Hexogen with Acetic Anhydride - Question

[simply RED]

The last months I really started to hate 100% nitric acid. It evaporates too much and it needs purging!
So, I found some acetic anhydride.

I am going to try to prepare RDX with acetic anhydride, paraformaldehyde and ammonium nitrate - Without nitric acid.

My question is: what is the best method for this (giving the best yield, of course) ?
If somebody has tried the process, it would be great to share some experience.

[quicksilver]

Do a quick patent search for: W.E. Bachmann, John C. Sheehan. These gentelmen have done the most work with the above process. There are 5 patents all together that enclude [two of which] just what you are looking for. Their work has been published in a number of journal articles as well.
Bachmann himself made a career working with RDX and is considered by some to be the "grandfather" of the various processes. They also worked with the unconventional techniques [none of which were adapted by industry] for increasing yield.

[simply RED]

I do have the exact process described several times, but as you may know, there maybe tricks doing the real thing. When RDX is made from hexamine dinitrate and HNO3 there are several tricks (purity of the acid, temperatures, speed of addition and so on).

I do not want to waste AcANH learning the tricks on my own, but rather learn from the experience of somebody that has made it before.

[quicksilver]

I can REALLY apprieciate that. I found something that may just be what you are looking for. If the attachment works; check out the PDF file enclosed. It deals with temp issue/yield etc.
I have found that some folks do great lab work but keep lousy notes, others post information that is of little use because they really don't pay as close attention to the little things or have such poor equipment that their temp readings suck, measurments are sloppy, etc.
One of the best threads here on another subject is frought with these issues and the subject matter is so interesting that it's a damn shame....

I hope this helps.....

[chemi]

The last months I really started to hate 100% nitric acid. It evaporates too much and it needs purging!
So, I found some acetic anhydride.

I am going to try to prepare RDX with acetic anhydride, paraformaldehyde and ammonium nitrate - Without nitric acid.

My question is: what is the best method for this (giving the best yield, of course) ?
If somebody has tried the process, it would be great to share some experience.

dear my friend ,
you must use fuming nitric acid (98%-100%) for prepration RDX from hexamine by disolving ammonium nitrate in nitric acid to make what is called ( dinitric acid) to make nitration to hexamine.
the only role of acetic anhydride is to absorb the formed water from the reaction instead of usining sulforic acid to avoide the formation of heat .
but paraformaldehyde is used in this reaction when you need to prepare HMX where paraformaldehyde is inrtoduced to form the eight membered ring of HMX and you do not need paraformaldehyde in prepration of RDX with six membered ring. this method is called Bechmann process
so how to you to make RDX without Hexamine ??
and how to you make bond cleavage to Hexamine molecule and to inroduce the nitro groups to form RDX without using the fuming nitric acid ?

regards ,
chemi

[nbk2000]

Newbies (chemi) need to realize that, unless they've already had 10 approved posts, that any posts they make have to be reviewed and approved by staff before it will be visible, even to the person who posted it.

So DO NOT repost the same thing a half-dozen times! Wait a few days, even a week if need be, keeping a copy of your posts saved on your computer, and wait for it to be approved (or not).

[simply RED]

Dear mods, may you approve the attachment by quicksilver.
Thanks

[nbk2000]

The G(m)ods have heard your prayers and BEHOLD!, the attachment was found worthy and approved. :)

[simply RED]

Quite interesting!
citing from memmory:
---Another option is to add NH4NO3 to conc.HNO3. Gasing occures and all(orange) nitric oxides are purged out. ---

Really?

[Maniak]

Yes, that's true. If you add AN into acid, fuming will proceed and the acid will become clear in 2-3hours. The problem is, wheather it can't add some water and dilute the acid by way like:
NH4NO3 + HNO2 < = > NH4NO2 + HNO3
NH4NO2 = > N2 + H2O
I don't believe that this reactions take place there, but haven't got better idea :o

Getting RDX - ross's process (AN + AA + paraformaldehyde) is not useful in my experience. The adition of BF3 which is used in original patent is probably quite important for RDX yield and mainly for its purity. Without BF3 you'll get poor yield with high content of acetyl-derivates (because of higher temperature used).
The method I prefere is so called bachmann-ross process (AN + AA + fuming NA + hexamine in acetic acid) which gives me the crop 70-74% of theory. It's safe and purity of the product is sufficient - it was proven by H-NMR to have 96%RDX + 4%HMX + traces of acetyl-derivates. Although there's need for both AA and fuming NA, the process can be economical when amounts of reactants are balanced.

