Sciencemadness Discussion Board

1,3,5-trinitrobenzene preparation from trinitrotoluene

Fery - 2-2-2022 at 06:03

I inherited a little of TNT which I believe was used in a granite quarry. As I'm not interested in explosives the only one usage was a preparation of 1,3,5-trinitrobenzene from it.
Searching sciencemadness forum I found these helpful threads/posts including also few pictures as trinitrobenzene easily deprotonates into red colored compounds:
https://www.sciencemadness.org/whisper/viewthread.php?tid=29...
https://www.sciencemadness.org/whisper/viewthread.php?tid=26...
https://www.sciencemadness.org/whisper/viewthread.php?tid=14...
https://www.sciencemadness.org/whisper/viewthread.php?tid=26...
https://www.sciencemadness.org/whisper/viewthread.php?tid=73...
https://www.sciencemadness.org/whisper/viewthread.php?tid=18...

The trinitrobenzene could be used for charge-transfer complexes with electron-rich arenes, or reduced to 1,3,5-triaminobenzene (precursor to phloroglucinol, but I have enough of phloroglucinol). Maybe it could be also used as an indicator for strong bases?

I followed the orgsyn method at 1/20 scale

http://www.orgsyn.org/demo.aspx?prep=CV1P0543
http://www.orgsyn.org/Content/pdfs/procedures/CV1P0543.pdf

Attachment: CV1P0541.pdf (299kB)
This file has been downloaded 248 times

http://www.orgsyn.org/demo.aspx?prep=CV1P0541
http://www.orgsyn.org/Content/pdfs/procedures/CV1P0541.pdf

Attachment: CV1P0543.pdf (299kB)
This file has been downloaded 204 times

18,2 g TNT was weighed (80 mmol), unknown purity, industrial origin, very old, very likely used in quarries.
27,0 g Na2Cr2O7.2H2O (90 mmol) purity 99,5%
H2SO4 96% pure 180 g (weighed it on scale, should be 98 ml, volume not measured)

