml H2O which seems unlikely, as a reaction volume of 930ml to later become filled with reaction product crystals from 5 grams of precursor. I hate imprecise language in experiment descriptions. That leaves math that doesn't add up and guess work about what the authors meant to say

I'm not sure about the deactivation, and have been reviewing what I already checked on the directing effect and I'm sure I wrote about it before.

Yes there is some overlap for the diazotizing and nitrating effect of NO2 that is not entirely selective. And some higher nitrations can occur under mild conditions from nitroso precursors, actually further oxidized by the same reagent used for "nitrosation".

As an example I think I posted an article describing salicylic acid sulfonate in dilute solution being nitrated all the way to picric acid by gas dispersion of NO2, and a nitroso intermediate there would likely be a given. The same would likely occur for resorcinol.

Nitric acid is unstable and exists as a sort of equilibrium mixture with nitrous acid, where either one can become the other as needed for a reaction.

I found where you and I were discussing this possible DDNR before with your post linked
http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

I have wondered what might be the effect of a sequence of addition for nitrating / diazotizing and cycling temperature might do.

A possibility may exist that a mixed product could be the result where you have p-DDNP "contaminated" with some percentage of DDNR...which might result in a better initiator "mixture" even though that mixture could be difficult to resolve into its separated components.

In this linked post I described the orienting effect of the 4 diazo on ring nitration for the third entering nitro and provided my opinion then, that on review I still believe is correct.

There seems to be a good probability that DDNR can result from an isopicramic acid precursor, but identifying the best reaction method there to accomplish the best yield of DDNR is perplexing. And certainly byproducts of unknown nature are a possibility.

http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...



[Edited on 2/3/2017 by Rosco Bodine]

So Simple .....isn't it!!! Ha! this one gives me a headache. Brain teaser.

What I found or dug up a reference for was that a diazo at 4 orients an entering nitro ortho-para which in this case would put the entering nitro at position 3 or 5 since the para position at 1 is already occupied by a hydroxyl which is nitro substitution inhibiting.

[Edited on 2/3/2017 by Rosco Bodine]

That would be quite novel would it, any ideas what this intermediate would look like, some NOx species involved? So if I would have to make a guess, you are wondering if nitration of isopicramic using 100% NA would provide a different result than either 65% NA or SA and small quantity of NA?


[Edited on 3-2-2017 by nitro-genes]

The soluble diazonium acid salt intermediate is described in the literature including I think some of the articles in this thread that detail the formation of the diazo oxide. The soluble diazonium acid salt intermediate is only transiently stable in highly acid solution and decomposes on dilution to precipitate the diazo-oxide. How reversible is that reaction I am not certain. I am not sure that redissolving p-DDNP in acid will reconstitute the soluble diazonium acid salt intermediate, even though the p-DDNP redissolves.

I am staying with the same analysis and thoughts I shared about this a year and a half ago. I would respectfully submit to scrutiny by my colleagues in theoretical chemistry, that this particular scenario represents a significant structural "perplexity" arising for the reported NMR data regarding DDNR and would wish everyone *good luck* with their own analysis. I am feeling a sense of genuine satisfaction that I have identified quite a puzzler for others' analysis and review / confirmation


[Edited on 2/4/2017 by Rosco Bodine]

Quote: Originally posted by nitro-genes

That would be quite novel would it, any ideas what this intermediate would look like, some NOx species involved? So if I would have to make a guess, you are wondering if nitration of isopicramic using 100% NA would provide a different result than either 65% NA or SA and small quantity of NA? [Edited on 3-2-2017 by nitro-genes]

It would seem counter intuitive to be able to add a 3-nitro to isopicramic acid by nitration under conditions that would immediately hydrolyse it as well since the 2,3 nitro presence is not favorable seen the observed ring strain and ease of hydrolysis, in this respect the patents going from nitroamidorescorsinol at 30% SA and boiling temperatures are noticeably different of course. There are many explanations why isopicramic can seemingly only form the 4-diazo 2,3,6 trinitrophenol from SA and 2 eqvt of KNO3 . Maybe indeed only some intermediate can be nitrated (nitramine f.e.) further, although I would think this would be strange considering DNAc can't be nitrated further. It is also possible that oxidation like Philou has shown is simply competing for the nitric, and 2 mole eqvt's happens to be sort of a coincidental sweetspot for obtaining some of the 4-diazo 2,3,6 trinitrophenol. I must say, I thought usually, 50-65% nitric-water was considered most oxidative and isopicramic is pretty stable up to temperatures of 60+ C. at these nitric concentrations. Then again, oxidation in 65% nitric may not be comparable in this case, since the water content would lead to diazotization of part of the product, which would resist any further reaction. Maybe overnitration and sponaneous hydrolysis of a putative tetranitro are more plausible side reactions competing for nitric (these products would also be oxidized further probably), or further reactions/stability of any of the products or intermediates in the nitration mix. Maybe the KHSO4 present acts as a solubility modifier allowing nitration as the free acid, maybe something more like a radical nitration adds the 3 nitro, or catalysis from nitrosylsulfuric present. How soluble is NO2/N2O4 in sulfuric, does it form some anhydride in there? Seems like a lot of possibilities, and each one and combinations of these would require a different adjustment of reaction conditions, so I wish anyone good luck trying.

Maybe lower temperatures, thoroughly dissolving the isopicramic before adding any nitrate and studying the effect of performing the reaction in a range of 100%-92% sulfuric may help get some more product.

[Edited on 11-2-2017 by nitro-genes]

Regarding the diazo at 4 / versus diazo at 6 structural perplexity / anomaly that I have CORRECTLY identified with regards to the classical theory conflict with currently reported NMR data for DDNR >

Honest actual scientists are here required to acknowledge the obvious conflicting reported NMR data that has been reported.

I have NOT irrationally "imagined" this PERPLEXITY to exist because of impaired discerning or deficiency of critical thinking. I'm 100% sure that I recognize a very odd scenario and GOOD QUESTION arising in the current "trusted literature" for the diazo compound DDNR identified 136 years ago by Benedikt and Hubl that EXPLICITLY identifies an unreconciled DEPARTURE for reported NMR instrumental data in direct conflict with established classical theory.

There is so much "theoretical" contemporary "science" incorrectly proclaimed to be "settled science" about a variety of things that are quite reasonably still debatable, that I find the irony of identifying this particular DDNR structural anomaly delicious. Indeed I know I am sending some people "back to school" by my identification of an obvious gap in the science that shows the science is simply NOT settled at all. I have thrown my colleagues in Munich quite a curve. So you all are invited and welcome to take your best shot at it. I don't really know myself the answer or explanation for this structural anomaly and perplexity so I can't explain it, other than to point to the anomaly and say it exists.

Quote: Originally posted by Marvin

I'm really confused. If the argument is what I think it is... Klapotke thinks the structure is 6-Diazo-3-hydroxy-2,4-dinitrophenol aka 6-diazo-2,4-dinitroresorcinol ...and this fits the NMR... Rosco thinks the structure is 4-diazo-2,6-dinitroresorcinol. How is this different?

You don't know? And stil you wrote correctly the two isomers...if they have different names...there is a chance those are two different molecules or there must be a reason...
A little drawing is better than a long speech so...


Do you get the difference now?
Are they superimposable by flipping or rotating...?
No --> different compounds.

[Edited on 21-2-2017 by PHILOU Zrealone]

Quote: Originally posted by Rosco Bodine

Rosco also believes that for DDNR the diazo-oxide forms straight across the ring as a para structure, not bonding with the hydroxyl that is ortho, adjacent as occurs for o-DDNP, but like the structure of the diazo-oxide found with p-DDNP. The tension across the ring bends the entire ring structure like a hexagonal spring wave washer being bowed by tension across its opposite vertices at para positions 1 and 4, and the diazo-oxide structure is the bowstring. [Edited on 2/21/2017 by Rosco Bodine]

Simply ask and voilà added

The structure of trans-diazo is stil a bit mysterious...in reallity it must display a trans quinonoid structure so that both sides can bend onto the same side of the paper plane while the diazo connects to the opposite side...but connectivity is a bit hard to figure out. Into the book Traité de chimie organique Azoiques-diazoiques,.... they mention that the probable connectivity passes via a direct link between the N2 and the Cs on both sides thus without the oxygen being involved...a bit a cage structure related to DABCO(DiAzaBiCycloOctane) but with the Ns at other places and with unsaturations...anyway this is highly stressed...

Quote: Originally posted by PHILOU Zrealone

Quote: Originally posted by Rosco Bodine   Rosco also believes that for DDNR the diazo-oxide forms straight across the ring as a para structure, not bonding with the hydroxyl that is ortho, adjacent as occurs for o-DDNP, but like the structure of the diazo-oxide found with p-DDNP. The tension across the ring bends the entire ring structure like a hexagonal spring wave washer being bowed by tension across its opposite vertices at para positions 1 and 4, and the diazo-oxide structure is the bowstring. [Edited on 2/21/2017 by Rosco Bodine] Simply ask and voilà added The structure of trans-diazo is stil a bit mysterious...in reallity it must display a trans quinonoid structure so that both sides can bend onto the same side of the paper plane while the diazo connects to the opposite side...but connectivity is a bit hard to figure out. Into the book Traité de chimie organique Azoiques-diazoiques,.... they mention that the probable connectivity passes via a direct link between the N2 and the Cs on both sides thus without the oxygen being involved...a bit a cage structure related to DABCO(DiAzaBiCycloOctane) but with the Ns at other places and with unsaturations...anyway this is highly stressed...

