Sciencemadness Discussion Board

Trinitrobenzene without nitric acid or benzene

Chisholm - 5-4-2017 at 09:10

The sugar-substitute inositol (cyclohexanehexol) can be dehydrated via concentrated sulfuric acid into phloroglucinol (1,3,5-benzenetriol), along with some amounts of the isomers hydroxyquinol (1,2,4-benzenetriol) and pyrogallol (1,2,3-benzenetriol).

Phloroglucinol (via its keto form) reacts with hydroxylamine under basic conditions to give the tris(oxime).

Could oxidizing agents such as manganese dioxide, potassium permanganate, sodium hypochlorite, potassium chlorate, or hydrogen peroxide convert the oxime to the nitro compound?

Praxichys - 5-4-2017 at 09:49

It certainly sounds possible but I think yields will be impractically low.

Assuming you can get anything but tar from the inositol dehydration, separation of the isomers will be difficult.

Further, the only references I can find about oxidizing oximes to nitro involve expensive and exotic things like peroxytrifluoroacetic acid and peroxyacetic acid, and even then the yields are only 20-40%. One method uses urea peroxide and trifluoroacetic anhydride. (!) At this level of danger and expense, you might as well have just nitrated benzene with WFNA and oleum and gotten 80 or 90% yield in one step.

PirateDocBrown - 5-4-2017 at 14:02

I've seen literature where that oxidation is done with sodium perborate.

Praxichys - 6-4-2017 at 06:25

Yes, but the perborate oxidation uses glacial acetic acid as a solvent, effectively making peroxyacetic acid the oxidant. The yields are still terrible, and I'm not sure it's wise to be near a peroxyacetic acid bath full of trinitrobenzene and the oxime at 50-60°C.

DOI: 10.1055/s-1992-22006

PHILOU Zrealone - 6-4-2017 at 14:53

The best alternative route to TNB (trinitrobenzene) instead of nitration I know of is from trimerization of nitro-malonaldehyde (2-nitropropan-1,3-dial).

The reaction is complex but seems to proceed via condensation of the aldehyd moeity and the hydrogen of the carbon holding the NO2 and strongly activated by 3 electron-withdrawing groups (2 aldehyd and 1 nitro).

Thus multiple Michael additions, dehydration, aromatisation and autoxidation (via air) of aldehyd to carboxylic and decarboxylation.

I have no clue about the overal yield because if TNB is a product of the reaction a linear polymer may also be present.

Another plausible way would be trimerization of nitroethanal (O2N-CH2-CH=O)...
O2N-CH2-CH=O + CH2(-NO2)-CH=O --> O2N-CH2-CHOH-CH(-NO2)-CH=O --> O2N-CH2-CH=C(-NO2)-CH=O + H2O
and so on until TNB or (-CH=C(-NO2)-)n (polynitroethyne).

Finally purely theorically...phloroglucidol tris-oxime (cyclohexa-1,3,5-trione tri-oxime) should display the same feature as usual oxime to be oxy-nitratable by concentrated HNO3 or N2O4 to yield geminal dinitromethylenes (amongst other oxydation products)
Ar-CH=N-OH --HNO3 or N2O4--> Ar-CH(NO2)2
So from our tri-oxime one may expect some 1,1,3,3,,5,5-hexanitrocyclohexane...the later may eventually lose some of its NO2 as HNO2 (nitrite anion and then H(+)) and in fine yield some TNB.
But this will be far less efficient than benzene + N2O5 or (conc HNO3/conc H2SO4 (+SO3))

[Edited on 6-4-2017 by PHILOU Zrealone]

Boffis - 7-4-2017 at 02:22

@ Chrisholm, do you have a reference for the dehydration of inositol, I would be interested in reading it.

I came across the formation of trinitrobenzene from nitromalonaldehyde during my research into fulminic and fulminuric acids. The two papers of interest are in the American Chemistry Journal by Hill & Halle and Hill & Torrey (refs below), The whole of this journal is available free on line. Nitromalonaldehyde is a very transient species with a fascinating chemistry, it not only condenses with itself to form TNB but if formed in the presence of some ketones and related compounds it give derivatives of 4-nitrophenol, so ethyl methyl ketone give 4-nitro-o-cresol and acetone dicarboxylic acid give 2-hydroxy-5-nitrobenzene-1,3-dicarboxylic acid.