18,2 g TNT was crushed into smaller pieces (read further how fine it should be crushed) and transferred into 250 ml FBF. Magnetic stirbar was inserted, the flask was put into an empty bath from plastic dish and this was placed on magnetic stirrer. 180 g 96% H2SO4 was added, thermometer inserted and magnetic stirring commenced. 6,5 g of Na2Cr2O7.2H2O from total 27,0 g was slowly added during first 10 minutes at which T raised from 15 to 25 C in t+10 minutes and later to 50 C in t+15 minutes. At that moment 10 C cold water was added into bath, T in reaction flask fall to 45 C at t+18 minutes. As T should be kept at 45-55 C during whole oxidation process the water bath was quickly siphoned out. A little of dichromate added in small circa 2 g portions during whole oxidation. When T raised to 55 C a little of cold water was added into bath but now only circa 1 cm high layer in the bath (level of reactants in the flask circa 5 cm). T during reaction was kept at 45-55 C, when fall to 45 C, 2 g portion of dichromate added. Addition of dichromate lasted 1,5 h and water had to be repeatedly added into and drained from bath, T always kept at 45-55 C. During second half of the addition the mixture frothed a little (? impurities in TNT ?) and become thick, but powerful stirrer and medium size stirbar were able to continue stirring.
After the addition of whole dichromate, warm water was added into bath and reaction continued stirring at 45-55 C (thermometer submersed into reaction mixture) for 2 hours, during which the water in bath was repeatedly replaced with fresh warm water (alternatively aluminium or glass bath and heating could be used too).
The reaction mixture was poured onto 200 g of snow (better than crushed ice due to much smaller particles, note that every overheating could decarboxylate the trinitrobenzoic acid) while stirring with inserted thermometer, final T 20 C. A little more of snow added (circa 25 g) so the final T 0 C.
Filtered on sinter. 20,6 g wet crude product obtained. Air dried. The product 1,3,5-trinitrobenzoic acid still not completely dry the next day, weighed 17,8 g (there is no need to dry the crude product).
Circa 3 g of 2-3 mm pieces of TNT stayed in bottom of reaction flask - crushing TNT at the beginning into more fine particles is highly recommended. It seems that crushing TNT into mixture of dust with also too big pieces circa 3 mm at the beginning was not enough and it had to be crushed to only fine particles completely as these 3 mm pieces stayed unreacted (3 g of these pieces recovered from 18 g of reactant).
15 % NaOH solution was prepared by dissolving 4,0 g NaOH in 26 ml H2O and cooled down to 0 C.
Crude product (trinitrobenzoic acid) was partially dissolved in 100 ml water and warmed to 35 C. It was then neutralized with 15% NaOH solution at 35 C (thermometer inside) dropwise by pipette and stirring in cold water bath, circa 1/3 of the NaOH consumed when color changed to red, then immediately 2 drops of acetic acid added and color reverted back to yellow. The thermometer was submersed into the neutralized mixture during whole time as overheating could decarboxylate the acid. Too low temperature is also undesirable as then the solubility of the acid is low.
The mixture was gravity filtered through filter paper into 250 ml FBF. 3,5 ml acetic acid added into the filtrate. Condenser attached and flask carefully slowly heated (CO2 effervescence) and then refluxed for 3 hours.
Cooled in cold water bath while stirring by hand until product solidified. Gravity filtered on filter paper, air dried.
Crude 1,3,5-trinitrobenze product obtained in yield 7,3 g.
Product dissolved in 5,7 g acetic acid under reflux (yeah too much soluble in this solvent).
Cooled outside (winter 0 C temperature) at which also acetic acid crystallized so it was necessary to repeat the reflux and then cooled only to room temperature 20 C.
Solubility seems to be approximately 30-40 g of trinitrobenzene in 100 g of acetic acid at 20 C.
Crystals vacuum filtered on sinter, washed with little of acetic acid, sucked to dryness. Air dried. The color of the product was not satisfactory (a mixture of paler crystals with darker pieces).
Into the mother liquor was repeatedly (5x) added circa 1 ml of water at which substantial amount of impure trinitrobenzene recovered.
So maybe it could be recrystallized from diluted acetic acid but this was not tried. Solubility in diluted acetic acid should be lower and the diluted acid (maybe 90, or 80, or 70% ???) should not crystallize at 0 C so cooling to lower temperatures could crystallize more of the product from the solvent.
So far 5,0 g of trinitrobenzene obtained and some recovered from acetic acid mother liquor.
The 5,0 g trinitrobenzene was recrystallized from 40 g methanol (weight on scale, volume was not measured) - dissolved at reflux temperature, cooled to 0 C outside (winter temperature), vacuum filtered on sinter at 0 C outside, washed with little of 0 C cold methanol on sinter, sucked to dryness. Air dried for 2 days, second day the weight did not decrease anymore.
Yield 3,3 g.
m.p. 121,5-123 C (lit. 123.2 °C)
Recovered crop from acetic acid mother liquor was dissolved in methanol mother liquor and volume reduced to half by evaporation on hot plate. After cooling down to 0 C, totally 2,4 g impure crop obtained which was recrystallized from 20 g methanol. 1,6 g trinitrotoluene obtained with somewhat discoloration, but m.p. was still good, this crop seemed to be still pure enough (just a little discolored).

photos will follow soon

Fery - 2-2-2022 at 07:46

reactants

IMG_20220125_175942_0.jpg - 49kB



TNT + H2SO4, note the TNT had to be crushed to smaller pieces

IMG_20220125_181319_8.jpg - 37kB IMG_20220125_181638_4.jpg - 36kB IMG_20220125_181652_7.jpg - 31kB



the T during oxidation should be kept in 45-55 C using cooling water bath and the rate of addition of dichromate

IMG_20220125_183228_1.jpg - 23kB IMG_20220125_183716_8.jpg - 33kB IMG_20220125_185017_9.jpg - 37kB



the reaction at the end become thick and frothy, it was then poured onto snow

IMG_20220125_214032_0.jpg - 20kB IMG_20220125_214042_6.jpg - 37kB



1,3,5 trinitrobenzoic acid precipitated and the mixture was then further cooled down to 0 C by addition of little of snow