That looks good except for the first entry should be 2-diazo-(1-oxide)-4,6-dinitrophenol instead of the 3-oxide you show.

The middle compound is synonymous with what Klapotke called HODDNP

Looking at the last listing, the 6-diazo-(3-oxide)-2,4-dinitrophenol nomenclature would work there for the structure inverted and rotated counter-clockwise 30 degrees.

It is the stressed structure that accounts for the surprising energy as an initiator compared with an ordinary styphnate.

Anyway, the convoluted nomenclature is in part or completely explained by not associating the compound with resorcinol as a derivative.

I think that different researchers working from different precursors and not making the association with resorcinol is why the failure to make attribution to the earlier references about DDNR that trace back to Benedikt and Hubl in 1881.

It could be the NMR data is correct but is referenced to structure nomenclature for phenol rather than resorcinol, then having all the expected resorcinol associated position numbers made into alternate expressions. Also, Klapotke identifies ortho bonding with the adjacent hydroxyl to the diazo as occurs with the usual o-DDNP. But the scenario we have involving p-DDNP is more similar and strongly suggests a diazo-oxide that is the same as DDNR, even though it would seem either structure is possible.

[Edited on 2/21/2017 by Rosco Bodine]

Thanks to Louis for the synonymous expressions. In the context of the alternative synonymous naming for DDNR, the issue with the NMR data for the diazo position is reconciled as not being a problem about the NMR data but the differing nomenclature, recognizing the same NMR data could correspond with a diazo at position 6 or 4 described by different diagrams orientation corresponding to different long names and "parent" compounds. There is still a question about the diazo-oxide configuration, whether it is the hydroxyl oriented ortho or para to the diazo, that becomes an anhydride in forming the diazo-oxide.

The differing nomenclature for the same DDNR compound seems to be the best explanation why all the older prior art references were not included by Klapotke, and why the DDNR is being called a "hydroxyphenol" which is technically correct, but diverts attention away from the prior art for DDNR referenced as a derivative of resorcinol.

Making this still more confounding is that isomers are possible, for the DDNR, although in this particular case I believe the compound [8] identified by Klapotke is the same compound identified in 1881 by Benedikt and Hubl, later by Mendola and others.

I think the uncertainty of which hydroxyl (ortho or para to the diazo) is intact and which hydroxyl becomes the anhydride associated with the diazo has contributed to the confusion.

I think these 2 linked posts were the earliest posts where I was questioning the structure being identified, and saying I thought the hydroxyl and anhydride were transposed
http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

The significance of what I am trying to say is that Klapotke's compound [8] and DDNR are one and the same.

Based on experiments reported by nitro-genes there is a niche reaction condition that allows for (probable) DDNR to be formed from a paracetamol precursor, without having to first further acetylate paracetamol to a diacetyl derivative.

The interesting and bizarre story here is that over a period of 134 years, four or five different researchers have made the exact same compound, (DDNR) by different methods and called it by different names, "reinventing the wheel" as if it were invented for the very first time And it seems nitro-genes has added another chapter to the story, not yet fully sorted out.

[Edited on 2/21/2017 by Rosco Bodine]

While the niche reaction conditions for best yield of DDNR from paracetamol not further acetylated before nitration are an interesting unexpected development reported by nitro-genes, how efficient will be that approach is unknown and may not ever work as well as proceeding from paracetamol acetate, which is Klapotke's compound [5].

It appears likely to me that the N NMR data identifying a diazo at position 6 for compound [8] of the Klapotke article is explained by nomenclature semantics that have inverted and rotated positions so that Klapotke's position 6 diazo for Klapotke's compound [8] (and likewise for unrecognized [8] is DDNR) is synonymous with what I am calling position 4 for the same diazo and is the same compound, DDNR, with confusion over nomenclature arising at compound [5] in the Klapotke article where p-aminophenol is converted to a diacetyl derivative that is O,N-diacetyl-p-aminophenol, or paracetamol acetate, or 4-Acetamidophenyl Acetate, but is identified synonymously by Klapotke as N-(4-Acetoxyphenyl)acetamide, which inverts the subsequent ring position numbering and causes the nomenclature confusion thereafter. Other confusion of the structure numbering would have occurred for nomenclature like 4'-Hydroxyacetanilide Acetate.

For example, in the alternative, if Klapotke's compound [5] had been named O,N-diacetyl-p-aminophenol and obtained as at least 100 grams yield from 97 grams of paracetamol substituted for the 35 grams of p-aminophenol described which yielded only 50 grams instead, everything to follow would have been simplified, efficiency would have improved and the nomenclature confusion about the 4 versus 6 diazo would have been entirely avoided.

Likewise, it is the potassium derivative of Klapotke's compound [8] that is of particular interest for testing initiator properties identified by Von Herz and physical properties identified by Benedikt and Hubl, Meldola, et al, in order to ascertain if indeed the compound [8] of Klapotke is DDNR as I believe it is.

Klapotke's compound [8] 6-diazo-3-hydroxy-2,4-dinitrophenol (HODDNP) is I believe the same DDNR identified by Benedikt and Hubl and Meldola and others.





Here attached again are the two Klapotke articles

Attachment: Synthesis & Energetic Properties of 4-Diazo-2,6-dinitrophenol and 6-Diazo-3-hydroxy-2,4-dinitrophenol.pdf (286kB)
This file has been downloaded 698 times

Attachment: Synthesis and Initiation Capabilities of Energetic Diazodinitrophenols.pdf (754kB)
This file has been downloaded 898 times

[Edited on 2/22/2017 by Rosco Bodine]

Quote: Originally posted by Marvin

PHILOU, Where did 2-diazo-(3-oxide)-4,5 dinitrophenol / 6-diazo-(5-oxide)-2,6-dinitrophenol come from? I didn't write that. Who thinks that is the structure?

You indeed wrote:
6-diazo-3-hydroxy-2,4-dinitrophenol aka 6-diazo-2,4-dinitroresorcinol
and
4-diazo-2,6-dinitroresorcinol.(*)

Those are ambiguous names as seen in the practice by all of us when trying to understand each others between forum members, but also when trying to understand and correlate the research of Klapote's team and previous early works from the past.(**)

So your wrongly named and ambiguous 6-diazo-3-hydroxy-2,4-dinitrophenol is the same as 6-diazo-3-hydroxy-2,4-dinitrophenoxide (an ortho diazo-oxide derived from 4-amino-1,3-dihydroxy-2,6-dinitrobenzen); while your wrongly and ambiguous 4-diazo-2,6-dinitroresorcinol is in fact refering to 4-diazo-3-hydroxy-2,6-dinitrophenoxide (a para diazo-oxide derived from the very same 4-amino-1,3-dihydroxy-2,6-dinitrobenzen)

As I exposed by drawing:
From trinitroresorcinol (1,3-dihydroxy-2,4,6-trinitrobenzen) by monoreduction of a NO2 group you only get two possible isomeric amino-dinitroresorcinol:
1°) the amino comes from reduction of the nitro trapped between te two NO2
--> 2-amino-1,3-dihydroxy-4,6-dinitrobenzen
2°) the amino comes from reduction of one of the two others NO2 that are next to one OH (in ortho) and in para of the other OH...and those only provide one isomer by degenerescence linked to the symetry of the molecule (the axis being between the two OH)...so reducing the 4 or the 6 nitro will always lead to 4-amino-1,3-dihydroxy-2,6-dinitrobenzen

From 1°) aka 2-amino-1,3-dihydroxy-4,6-dinitrobenzen
by diazotation you will get only one isomeric 2-diazo-oxide because the -N=N-OH may react (anhydrize) with the -OH in position 1 or with the -OH in position 3 what is the same (again degenerescence) also because of the inter-OH molecular symetry...
So in fact:
2-diazo-(1-oxide)-3-hydroxy-4,6-dinitrobenzene
is equal to
2-diazo-3-hydroxy-4,6-dinitrophenoxide (not phenol because the OH in position 1 is not present anymore and phenoxide means implicitely that the oxide is in position 1)
what is equal to
2-diazo-(3-oxide)-4,6-dinitrophenol (the OH in position 1 implicitely)
and if you flip along the free OH axis this is equal to
6-diazo-(5-oxide)-2,4-dinitrophenol (cf supra)

From 2°) aka 4-amino-1,3-dihydroxy-2,6-dinitrobenzen
by diazotation you will get only two isomeric 4-diazo-oxide because the -N=N-OH may react (anhydrize) with the -OH in position 1 (thus in para position vs itself) or with the -OH in position 3 (what is ortho position vs itself)...
So in fact:
para case
4-diazo-(1-oxide)-3-hydroxy-2,6-dinitrobenzene
is equal to
4-diazo-3-hydroxy-2,6-dinitrophenoxide (not phenol because the OH in position 1 is not present anymore and phenoxide means implicitely that the oxide is in position 1)
what is equal to
2-diazo-(5-oxide)-4,6-dinitrophenol (the OH in position 1 implicitely)
and if you flip along the free OH axis this is equal to
6-diazo-(3-oxide)-2,4-dinitrophenol (cf supra)

ortho case
4-diazo-(3-oxide)-1-hydroxy-2,6-dinitrobenzene
is equal to
4-diazo-(3-oxide)-2,6-dinitrophenol
what is equal by flipping along OH axis to
4-diazo-(5-oxide)-2,6-dinitrophenol
what is equal to
2-diazo-(1-oxide)-5-hydroxy-4,6-dinitrobenzene
what is equal to
2-diazo-5-hydroxy-4,6-dinitrophenoxide
what is is equal to
6-diazo-(1-oxide)-3-hydroxy-2,4-dinitrobenzene
or
6-diazo-3-hydroxy-2,4-dinitrophenoxide

Yes you start to feel dizzy and suffer from vertigo...your brain is overheating and fuming and this gives you a better view of the problem...