Nitromalonaldehyde is derived from mucobromic or mucochloric acids (4) which are themselves produced by exhaustive halogenation of furfuraldehyde (3). The preparation of the halomucic acids is described and is available from Org Synth. I have run the mucobromic acid synthesis several time, works OK but involve handling a lot of bromine and requires fairly involved equipment. As a preparative route to TNB however, its a non-starter, the yield from furfuraldhyde to TNB is probably less than 10%.

Bed-time reading suggestions :)
1) H. B. Hill and J. Torrey, Amer. Chem. J. 22, p89 (1899).
2) H. B. Hill and W. Hall Amer. Chem. J. 29, p253 (1903)
3) Mucobromic acid from furfural Org Synth Col Vol 4 p688
4) Sodium nitromalonaldehyde hydrate from mucobromic acid Org Synth Col Vol 4 p844

Chisholm - 7-4-2017 at 07:17

Quote: Originally posted by Boffis  
@ Chrisholm, do you have a reference for the dehydration of inositol, I would be interested in reading it.


I don't; however, apart from the problem of finding the right concentration of SA to avoid forming tar, there's no reason the dehydration wouldn't work. There would be plenty of side products, though.

Boffis - 8-4-2017 at 09:08

@Chisholm, I may try this reaction when I get chance but I can see some problems.

Phloroglucinol is rather sensitive to oxidation so it may be simply oxidized by the conc sulphuric acid being used as dehydrating agent. I may also react with excess acid to form a sulphonic acid or poly-sulphonic acid and then there is the issue of which isomers are produced as you alluded to above.

PHILOU Zrealone - 8-4-2017 at 16:11

Phenols are ketons in disguise under their enol form (hence the name Ph-enol).
So phloroglucidol may be seen as cyclohexa-1,3,5-trione.

If a keton is put into a dehydrating agent if often is subjected to (poly)condensation-crotonisation...a typical example is aceton that turns into mesityl oxide, phorone, isophorone and mesitylene (1,3,5-trimethylbenzene) and polymeric tarry materials.

Cyclohexanone and cyclopentanone also form after tricondensation-dehydration a central aromatic ring with 3 cycloalcanic bridges between positions 1-2 ;3-4 ; 5-6.

If we imagine the same with cyclohexa-1,3,5-trione then we should get a graphite like end product with tarry polymers...

Phloroglucidol (1,3,5-trihydroxybenzene) is sensitive to oxydizing réactants...but HNO2 allows easy conversion to trinitrosophloroglucidol and dilluted HNO3 (because less oxydizing than conc HNO3) turns it to trinitrophloroglucidol.

Boffis - 9-4-2017 at 07:24

@Philou Zrealone, good thinking Batman!

PHILOU Zrealone - 14-4-2017 at 04:57

Quote: Originally posted by Boffis  
@Philou Zrealone, good thinking Batman!

:)

@Boffis,

What part of the thinking?

And why Batman?

Boffis - 17-4-2017 at 03:39

Don't you remember the old Batman and Robin cartoons strips? Robin was always saying "good thinking Batman" :)

PHILOU Zrealone - 17-4-2017 at 15:45

Quote: Originally posted by Boffis  
Don't you remember the old Batman and Robin cartoons strips? Robin was always saying "good thinking Batman" :)

Now that you mention it...yes I do remember those old cartoons...:);):P:cool:

clearly_not_atara - 17-4-2017 at 16:59

Quote:
the only references I can find about oxidizing oximes to nitro involve expensive and exotic things like peroxytrifluoroacetic acid and peroxyacetic acid, and even then the yields are only 20-40%.


There is a paper by Corey in which this is done with hypochlorite under PTC:

http://www.sciencedirect.com/science/article/pii/S0040403901...

The oxidant is reported as tetra-n-butylammonium hypochlorite although presumably other sources of ClO- which are soluble in inorganic compounds may be equally applicable.

I am not sure however if the reaction will be applicable to phloroglucinol. Notably the product is the alpha-chloronitro compound although in this case the compound 1,3,5-trichloro-1,3,5-trinitrocyclohexane will rapidly eliminate three equivalents of HCl to give trinitrobenzene, if it does not decompose spontaneously, that is.