IMG_20220125_214759_6.jpg - 46kB IMG_20220125_214805_2.jpg - 51kB IMG_20220125_215618_7.jpg - 31kB



the acidic solution was then filtered on sinter and washed with ice cold water to remove traces of Cr3+, note the Cr3+ is green and after washing it out the washings become yellow perhaps due to trinitrobenzoic acid coloration, do not wash more than necessary, but wash out all Cr3+ (green color, testing by precipitation with diluted cold NaOH solution, trinitrobenzoic acid reaction with cold solution of NaOH does not produce precipitate, at elevated temperature it decomposes to almost white product unlike Cr3+) - I forgot to mention that in my initial post



100 ml of water was added to 1,3,5-trinitrobenzoic acid and heated to 35 C in warm water bath (do not overheat as then the acid decarboxylates, do not let too cold as then the acid solubility is low), always keep inserted thermometer

IMG_20220126_081852_4.jpg - 20kB



the neutralization, 1 last drop of 15% NaOH is capable to change color to red, note do not add more than that last drop!!!, stir the mixture well during whole time and cool down if necessary so the neutralization is performed at 35 C (low temp = low solubility of acid, high temp = decarboxylation of the acid)

IMG_20220126_083322_2.jpg - 34kB



then quickly add 2 drops of acetic acid at which the color reverts back to yellow, leaving the previous red color for long time allegedly has bad impact on the quality of the product

IMG_20220126_083344_0.jpg - 35kB IMG_20220126_083901_5.jpg - 28kB



gravity filtered, only little captured on filter paper

IMG_20220126_091137_0.jpg - 44kB

Fery - 2-2-2022 at 08:05

decarboxylation under very weak reflux, then full reflux for 3 hours

IMG_20220126_095149_2.jpg - 34kB



after cooling down the 1,3,5 trinitrobenzene solidified and was filtered out, then air dried to the next day

IMG_20220126_172135_1.jpg - 23kB IMG_20220127_042257_8.jpg - 35kB



crystallization from acetic acid

IMG_20220127_183619_2.jpg - 27kB IMG_20220130_135546_0.jpg - 47kB



some impure product recovered from acetic acid mother liquor

IMG_20220130_140118_4.jpg - 48kB



crystallization of the main crop from methanol

IMG_20220130_220217_7.jpg - 33kB IMG_20220131_104711_3.jpg - 42kB IMG_20220131_104816_5.jpg - 20kB



impure product recovered from acetic acid mother liquor dissolved in methanol mother liquor, evaporated to half of its volume, crystallized on cooling, filtered, then crystallization from 20 g of methanol, here the final crystallization from 20 g of methanol

IMG_20220201_103912_9.jpg - 24kB



comparison of main crop (pale yellow, left beaker on 1st and 2nd picture where photo taken against sunlight, right beaker on 3rd picture where photo taken with sunlight from the back) and second crop (a little darker, but still good m.p.)

IMG_20220202_122340_8.jpg - 33kB IMG_20220202_122347_3.jpg - 40kB IMG_20220202_122405_3.jpg - 33kB

Texium - 2-2-2022 at 12:22

Very nice work! Be careful with that product though, as 1,3,5-trinitrobenzene is more explosive than TNT and can detonate from shock in solid crystalline form like you have. Take care not to grind it or handle it roughly.

Fery - 2-2-2022 at 22:16

Hi Texium, thank you, I did not know that it is so much sensitive. I always thought that it requires primary initiator like Pb(N3)2 or Hg fulmitate to detonate. Btw for crushing TNT I used common pliers from hand tools shop, as the TNT was in bigger pieces of few grams. But apparently I did not crush TNT thoroughly enough. Very likely I had to sieve the crushed TNT through fine sieve (at most 1 mm) and crush again all pieces bigger than 1 mm.
Do you know any interesting usage of trinitrobenzene? I mean in chemistry (organic, analytic etc), not as an explosive. So far I saw only its deprotonated form (that picture with red color during neutralization).