(*) So what you wrote implicitely is exactly what I wrote explicitely...
I deliberately noted the oxide place and number to make things clearer...because this is the main source of confusion...
I really think that everybody here should always do so...to be sure what we speak about.
(**) If there is little ambiguity (***) when dealing with DDNP beause the diazo can only form a single diazo-oxide with the only OH present; when dealing with dihydroxyphenol (resorcinol)...confusion arises and speaking about DDNR becomes a problem.
In common usage for the future we should refer DDNP to be Diazo-Di-Nitro-Phenoxide and not Diazo-Di-Nitro-Phenol because the -OH is inexistant into the molecule...for analogy it is like saying that CH3-O-CH3 is methyl-methanol...while it is an ether (dimethyl-ether).
(***)The tiny ambiguity may come from the various true isomers and from the effective numerotation/nomenclature used...
so a little molecular drawing is never too much.


[Edited on 22-2-2017 by PHILOU Zrealone]

It is hard to believe there really exists a N=N-O bond for diazophenols:

1. The ring strain produced for pDDNP would be really immense for an aromatic and perhaps more importantly, much higher compared to oDDNP. Yet, the difference in energy release and initiating properties between the two seems very similar. (Although the latter still needs proper verification, I would say pDDNP is somewhat more energetic, but perhaps also seems so due to mono-nitro present)
2. One would expect that Tdec would be lower for pDDNP due to increased ring strain, while the opposite is true
3. In all likelyhood, both DDNP's dissolve in strong acids to form a diazonium salt
4. Even moderately strong nucleophiles can easily react with the diazogroup leading to substitution, even under (almost) waterfree conditions
5. No other compounds containing a R-N=N-O-R group have been found (fafaiaao), while diazotates are charged (behaving more like a OH- salt IIRC)

I think depending on conditions, DDNP's represent something between a zwitterionic and quinoid structure. In very strong concentrated acids it can behave as the former, but the electrons are not completely sure where they should reside the most.

So where does the extra energy for pDDNP come from in Klapotkes calculations? From assuming the NNO bridge?

[Edited on 24-2-2017 by nitro-genes]

In the case of the p-DDNP the diazo-oxide bridge has only one path to form which is straight across the ring, since there is no other hydroxyl available. And I think the nitros on each side of that para hydroxyl amplify the attraction and steer that diazo-oxide linkage formation there sort of like magnets steering an electron beam. And I think the combined effect of those "steering groups" is stronger than the attraction of an adjacent hydroxyl as a prospective target, particularly if that hydroxyl hasn't formed yet because it is a nitro at 3 and must decompose to a hydroxyl subsequent to the formation of the para diazo-oxide linkage across the ring. I could be wrong, but that is my guess what is occurring. If the second hydroxyl was already present before the diazo-oxide linkage formed, then it might preferentially form the linkage with the adjacent hydroxyl similarly as is the case for o-DDNP. There is an uncertainty about these reactions and sequence that could make possible different isomers. To sort out what is the whole true story would require performing different paths of synthesis and comparing the products to see which are the same and which may be different.

I think the extra energy for the p-DDNP comes from the para diazo-oxide bond and added stability comes from the symmetry.

Thanks to Boffis for the file for the Klapotke article that is published. We have been discussing what was information in an early preview draft that had not yet published so there may be unknown changes or corrections. Attached is the file for the article as published.

Attachment: 4-Diazo-2,6-dinitrophenol and 6-Diazo-3-hydroxy-2,4-dinitrophenol.pdf (432kB)
This file has been downloaded 565 times

I already see what I believe is a problem in the NMR data table figure description.

The first two compounds based on identification of structure diagrams appear to be transposed. The legend shows compound [7] as the middle compound, but the structural diagram is for iso-DDNP which is compound [4]

Likewise for the first structural diagram shows compound [7] but the legend for the table says it should be compound [4].

So the table legend for Figure 3 is wrong for the first two compounds. I am pretty sure the top graph is for [7} and the middle graph is for [4].




[Edited on 2/24/2017 by Rosco Bodine]

Okay I think I have sorted out what has been puzzling about this Klapotke article and why his reported compound [8] may differ from Meldola's p-DDNR structurally with regards to the diazo-oxide appearing as an ortho linkage to the adjacent hydroxyl to the diazo, instead of appearing as a para linkage diazo-oxide to the hydroxyl across the ring directly opposite the diazo.

If what Klapotke is reporting is accurate, (and I now believe it is), then it is possible to produce either isomer depending upon the method of diazotization applied to the compound [7] which is 4-amino-2,3,6-trinitrophenol. Following the method of diazotization of Klapotke depends upon the decomposition of the nitro at 3 to diazotize the amino, so that sequence makes available the ortho hydroxyl adjacent to the diazo as soon as it forms. In that case, if that is what does occur then indeed HODDNP could be the result.

In an earlier post linked here Meldola used a different means of diazotization applied to the same 4-amino-2,3,6-trinitrophenol, and used small portions of solid nitrite added to a solution of the 4-amino-2,3,6-trinitrophenol in cold H2SO4, under conditions where the 3 nitro remained intact as the diazotization was
performed. In that circumstance there was only one hydroxyl available for the formation of the diazo-oxide bridge and that would be the para, just the same scenario as occurs for p-DDNP.
http://www.sciencemadness.org/talk/viewthread.php?tid=439&am...

Meldola actually isolated the 4-diazo-2,3,6-trinitrophenol(1-anhydride) as a dense yellow microcrystalline powder precipitated from dilution of the soluble diazonium sulfate. Boiling that material with a strong solution of sodium acetate decomposed the 3 nitro and produced the sodium salt of p-DDNR, even though it is clear Meldola was not certain himself of the structure, and did not positively identify one isomer. Given the reaction sequence as the most logical influence, it would be likely that Meldola was describing the para isomer variant different from the ortho diazo-oxide DDNR of Bendikt and Hubl. The DDNR of Benedict and Hubl has the ortho hydroxyl already present, as would other diazotization schemes applied to styphnamic acid. So evidently the ortho bridge for the diazo-oxide would be the usual, "normal" structure for DDNR and is the exact same compound as Klapotke's HODDNP.

Klapotke's HODDNP = Benedikt and Hubl's DDNR

Meldola describes what would be an iso-DDNR or p-DDNR variant that has the same para diazo-oxide bridge as p-DDNP.

So the normal, more usual diazo-oxide bridge is the same ortho structure for DDNP and DDNR. The classical DDNP of Griess is o-DDNP and the classical DDNR of Bendikt and Hubl is o-DDNR.

The para variant of DDNR would then be the para "iso" structure for the p-DDNP and for the p-DDNR. The enhanced energy and stability that appears for p-DDNP would reasonably apply also for the p-DDNR.

So at this point, the logical hypothesis for me to make is that depending upon the method of diazotization applied to Klapotke's compound [7] 4-amino-2,3,6-trinitrophenol it is possible to obtain either isomer desired. Following the method of Klapotke the product will be HODDNP, o-DDNR, but in the alternative, following the diazotization method of Meldola and hydrolyzing the product will be p-DDNR, or iso-DDNR, obtained as the sodium or potassium salt, with the free acid p-DDNR obtained by acidification with HCl or other acid.

So I think ...the mystery is solved!!!! See how easy that was

Hey it was so much fun ...let's do it again

To summarize what has been distilled from the study of the available references, the nexus which I had identified for the compounds described by Benedikt and Hubl, Von Herz GB207563,
Hagel and Redecker US4246052, and Klapotke compound [8]
appear to be all DDNR ....and about that nexus I was correct.

However, all of those examples of DDNR (referenced to ring positions for resorcinol) are the ortho structure for the diazo-oxide from what would be position 4 of the diazo to the adjacent 3 position hydroxyl, made into a 3 anhydride.

The DDNR reported by those several authors is

4-diazo-2,6-dinitroresorcinol-(3-anhydride) o-DDNR = HODDNP

I was incorrect in my early impression that the "normal" DDNR diazo-oxide structure would be the para structure that would be from the diazo at 4 to the hydroxyl at 1, forming a 1 anhydride.
That is the exceptional structure obtained by an alternate scheme for diazotization used by Meldola.

The p-DDNR of Meldola would be.

4-diazo-2,6-dinitroresorcinol-(1-anhydride) p-DDNR or iso-DDNR

The name is somewhat misleading for p-DDNR since the compound is not strictly a resorcinol "derivative" but is structurally similar.

That para diazo-oxide iso-DDNR or p-DDNR is the "special case" isomer obtained by a modified diazotization scheme of Meldola, applied to a precursor that is Klapotke's compound [7].

Paracetamol is the starting material for p-DDNP and p-DDNR and each are expected to have enhanced properties over their more commonly known analogous ortho isomers.

[Edited on 2/26/2017 by Rosco Bodine]

Why the reaction of trinitrophenol and cyanide occurs so readily to form isopurpuric acid is beyond me, but the product upon partial hydrolysis of one of the nitrile groups and diazotization to the benzazimide seems OTC. "explodes violently at 210 deg C" sounded promising in any case .

Seems the reaction of nitrophenols with cyanides to purpuric acids was already discussed to some extent in this thread:

https://www.sciencemadness.org/whisper/viewthread.php?tid=10...

Though no mention of a diazogroup or benzazimide, so I therefore decided to post this in the DDNP thread. Technically, the compound contains a diazogroup, two nitro groups and a hydroxyl, so posting this here seemed appropriate.

Haven't seen much on benzazimide derivatives as energetic materials before. Also curious what the reaction of cyanide and TNP would do when no excess of cyanide is used. Interesting reaction, probably occuring via a meisenheimer complex due to basicity and nucleophilic attack by the cyanide ion like in VNS, but if so why can't you directly add an azide group on TNP? The reduction of one of the nitrogroups is maybe helping here, but where did the missing oxygen go, is cyanate maybe acting as the nucleophile here?





[Edited on 5-10-2017 by nitro-genes]

Thank you Nitrogenes for that sharing...
Very strange molecule (those iso and meta-purpuric acids) and interesting reaction indeed... never read about it...seems so easy

Previously I attempted the further nitration of isopicramic acid using KNO3/SA and obtained a small yield (<10%) of 2,3,6 trinitro 4-diazophenol for which the 3-nitrogroup easily hydrolyzes producing 2,6 dinitro 4-diazo resorcinol. I was curious if using nearly anhydrous mixed acids and very low temperatures to suppress oxidation as much as possible could produce a higher yield.

2 grams of isopicramic acid (0.01 moles) was dissolved in 30 grams freshly distilled azeotropic sulfuric (98.3%) at room temperature. The dissolution itself was not only somewhat exothermic but also released some gasses. the clear orange solution was then put in the -23C freezer and when fully cooled, 1.39 (0.022 moles) grams of ~98% nitric was added, briefly swirled and put back in the freezer. A small sample was taken, crashed on ice and diazotized every day, the first few days, mainly p-DDNP was produced, indicating that nitration at these temperatures is VERY slow. After 5 days in the freezer still large amounts of a light yellow compound were present (p-DDNP, 2,6 dinitrobenzoquinone, maybe the triazene Philou mentioned). Reckoning the nitration would take a month or so or wouldn't proceed at all at these temperatures, I decided to led it further react for 24 hours at 4 deg. C. in the fridge. The colour had then changed to a dark red and small bubbles were visible in the mix. While in the icebath, the mixture was then diluted dropwise with an icecold water/sodium nitrite solution to a total volume of 100 ml. Upon adding ethylacetate for the extraction, a pale yellow compound precipitated at the interphase, which I suspect to be the 2,3,6 trinitro 4-diazophenol itself. (Maybe it would precipitate from the dilute sulfuric as well with some more patience ). Extractions were continued until all precipitate was extracted. When all ethyl acetate had evaporated, 20 ml of distilled water was added and left to stir at room temperature for 1 hour, the precipitate washed with another 10 ml cold water and dried. Yield was 0.37 grams of a light yellow crystaline solid, presumably 2,6 dinitro 4-diazo resorcinol.

I tried to grow well defined 2,6 dinitro 4-diazo resorcinol crystals from slow evaporation of a MEK solution, but it seems to react with the solvent itself, producing an orange coloured impurity. Similar to DDNP's, the best method of recrystallization seems from 65% nitric since this oxidizes most impurities. 100 mg of crude 2,6 dinitro 4-diazo resorcinol was added to a 20 ml beaker and made into a paste with a few drops of water. Then 65% nitric was added with slight warming untill everying dissolved (2 mls or so). Hot water was added to a total volume of 20 ml and put in the fridge for several hours. The DDNR crystallizes as long bright yellow plates (attachment).







Attachment: 2,6-dinitro-4-diazoresorcinol - Copy.avi (2.8MB)
This file has been downloaded 1059 times

[Edited on 31-10-2017 by nitro-genes]

Was mainly curious if the nitration of isopicramic could be improved somewhat, any ideas if/how this could be further improved?. Curious what exactly causes the side occuring oxidation of isopicramic to the quinone imine, maybe some direct rearangment of a nitramine instead of hydrolysis, if so would theoretically a stronger acid than sulfuric help prevent this instead of anhydrous conditions as I assumed previously?

Would be interesting to look at possible double salts etc, though accurately experimenting with sensitive salts of these compounds would be difficult at a scale I would be comfortable with. Briefly touching it with a glowing splint seemed effective enough in desensitizing it completely

Quote: Originally posted by nitro-genes

No, the salts are supposedly extremely sensitive to friction and can explode upon crystallization, so I never made more than a few mg's. Even when a suspension of 2,6 dinitro 4-diazo resorcinol in water is stirred briefly with a neutral solution of KNO3, the potassium salt is already formed too some extend and the resulting product makes DDT in sub-mg amounts, so care should be taken to avoid any contact with metals or salts. I'm not entirely sure this is 2,6 dinitro 4-diazo resorcinol as well, the product I obtained seems more easily decomposed in hot water and not sure if the reactivity towards MEK is also expected for the 2,6 dinitro 4-diazo resorcinol.

Thanks Philou for explaining the probable cause of the reactivity with butanone. I've used acetone for DDNP recrystallization regularly and never really observed any reaction there, maybe the increased acidity of DDNR is contributing here, or maybe because acetone can form a terminal -ene in contrast to butanone?

Regarding the ease of going from the O-acetylation of acetaminophen to DDNR: Taking "Synthesis and Energetic Properties of 4-Diazo-2,6-dinitrophenol and 6-Diazo-3-hydroxy-2,4-dinitrophenol" as an estimate, the total yield would be something like: 0.81*0.56*0.76*0.75=26% overal yield. This is probably biased by the small scale and focus on purity, though the nessecary 4-amino-2,3,6-trinitrophenol intermediate is not very stable and likely dangerous to store. The nitration of acetaminophen can probably be optimized to 90+ yields with good temperature control, high purity acetaminophen and a nitrous scavenger (like guanidine nitrate in a previously posted paper), so even a 50% yield (doubt this is possible though) for the nitration of isopicramic to DDNR could be competitive with other published synthesis routes, perhaps even those from resorcinol/styphnic (mono reduction of styphnic to styphnamic using anhydrous stannous chloride also only had a ~60% yield IIRC). The nitration of isopicramic is certainly the most interesting route, since it has not been published or examined previously, besides, it is the only completely OTC route possible.

That is...if it really is DDNR...one of the things that strikes me is that the burn video of the pure compound isolated from the nitration of isopicramic shows hardly any sooth, despite the fact that DDNR would only have a marginally better oxygen balance compared to DDNP. The presence of the diazo group can be tested for probably, though another possible hydroxyl, the presence of an extra nitro group originating from a putative tetranitro 4-aminophenol intermediate would be hard to test for, since an ortho dinitro presence would be just as suceptible to nucleophillic replacement f.e as an diazogroup. If you assume however a tetranitro 4-diazophenol would actually exist, the diazogroup itself would likely be liable for decomposition, resulting in a tetranitro -p-hydroquinone, or tetranitro/pentanitrophenol. This seems very unlikely though, since the initial product of the nitration of isopicramic seems to produce NOx when reacted with water and not N2 gas, so this makes it unlikely that the product obtained is derived from a putative tetranitro compound.

I have wondered if it would be possible for the 2,3,6-trinitro 4-aminophenol to rearrange differently, maybe resulting in a furoxan or furazan group. Still wonder if DDNR would be able to form salts like is published, wouldn't this be hard to explain assuming a quinone or zwitter-ionic like structure? It is interesting to notice that the difference in elementary composition between an ortho-diazoquinone and a furazan would be negligible. Klapotke et al have found no furazan/furoxans though when heating 2,3,6 trinitro-4-aminophenol in 65% nitric. Besides, I doubt a furazan/furoxan derivative would behave as energetic in the burn-test-video I previously posted.

All in all the only likely possibility is that the compound is DDNR after all, pfff glad I could convince myself again.

[Edited on 11-11-2017 by nitro-genes]

Quote: Originally posted by Rosco Bodine

DDNP forms a 1:1 low melting eutectic with picric acid when solvent acetone used to wet the mixture to a paste is evaporated with warming, and this may involve a diazo coupling reaction. A similar effect may occur for DDNR and styphnic acid.