Texium - 3-2-2022 at 12:56

I don’t know. Besides reducing it to the corresponding triamino compound, which would give you some handles to work with, I don’t see it having many synthetic uses. It’s so strongly deactivated that it would likely be completely inert towards electrophilic substitution. The fact that it is easily deprotonated is interesting though. You may be able to use that to your advantage by trapping the anion somehow. I’ll do a SciFinder search later to see if anyone has done something like that before.

Boffis - 7-2-2022 at 22:49

Well done Fery; another nice practical write-up of an often discussed but rarely practiced reaction. I was actually planning to try this myself and your details with help me greatly. My target is not TNB but 4,6-dinitroanthranilic acid. My plan is to oxide the starting compound to the acid as you have done and then try o-selective reduction of the sodium salt with sodium sulphide and sodium bicarbonate (as a buffer) in a similar fashion to the preparation of picramic acid. I have not seen a reference to this use of sulphide reduction so the procedure is speculative.

Nice work, keep it up.

Fery - 8-2-2022 at 00:32

Hi Boffis, that could work. Just keep the 2,4,6 trinitrobenzoic acid always below 55 C otherwise it decarboxylates to 1,3,5-trinitrobenzene (you can easily filtrate the insoluble trinitobenzene off, but that reduces yield).
Also near the end of the addition of dichromate the mixture becomes quite thick.
Here they used hydrazine + cat. FeCl3 for the reaction -NO2 -> -NH2
https://sci-hub.ee/10.1134/s1070428002120254

Attachment: s1070428002120254.pdf (482kB)
This file has been downloaded 207 times

Boffis - 8-2-2022 at 11:01

Whoo Fery, that's a useful find! Save me a lot of experimentation. Many thanks.

Keras - 29-10-2024 at 06:37

Is there another way to synthesise TNB rather than going the round way through TNT? I would like to synthesise phloroglucinol, TNB + sodium hydrosulphite → 1,3,5 triaminobenzene + NaOH → phloroglucinol.

Boffis - 30-10-2024 at 02:02

Apparently TNB can be prepared by the spontaneous trimerisation and decarbonylation of nitromalondialdehyde salts when acidified.

Nitromalondialdehyde is prepared via the scheme:

A pentose sugar --> furfuraldehyde --> mucrobromic acid --> nitromalondialdehyde alkali salt. The first step has been well covered on SM before, the 2nd and 3rd are on the Org Synth Vol4 p688 and Vol4 p844. The final stage is discussed by Hill et al but I'll need a bit of time to dig out the reference.

There is of coarse the direct nitration of m dinitrobenzene but the yield is poor due to oxidation and the formation of large amounts of tetranitromethane by-product. The route via TNT looks like the best route though from a practical perspective,

Keras - 30-10-2024 at 05:20

Quote: Originally posted by Boffis  
Apparently TNB can be prepared by the spontaneous trimerisation and decarbonylation of nitromalondialdehyde salts when acidified.

Nitromalondialdehyde is prepared via the scheme:

A pentose sugar --> furfuraldehyde --> mucrobromic acid --> nitromalondialdehyde alkali salt. The first step has been well covered on SM before, the 2nd and 3rd are on the Org Synth Vol4 p688 and Vol4 p844. The final stage is discussed by Hill et al but I'll need a bit of time to dig out the reference.


Interesting!

Quote: Originally posted by Boffis  

There is of coarse the direct nitration of m dinitrobenzene but the yield is poor due to oxidation and the formation of large amounts of tetranitromethane by-product. The route via TNT looks like the best route though from a practical perspective,


Fair. Any other oxidant besides potassium dichromate? Potassium permanganate? Oxone? Ammonium persulphate?

Ormarion - 10-11-2024 at 05:25

Boffis, you can also make 3,5 dinitro anisole from trinitrobenzene using NaOMe in methanol, there is a procedure for it on orgsyn i think