Had some DDNR left from the previous nitration of isopicramic, it showed some discolouration from exposure to indirect sunlight, so decided to do some tests and be done with it. DDNR seems very soluble (even at room temperature) in a zinc acetate/acetic buffered solution at pH=6, so it appears the zinc salt could be used to produce other salts if needed. Below pH of 8, N2 gas is slowly produced, forming a brown compound and deep red solution as described earlier . Some NaN3 was also added to a cooled solution of the zinc salt, immediately leading to N2 production and upon standing a brown gelatinous precipitate was produced that wasn't energetic. The decomposition temperature of DDNR seems similar to that of p-DDNP when together on the hotplate, exploding at around 170-180 deg. C. Interestingly, both samples seemed to melt first, followed by explosion only seconds later. Together with the dissolution in concentrated acids, it is very likely to be DDNR after all.

Also made a few mg's of the potassium salt of 2,6-dinitro 4-diazoresorcinol (attachment), by precipitation from the zinc salt using a solution of potassium nitrate. It separates on strong cooling as yellow/orange triangular shaped plates that were incredibly static, jumping around under the binocular when touched. Rough ignition temperature was determined on the hotplate, surprisingly, the explosion temperature was a lot higher than for the acid DDNR itself, darkening at 150-160 deg C, explosion within 10 seconds at 215-225 deg C, immediate explosion at 240-250 deg C.



[Edited on 24-12-2017 by nitro-genes]

Hypochlorite mediated oxidation of 2,3,6 trinitro-4-aminophenol to a furoxan might be tricky due to the ease of hydrolysis of any possible quinone-imine formed under these conditions due to OH/NH2 group in para, likely formation of chloropicrin from nitrophenols in general and the well described incompatibility of the compound with bases, decomposing to form tarry products.

Going through the excerpts posted, it seems any possible furazan/furoxan from the nitration of isopicramic acid could be harder to distinguish from a diazophenol than I thought, potentially sharing similar melting points, colour, explosion temp, solubility in strong acids and light sensitivity.

I wondered if the trinitrodiazophenol obtained from the nitration of isopicramic acid may rearrange in water or under slightly basic conditions to a benzofurazan/furoxan as opposed to only displacement of the 3-nitro by water or OH-, but can't figure out a likely mechanism, since it would entail a series of complex rearrangements not generally known for diazo(nium) groups and unlikely maybe due to the para orientation of the OH/NH2 group. Couldn't find any examples of furazan/furoxan formation from diazonium precursors in literature at least, except via azide substitution of the diazo group and oxido-redox with an ortho nitrogroup, like in the synthesis of KDNP, which probably doesn't occur through a quinone-imine precursor in the first place. The latter also suggests that any furoxan formed wouldn't react further with the excess azide used in some reaction schemes outlined, which directly contradicts the observed formation of nitrogen for the compound obtained from further nitration of isopicramic and seems the best argument the compound truly is a diazophenol. Moreover, the in literature described "boiling dilute SA-excess nitrite treatment" of mono and dinitro-4-amino resorcinol derivatives also results in a compound that forms salts, which would be hard to explain as a furazan/furoxan without an extra nitro group present ortho relative to the amino group after nitration, which seems very unlikely to occur under these conditions.

As an interesting side note, I subjected 2-nitro 4-diazophenol to the "boiling 30% SA-excess nitrite" treatment and obtained a product which proved very similar to pDDNP with regard to oxygen balance, so definitely a second nitro group was introduced. Where was it likely to end up? Position 3,5 or 6?

So here a follow up of the conversion of picric acid to picramic acid using copper(II)sulfate and ascorbic acid from the short questions quick answer thread discussion with Rosco Bodine:

After some experimentation, it seems no necessity to increase the pH during the reaction at all. The entire reaction seems to be facilitated by ligand coordination of ascorbic acid by the copper(I) or (II) towards the picric acid, resulting in a very high specificity of the reaction, resulting in the reduction of 1 ortho nitro group of picric acid, producing picramic acid, which eventually seems to precipitate during the reaction as the insoluble copper(I) or copper(II) salt. The first step probably comprises forming a complex of Cu(I/II)(Ascorbic)x(Picrate), which upon heating for some time produces copper picramate.

https://books.google.nl/books?id=PshJAQAAMAAJ&pg=PA407&a...

Experimental 1:

1 gram of picric was added to 25 ml distilled water and brought to near boil. The picric solution was adjusted to a pH of 7 using 10% NaOH solution, then 1.1 grams of copper(II)sulfate pentahydrate (~1 molar eqvt relative to picric) was added at once. When all had dissolved into a greenish solution, 2.7 gram ascorbic acid (~3.5 molar eqvts) were added at once. The solution went a clear dirty greenish-yellow-brown first. After a few minutes of heating, large amounts of a greenish precipitate started to develop. The solution was kept near boil for another 15 minutes, which seemed to change the colour to a yellow-greenish (not sure yet the extended boiling time is necessary at all). During heating, lots of an odorless gas was produced, evident by a lot of foaming. Most of this had died down after the 15 minutes of heating. The solution was cooled and the yellow-greenish precipitate filtered off, which after drying weighed 0.77 grams (~67% yield, assuming pure copper picramate). 25 mls of 10% HCl was gently evaporated down to 10 ml. The greenish precipitate was added in small amounts untill no more was able to dissolve. Slowly the solution was diluted with distilled water which precipitated small needles of an orange-red compound. After washing thoroughly, this was diazotized at 0 deg C., producing a bright yellow compound that flashes a lot like DDNP.

An experiment looking at the catalytic potential of copper salts was performed earlier.

Experimental 2:

1. 0.25 grams of picric were added to 10 ml water and brought to 80 deg C. Then, 0.6 grams of ascorbic acid were added, which produced no obvious colour change. Using NaOH solution, the picric/ascorbic was adjusted to a pH of around 7, which also did not result in any significant change in colour within 5 minutes. A few crystals of copper(II)sulfate (~5 mg) were added, which immediately produced a clear solution changing rapidly in colour, going from yellow, to orange, to dark orange red, finally resulting in a fine orange-red precipitate at the bottom of the beaker (Seemingly too much to be explained by the copper addition alone, though would need to be weighed). Out of curiosity, another 0.6 grams ascorbic acid was added in small increments (everytime adjusting the solution again to pH7. Strangely, after only a few spatules of additional ascorbic, the solution began producing large amounts of some odourless gas, foaming significantly and changing to a very dark red-orange colour, with only a small amount of precipitate, that most likely is copper powder. The dark red-orange solution was acidified using sulfuric acid, resulting in a much lighter orange solution and a precipitate of brownish crystals, that on closer examination proved to be unreacted picric acid.


2. 0.6 grams of ascorbic acid was dissolved in 10 ml water and brought to around 80 deg C.. Then added a few crystals of copper(II) sulfate, there was no colour change. Then adjusted the solution to pH of 7, which produced a slightly opaque solution with slight orange sheen (Probably copper(I) oxide). A spatule of salicylic acid was added, resulting in no colour change. Finally, a spatule of picric was added, immediately going from orange to dark orange-red again.



It seems from the latter experiment that copper salts are truly increasing the reduction rate at a pH of 7. Very interesting, the question is what formed here, how specific the reduction is and how the conversion efficiency is (if it can truly acts as a catalyst) Probably not since presumably conversion to copper powder is also happening at this pH and temperature, explaining the unreacted picric recovered.

This may well be the weirdest way to produce picramic acid, though it is not entirely sure yet the compounds isolated are truly picramic and DDNP, a dinitro nitroso phenol could behave similar energetic maybe. Also not entirely sure what the bright orange-red compound is, presumably picramic, though the melting point seems to be on the low side. I've wondered if a picramic picrate salt could exist, since picric itself is a very strong acid and what it's properties would be. Thorougly washing the DDNP produced with warm water and measuring yield could provide an answer to that. Alternatively, maybe dehydroascorbic could condense with the picramic produced or something, which on diazotization reverts back to DDNP.

All in all, the ligand orientation mechanism seems most plausible for the reduction, since catalytic amounts of copper do not seem to have the same results till now, although this would need more experimentation. Alternatively the reduction may be catalyzed by copper(I) itself somehow.

[Edited on 1-2-2018 by nitro-genes]

IIRC, Axt posted a reference long time ago regarding the reduction of picric to picramic using just iron and water at 80 deg C. (somewhere in the picramic thread) and I think Rosco posted a brief reference regarding reduction using zinc metal/ammonia in boiling methanol for half an hour, so it might be possible.

No chemist myself and the specificity and precise reactions occuring during these different nitro reductions are still largely a mystery to me, so sorry if this seems like a load of unscientific crap:

Bechamp seems to work fine for most mono nitro arenes, though I thought in aquous solutions at least, there is no specificity for the reduction of a specific nitrogroup in the case of polynitroarenes. Not sure why this is though, maybe due to proton-tranfer or hydrolysis reactions, there seem to be 2 independ pathways during nitro reduction, one via direct reduction and one via condensations of partially reduced intermediates, maybe this is one of the reasons?

http://pubs.rsc.org/-/content/articlelanding/2014/cp/c4cp043...

In anhydrous solvents it seems possible to selectively reduce 1 single nitrogroup using nascent hydrogen, for example 1 nitrogroup of TNT can be reduced specifically using gAA/Fe(0). Since I lack any glacial acetic, I tried the reduction of TNT in ethanol/HCl/steelwool once, which seemed to work in producing 2-amino 4,6-dinitrotoluene (although not very pure). If the same could work for polynitrophenols...no idea...might be possible... one thing that might be different for nitrophenols is that the OH group itself may also form salts (that may precipitate from most OTC solvents for example) and maybe also change the way partially reduced intermediates would behave during the reduction, for example a nitrosophenol would be in equilibrium with a quinone-oxime. Maybe the reactions occuring may also largely depend on the pH and solvent used, so I wonder if one set of conditions would apply to different nitro arenes anyway.




Also did another experiment with the Cu(II)/ascorbic acid reduction of picric again. After adding the ascorbic acid to the copper sulfate/sodium picrate solution, the greenish precipitate was immediately filtered off and thoroughly washed with cold water after it formed (so without any further heating applied). It dissolves in 10% HCl easily, but only picric can be isolated from this and not a trace of picramic acid. So it seems likely a complex of ascorbic/Cu(I) or Cu(II) and picric is formed first which goes through some internal reduction forming picramic acid. Very interesting, although maybe not economical, never seen picramic of this purity before. It would indeed also be interesting to see if iron(II) instead of Cu(II) would behave similar under these conditions.

I'd like to scale this up a bit to be able to measure yields etc, though I'm hesitant due to the odourless gas produced during heating. My first guess would be carbon dioxide from decarboxylation of some dehydroascorbic acid derivative, but not sure if it could be something toxic as well (carbon monoxide, or some organocopper compound).

"Crystal structure of a copper complex of 2-carboxypentonic acid; A decomposition product of dehydroascorbic acid (DOI10.1039/dt9870002905)

Abstract
In acidic aqueous solution and in the presence of copper(II), ascorbic acid is rapidly oxidized to dehydroascorbic acid, which rearranges to give the branched-chain dicarboxylic acid 2-carboxypentonic acid (1,2,3,4-tetrahydroxybutane-1,1-dicarboxylic acid)(H3cpa). The ion cpa3– is sequestered by copper(II) to produce the insoluble crystalline product [Cu9Cl2(cpa)6(H2O)3]2–·xH2O"

Any ideas what kind of gasses could be produced? Is gas production/foaming also observed during the making of nano copper from Cu(II) and ascorbic for example?

[edit] What do you know, an OTC source of copper(II)sulfate! People actually just pour this down the drain?

https://www.doitbest.com/products/477796

[Edited on 4-2-2018 by nitro-genes]

After some experimentation it seems possible to use copper salts as a catalyst for the reduction of picric acid to picramic acid. Between 0.0001 and 1.1 molar equivalents (relative to picric) of a salt containing the Cu(II) or Cu(I) ion can be used as a catalyst for the reduction of picric acid (or any of its picrate salts) solutions in water to form picramic acid or any of its picramate salts in high purity and yield. The reaction can be performed at a temperature of 20 deg C. to 100 deg. Celcius, (preferentially between 60 and 90 deg Celcius), and at a pH of about 2-12 and using a variety of reducing agents, including ascorbic acid, glucose (and other reducing sugars), sulfites and other common reducing agents known in the art, for which between 1.5 to 6 molars are preferentially used for each mole of picric. Depending on the reducing agent and temperature utilized, reaction times can vary from 5 minutes to 8 hours. Depending on the reducing agent utilized, an aquous solution of the reducing agent is preferentially added to an aquous solution of picric acid or any of its picrate salts, in other reactions the reducing agent is directly added to the picric acid or picrate solution, and the catalyst solution containing Cu(II) or Cu(I) salts is gradually added over timeperiod of the entire reaction. The picramic acid obtained is of exceptional purity and can be directly used to produce diazodinitrophenol (DDNP) by diazotization using 5-20% HCl and a nitrite salt at 0-5 deg. C.. The light yellow DDNP thus produced is obtained as spherical crystal agglomerates of high bulk density, that can be directly used for primer applications.

It would also be interesting to see whether something like styphnic acid can similarly be reduced.

[Edited on 5-2-2018 by nitro-genes]

If you are getting picramic acid from an acidic reaction mixture down to a pH of 2, that is very interesting and contradicts a lot of literature that reports an alkaline condition picrate as a requirement. Maybe it is only the presence of some alkaline picrate in part as a component in a partially neutralized picric acid reaction system that is required, and the pH for the reduction mixture of a "partially neutralized" reaction system may still be acidic and not interfere with reduction of that component that is already neutralized. Glycine can be used to greatly increase the solubility of picric acid in an excess of picric acid required for formation of Diglycine Picrate, which would be an acidic mixture or greatly more soluble and concentrated picric acid. Glycine also complexes copper so this could be an interesting material for experiments as a possible additive that could possibly make the catalytic copper salt even more active. The copper glycine complex is reportedly not stable above 90 C so that factor should rule out use of an excessively hot reduction mixture. See attached articles.

Information from the articles indicates that the glycine cobalt complex could also be a candidate regenerable catalyst for reduction. Reportedly the Glycine Cobalt Picrate itself is not unstable to boiling heat. It is unknown how soluble is the different lower oxidation state cobalt glycine picrate or how soluble or reactive are either oxidation state in reaction with ascorbic acid or ascorbate. Here what is contemplated is that the soluble glycine cobalt picrate itself could possible function as a regenerable catalyst which could act as a reducing agent on the remaining 99% bulk of the picric acid or soluble picrate salt such as the sodium or magnesium salt.

The discussion posts related to reduction schemes for picric acid / picrates to picramic acid / picramate should be exported to the dedicated thread for the topic to keep the subject matter in one place. Here is a post link for that thread. Historical references being researched on some of these niche topics have for years been resulting in Google search hits for this science forum as a kind of repository for "obscure and interesting" collected references and experiments so we should keep things organized.

http://www.sciencemadness.org/talk/viewthread.php?tid=433&am...


Ferrous ammonium sulfate Mohr's Salt is another likely candidate for a reducing reagent, the double salt or complex of Ferrous Sulfate and Ammonium Sulfate, which upon oxidation to Ferric Ammonium Sulfate expels one Ammonia. References to "the ammoniacal liquid" without further details in historical references to reductions involving iron indicate that ammonia was used as a base but do not specify a step by step or reaction specifics to know exactly how ammonia was used.

A second reducing agent like ascorbic acid / ascorbate, or a sulfide added gradually can be used to regenerate the "ous" lower oxidation state iron or copper or other transition metal salt, from its oxidized "ic" higher oxidized state so it acts as a regenerable catalyst and the reduction continues so long as the catalytic salt does not further react to form some inactive byproduct, or precipitate as an insoluble oxide, which in effect would "poison" the catalyst and require more of it to be added to continue the reduction.

Glycine is something that has caught my attention as possibly useful as a chelating agent that has potential value because glycine is also reactive with picric acid, forming diglycine picrate and increasing the solubility of picric acid by that association, particularly in solution where the picric acid is in excess.

Glycine also complexes copper and is known to inhibit the precipitation of some metal hydroxides up to an alkaline condition of 10.3 pH. Because of the activity of glycine it could have usefulness as a protector of the catalyst by inhibiting precipitation of an inactive byproduct and counter the tendency of the catalyst to be deactivated by undesired further reaction. The glycine could also serve as a buffer for the pH or as part of a buffering scheme in mixture with other salts of possible benefit also.

There is probably an optimum process scheme that varies somewhat for the particular transition metal salt being used. Solubility of the reactants is an important factor and greater solubility is generally better because more concentrated solutions generally react more quickly and completely.

Attachment: diglycine picrate J. Biol. Chem.-1912-Levene-285-94.pdf (554kB)
This file has been downloaded 467 times

Attachment: glycine picrate related J. Biol. Chem.-1958-Selim-157-62.pdf (271kB)
This file has been downloaded 523 times


[Edited on 2/5/2018 by Rosco Bodine]

Quote: Originally posted by nitro-genes

After some experimentation it seems possible to use copper salts as a catalyst for the reduction of picric acid to picramic acid. Between 0.0001 and 1.1 molar equivalents (relative to picric) of a salt containing the Cu(II) or Cu(I) ion can be used as a catalyst for the reduction of picric acid (or any of its picrate salts) solutions in water to form picramic acid or any of its picramate salts in high purity and yield. The reaction can be performed at a temperature of 20 deg C. to 100 deg. Celcius, (preferentially between 60 and 90 deg Celcius), and at a pH of about 2-12 and using a variety of reducing agents, including ascorbic acid, glucose (and other reducing sugars), sulfites and other common reducing agents known in the art, for which between 1.5 to 6 molars are preferentially used for each mole of picric. Depending on the reducing agent and temperature utilized, reaction times can vary from 5 minutes to 8 hours. Depending on the reducing agent utilized, an aquous solution of the reducing agent is preferentially added to an aquous solution of picric acid or any of its picrate salts, in other reactions the reducing agent is directly added to the picric acid or picrate solution, and the catalyst solution containing Cu(II) or Cu(I) salts is gradually added over timeperiod of the entire reaction. The picramic acid obtained is of exceptional purity and can be directly used to produce diazodinitrophenol (DDNP) by diazotization using 5-20% HCl and a nitrite salt at 0-5 deg. C.. The light yellow DDNP thus produced is obtained as spherical crystal agglomerates of high bulk density, that can be directly used for primer applications. [Edited on 5-2-2018 by nitro-genes]

Did some experiments this weekend:

First I tried to determine the minimal temperature for the reduction on a very small scale using 1 mole eqvt of Cu(II)sulfate and 3.1 mole eqvts of ascorbic acid. At 40 deg C., the complete formation of the initial green precipitate (as evident by a nearly colourless filtrate) takes about 15 minutes. At 40 C, there is hardly any reduction going on and the precipitate remains bright green if kept at this temp for 60 minutes. The temperature was then raised gradually to around 60 deg C., gas formation became evident and within another 20 minutes or so the precipitate attained a more yellowish colour. So 60 C. seems minimal for the redution to take place and is concomitant with the gas formation.

The gas is probably carbon dioxide, resulting from oxidation of ascorbic to dehydroascorbic, which hydrolyses to 2,3 diketo gulanic acid, which decarboxylates further to various other products (attachment), which probably explains the brown colour obtained from prolonged heating of the reaction.




Also did the reduction on a slightly larger scale, determining yield of the picramic obtained and tried the reaction with 1 and 0.5 molar eqvts of copper(II)sulfate to try and minimize the amount of copper salts needed.

Exactly the same reaction conditions were used:

2.3 g picric was suspended in 25 ml dH2O and warmed to 60 deg C. A 10% NaOH solution was gradually added to a pH of 7. When done slowly, no pH paper is needed, the picric itself functions as a reliable indicator of pH. Upon each addition of the NaOH, an orange colour is observed, which quickly reverts back to a bright yellow again when neutralized by the undissolved picric. When all the picric has dissolved, a few extra drops of NaOH are added until a permanent slight orange colour is obtained. Then Cu(II)sulfate pentahydrate was added (2.5 g in Exp 1 vs 1.25 g in Exp 2), which dissolved into a clear bright green solution. Then 5.46 g (3.1 molar eqvt) of ascorbic acid was added. After a few minutes or so, an abundant bright green precipitate started to form and a mild exotherm could be observed. The solution climbed to 70-75 deg C on its own and was kept there for 30 minutes. The precipitate changed to a yellow-greenish colour, was filtered and washed 3 times using cold water.

After weighing and drying, the precipitate was dissolved in the least amount of boiling 20.2% HCl, (about 15 ml). Upon dilution with icecold water to 100 ml, small bright orange-red needles of putative picramic acid were obtained, which were washed, dried and weighed. The filtrates were saved and copper(II)oxide can be retrieved from them by basifying with NaOH and boiling.

Results:

Exp 1: 2.22 g of putative copper (I/II) picramate, 1.32 g picramic acid (66.3% yield).
Exp 2: 1.65 g of putative copper (I/II) picramate, 1.00 g picramic acid (50.2% yield).

The filtrate of the reduction experiment 1 was saved and a dark brown in colour (probably from dehydroascorbic decomposition products), upon basifiying with NaOH, the solution became a very dark red colour (maybe some diamino nitrophenol) and some residual Cu(I)oxide was present as a light yellow precipitate.

Any ideas how to maximize the yield further? In both cases, after 30 minutes into the reduction, there was still lots of gass formation. Maybe it just needs longer reaction time? Not sure if the decarboxylation observed is a necessity for the reaction to proceed and is a direct indicator of the formation of dehydroascorbic acid, which probably leaves the complex and allows another ascorbic acid to complex. Depending on the stability of the copper picramate, maybe keeping at 60 C. until no more decarboxylation is observed results in higher yield. Maybe larger dilution of the reaction also helps.

[Edited on 11-2-2018 by nitro-genes]

Considering that the picramate of copper is insoluble it would rule out that copper can be used as a regenerable catalyst because ALL of the copper in equimolar proportion is subject to sequestration as insoluble picramate byproduct. The copper has catalytic activity with respect to the ascorbic acid and to keep shifting the equilibrium for the reduction by precipitation of the picramate as a cuprous salt. But in terms of efficiency as a catalyst to be regenerated, iron or manganese or cobalt or nickel salts would be better choices in terms of potential efficiency for the reduction, because the picramates for those metals should not precipitate as occurs for the copper picramate.

Girard, Comptes Rendus, March 7, 1853 pg. 421

If your product of reduction is cuprous picramate then half of the reducing power of the copper is being sequestered in the product.
At the end of the reduction reaction, copper values in the spent reaction system would desirably be the +II cupric salts, but that may not be the case if the precipitate is instead the cuprous picramate. It may be unavoidable that it happens that way if that is what is occurring and changing the pH to more alkaline may or may not help. This reaction involves a set of factors that are variables difficult to predict what reaction wins out, so that conditions can be optimized. Guesses and experiments are ahead.

Suggest to increase the amount of CuSO4 150% of the 2.5 g of Experiment 1 and compare and then a second experiment with the same 150% increase applied to the ascorbic acid as well. As a variation the added amount of CuSO4 could be mixed with the ascorbic acid and the mixed reagent added gradually as a 3rd experiment. Adjusting the pH conditions of the reducing reagent solution, or adjusting the pH of the picrate reaction mixture with a buffer like magnesium oxide in suspension in the stirred mixture is another idea. The addition of a chelating agent like glycine is another idea for possibly increasing the efficiency of the copper catalyst which appears to be tied up as a reagent as much as it is acting as a catalyst. It seems it is capable of acting as a catalyst but is also being consumed as a reagent, and sequestered as a byproduct, so that an additional available quantity that can regenerate may improve the yield.

The reducing "power" of ascorbic acid is greatest for the free acid at about pH 2 and between 2 and 4.1 is the most active range if what I seen reported is accurate. In that acidic range it appears the reducing power is effectively double what is the reducing power for ascorbate ion in an alkaline pH range, where half the available reducing hydrogen is already consumed in association with a base like sodium if the ascorbate being used is sodium ascorbate. The available reducing hydrogen reduction "capacity" then exists for a first order "mono"-dehydroascorbate intermediate that is further reduced to dehydroascorbate which has formed from a loss of TWO reducing hydrogens with respect to ascorbic acid. Evidently the distinction between "protonation" by an acid hydrogen and an added hydrogen by "reduction" is a variable that depends entirely on pH. Evidently in an acidic pH reduction reaction system a +2H reducing power exists for free ascorbic acid. At intermediate pH there is a +1H reducing "potential" and at a strongly alkaline pH there is a zero reducing potential with the ascorbic acid appearing as an inactive di-acid salt.

The pH factor is not conclusively identified to be strongly alkaline for the reduction reaction system of where picric acid / picrate is the target for reduction, and there may be a tradeoff for the best reaction system where a slightly acidic reduction mixture benefits increased reducing power for ascorbic acid, but an increased alkalinity may benefit the reduction and selectivity for production of the target picramate. Somewhere is an intermediate range of pH and temperature and concentration that "just works" best for the
process using a combination of reagents. Working out what is that optimum combination for the best result is a lot of trial and error until it gets "dialed in" what is the "secret formula" that works best.

Have you been curious enough to do a glowing splint test for oxygen on the gas?

I have a suspicion your colorless gas evolution may be oxygen. And if it is oxygen then it
will take a lot less ascorbic acid for the reduction according to theory than would be required for reduction by a sulfide. My suspicion is that the ascorbic acid is manifesting the surgical precision of an enzyme, with one or a pair of ascorbate ions operating directly on the N=O2 group and cleaving, splitting that radical without actually combining with the freed oxygen, attaching an H or H2 hydrogens directly to the N to form the (di-) hydrogenated N, (NH2) amino radical. It would be a small wonder if that is indeed the mechanism occurring.







Attachment: Chemical_Gazette 1853 article Aime Girard picramic acid.pdf (284kB)
This file has been downloaded 483 times

After reviewing the descriptions for pH affecting ascorbic acid, and noting the insolubility of copper picramate, it seems that use of copper may be better applied in the isolation and purification step where a soluble picramate in solution with other reduction reaction byproducts is mixed with a soluble copper salt to precipitate the insoluble copper picramate filtered out to isolate the picramic acid value as the insoluble copper salt, washed and redissolved in an acid to resolve the solution into its separated components with free picramic acid precipitated from a solution of the acid salt of copper that remains dissolved.

Magnesium seems to be the most likely candidate for a picrate to be reduced by ascorbic acid due to the high solubility of the magnesium salt and the limited alkalinity which would be inherent due to the low solubility of magnesium hydroxide and the limit of a 10.3 pH that would occur for a solution of magnesium hydroxide alone. A solution of magnesium picrate would be much closer to neutral, and would be an excellent target in the pH range "sweet spot" for reduction by free ascorbic acid alone or partially or fully converted to a sodium or magnesium ascorbate. HCl would be cheap acid that should work fine for adjusting pH for the reduction or for the later isolation and purification, that could probably done with or without using a copper salt to obtain an insoluble picramate intermediate.

Also still of interest is something I mentioned before about the reported increased solubility of picric acid associated with glycine to form diglycine picrate with the picric acid in excess to possibly more equivalent what would be a mono-glycine picrate. The pH of such a mixture would definitely be acidic and might be susceptible to reduction by ascorbic acid directly, and may precipitate free picramic acid directly as fast as it forms ....if there is not a similar association with glycine causing increased solubility for picramic acid....then the free picramic acid should precipitate directly in pure form, from an efficient ascorbic acid reduction that may be optimized in the acidic pH reduction scheme. This is theoretical and would depend on the selectivity for reduction to picramic acid and would it proceed at an acidic pH similarly as occurs in the experiment where a copper salt is used. It would definitely be an experiment worth trying since it could be exactly the reaction condition that is desirable if it does work as anticipated. It could be a model lab method for easy synthesis of pure picramic acid as a one pot synthesis, and may possibly even be extended further to a one pot synthesis of DDNP.

What I am thinking about as an extension even further of the "one pot" reaction scheme possible is the salicylic acid or possibly aspirin starting material which is first sulfonated and then under mild conditions in dilute solution nitrosated and nitrated to picric acid quantitatively....not even isolating the picric acid...add glycine and proceed with an ascorbic acid addition, followed by sodium nitrite or the variant for diazotization using a nitrate and copper wire. DDNP could possibly be obtained by a much simpler path of reactions under milder conditions than the more usual and better known methods.

[Edited on 2/12/2018 by Rosco Bodine]

Ascorbic acid seems to be the exception to the widely "accepted" narrative in the literature about reduction conditions for producing picramic acid, all of which references I have ever seen are reporting alkaline condition as a requirement. Once again it seems the highlighter is being applied to "textbook" generalizations that have a SM footnote inserted regarding a HUGE exception to the "rule" having been identified by this science forum. Go figure. So much for academic preeminence "authority" meeting up with test tube anarchists leading a molecule rebellion. It's a veritable ionic uprising.

Joking aside....Ascorbic acid is reportedly heat sensitive which is why cooking foods have reduced vitamin C levels compared to raw uncooked food. So even though the activity and reaction rate increases with temperature, there is a half life for ascorbic acid that shortens with elevated temperature, and it is not heat stable. I'm not sure what that parameter is exactly. Also it occurs that most of the studies about ascorbic acid are in vivo and there can be different behavior in vitro. So there is not necessarily a specific rule that what happens in a biological reaction system also happens the exact same way in a laboratory synthesis. In a living organism other enzymes interact with ascorbic acid to do various things so isolating what occurs for ascorbic acid in a reaction in a flask is a different kettle of fish.

[Edited on 2/13/2018 by Rosco Bodine]

Are basic conditions really a necessity for the reduction of picric to picramic? What would happen if a solution of picric acid in gAA or other solvent would be reduced with H2S, stannous chloride, zinc or iron powder e.g., never seen any references for this at least. My uneducated thinking (without much reading) was that the impurities arising from the sulfide reduction are actually due to the high pH employed during the reaction, largely due to overreduction or meisenheimer complex formation (HS-) and hydrolysis of a benzoquinone oxime intermediate. The economy of an aquous solution and a cheap industrial reagent like NaS are just a logical choice, so not a lot of research would have been done if such a proven synthesis exists.

The chemistry of ascorbic and dehydroascorbic at different pH levels is immensly complex, it indeed seems that ascorbic acid itself can form various products when heated in aquous solutions, depending on pH and air exposure. This was why I was contemplating running the reaction at the lowest temp possible for the reduction, namely 60 C.

"Degradation of Ascorbic Acid in Aqueous Solution
Jian-Ping Yuan, and Feng Chen*
Department of Botany, The University of Hong Kong, Pokfulam Road, Hong Kong
J. Agric. Food Chem., 1998, 46 (12), pp 5078–5082"

Considering the additional presence of Cu(II), Cu(I) being able to complex with some of these products and the formation of condensation products, schiff bases and additional reducing agents from dehydroascorbic and decomposition products, it seems a small miracle picramic acid can be obtained in this purity and yield.

Something I would still like to test are other transition metals (especialy iron) and whether the specificity of the reduction is largely a feature of Cu(I) by using some other reducing agent. I can't see a direct reason though why Cu(I) would display such specificity. Anyway, if not, this would also be further evidence the entire reaction may indeed be facilitated by the ligand coordination of ascorbic mediated by an insoluble Cu(I)(ascorbic)picrate complex. Although far from proven at this point, it would be a sexy mechanism (Admittedly, I'm biased), the Cu(I) bringing those reductive ascorbic hydrogens in perfect position for the reduction, almost like an enzyme would indeed.

https://en.wikipedia.org/wiki/L-ascorbate_oxidase

[Edited on 13-2-2018 by nitro-genes]

Quote: Originally posted by nitro-genes

Are basic conditions really a necessity for the reduction of picric to picramic?

Quote:

"Degradation of Ascorbic Acid in Aqueous Solution Jian-Ping Yuan, and Feng Chen* Department of Botany, The University of Hong Kong, Pokfulam Road, Hong Kong J. Agric. Food Chem., 1998, 46 (12), pp 5078–5082" Considering the additional presence of Cu(II), Cu(I) being able to complex with some of these products and the formation of condensation products, schiff bases and additional reducing agents from dehydroascorbic and decomposition products, it seems a small miracle picramic acid can be obtained in this purity and yield.

Hmmmm how about glycine ? Take a shot in the dark...and listen...maybe then a scream is heard in the distance

Magnesium is even more sure to make somebody holler.

http://www.sciencemadness.org/talk/viewthread.php?tid=26572&...

FR2106904 French patent DDNP

Thanks for the patent and all the other information, not much time is left in the day and most of what is left is spent in "the lab". The plan is to eventually start reading as well.

Did some small scale experiments again:

1. The initial light green precipitate formed from the copper/ascorbic reduction seems to contain unreacted picrate and Cu(I) ions

2. Although the initial complex (imediately after it formed) contains hardly any picramic, keeping at 40-50 seems to produce picramic acid eventualy, while no gas formation is observed, Either this is occuring at a rate too slow to be noticed, or the gas formation is not a requisite for reduction at all, in which case a longer heating period at 50 deg C could produce a purer product in higher yield. (To be tested at slightly larger scale)

3. After the heating period, the yellow-green precipitate also seems to contain Cu(I), probably a complexed or uncomplexed Cu(I)picramate. (After HCl treatment and adding NaOH, a yellow precipitate of Cu(I) oxide is formed, adding HCl again and oxidizing the Cu(I) with nitrous acid and adding NaOH again, greenish blue Cu(II)hydroxide is formed, which upon heating forms black Cu(II)oxide). This explains why at least 2 molar equivalents of Cu(II) salts relative to picric are needed, and this would indeed mean that a larger amount of ascorbic acid is needed (>3.5 eqvts), like Rosco suggested.

4. A strange observation, somehow, the dark brown filtrate after the reduction at 70 deg C, reacts with NaCl to produce a clear orange solution, while no Cu(I)chloride is precipitated. This suggest all copper is sequestered in the precipitate and a slight excess may produce higher yields. Although the 0.01 molar scale is perhaps too small to produce accurate yield measurements, the ratio of the putative copper picramate and picramic obtained after HCl treatment suggests that not all the copper picramate truly is copper picramate. Perhaps a part of the picramic remains in the dilute HCl solution, or maybe a complexed copper picramate is obtained, or another copper salt is present from decomposition products of the ascorbic, or unreacted picrate is still present, not sure. Regarding the reaction with Cl-, would a dinitro benzoquinone or an oxime or something be able to react with Cl-?

5. Cu(I) and Cu(II) chloride are very soluble in excess HCl, stirring the copper picramate obtained with 10% HCl at 20-60 deg C is enough to displace all the copper. About 1 gram of the putative copper picramate was added to a beaker and with slight heating, 10% HCl was added in small increments. The displacement can be followed by watching the edge of the fluid level in the beaker. As long as a milky greenish colour is observed it needs more HCl. In total 1 gram copper picramate needed about 12.5 ml of 10% HCl. Amazingly, from an amorphous copper picramate, reasonably large and compact crystals of picramic seem to be formed this way, that take up much less volume then picramic obtained from the boiling HCl-dilution method I posted earlier. It collects as a dense layer of beautiful pom-grenade coloured crystals at the bottom of the beaker, which can be decanted to near dryness after only few minutes of standing. If a high yield synthesis is realized, it might be enough to simply wash and decant the copper picramate a few times and then add HCl to directly yield picramic without any filtering and drying steps.

6. Copper(I)picramate seems to be oxidized in the air on longer storing. ALthough this doesn't seem to affect the picramic obtained, it seemed Cu(I)picramate needed less HCl to displace the copper than a longer stored and presumably oxidized Cu(II)picramate.

[Edited on 17-2-2018 by nitro-genes]

I call it a day, lump-and-dump-easy-OTC-picramic...whoohoooo

Suggestions welcome

[Edit, new file uploaded, some mistakes corrected]




[Edited on 5-3-2018 by nitro-genes]

Attachment: Reduction of picric to picramic using copper sulfate and ascorbic acid - Copy.pdf (885kB)
This file has been downloaded 692 times

Some mistakes were present in the first version, uploaded a new version in the previous post.

@Boffis

Reprtedly, the sulfide methods do give higher yields, though this is an easy reaction indeed using OTC chemicals. Higher yields may be possible somehow, though optimizing such undescribed reactions is very labour and material intensive, especially as an amateur. Would be interesting to hear if you have some additional ideas how to possibly increase yield further. Maybe a buffer solution at a pH of 3 like rosco suggested and gradual additions of ascorbic would be an improvement. Any suggestions?