[me234]

Maniak, could you tell me how you used H-NMR to distinguish between RDX and HMX? As far as I can tell, all protons in both molecules will be equivalent to each other within the molecule, and also equivalent to the protons in the other molecule.
Examination of the structures shows that comparing any proton containing group (i.e. the methylene groups) in either molecule, gives each methylene group flanked by identical nitramine groups, and no non-equivalent groups are present in either molecule which might couple with the methylene protons.
Unless of course something untoward happens, like the nitro groups being in some form of staggered conformation (but these aren't really close enough to couple).

Then again, I don't think I've ever seen what RDX and HMX look like in 3-D; maybe there is something that could create non-equivalence there.
Anyway, if you could post how you differentiated between the two I'd be grateful.

[Maniak]

The tests were done in university lab (I haven't got my own NMR yet:D ) and RDX wasn't mine, but just prepared according to my version of bachmann-ross, so I can't describe it exactly (also because of my lack of knowledge:o ).
But I think there are important differences between RDX and HMX in 3-D. See Urbanski IV (pgs 385 and 372) - HMX may exist in more (4?) steric modifications, but RDX has just one. The most stable beta-HMX conformation, which is created by bachmann-ross process, is quite different from the only one existing conformation of RDX.
Anyway, my friend at NMR (wocas@bubakov.net - many thanks for his help:) ) used exact values for comparison from:
A.R.FARMINER, G.A.WEBB - NMR Studies On Some N-Nitramines and N-Nitrosamines, 1975

[simply RED]

Yes, that's true. If you add AN into acid, fuming will proceed and the acid will become clear in 2-3hours. The problem is, wheather it can't add some water and dilute the acid by way like:
NH4NO3 + HNO2 < = > NH4NO2 + HNO3
NH4NO2 = > N2 + H2O
I don't believe that this reactions take place there, but haven't got better idea :o



I think the yield of this reaction is extremely low.
Most possibly when NH4NO3 is dissolved in HNO3 the NOx are less solvated and less soluble, so they evaporate.

[Chris The Great]

With H-NMR, RDX will give a peak of 6 while HMX will give a peak of 8 for the CH2 groups. I'm not sure if you could tell the difference in a sample though if there is only a few percent of HMX in there.

The four different "forms" of HMX also have very different molecular forms so I would think they would be different than for RDX on NMR. However Urbanski doesn't have a structure of RDX, only a charge distribution diagram which doesn't show the 3D structure.

[sprocket]

You'd think the chemical shifts would be similiar, but no... Anyway, this paper seems to be relevant to the subject, but unfortunately my university doesn't subscribe to the journal:

A. G. Landers, T. M. Apple, Cecil Dybowski, T. B. Brill, <SUP><FONT SIZE='-1'>1</FONT></SUP>H nuclear magnetic resonance of α-hexahydro-1,3,5-trinitro-<I>s</I>-triazine (RDX) and the α-, β-, γ- and δ-polymorphs of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), Magnetic Resonance in Chemistry, 1985, 23(3), 158.

Does anyone have access to this journal?

[Maniak]

I attach NMR spectra of RDX by KA method (german origin, but also described in bachman-ross's patent) - there is more visible HMX peak because of its higher content in contrast to other method we tried. Other acquired spectra differs in height of HMX peak.
The values of -CH2- in RDX and HMX are 6,09 and 6,04 ppm according to my above cited literature. The literature doesn't say which form of HMX it is, so I'll try to get Landers' article and find solution.

[megalomania]

The evaluation of structural isomers of molecules is difficult to do without having some experience with an NMR, but the NMR is just the tool to do it best without some X-ray crystal spectroscopy. I am no expert on the NMR, but I do get to use one from time to time. One method to evaluate structural differences by calculating the differences in J values between the spectra. Different structures will have different neighbors influencing them, and therefore no two spectra will be identical with two different molecular structures. The chemical shifts will also be slightly different.

The difference in J values and chemical shifts will be subtle, but that’s why there are software tools to help you elucidate the data. The higher the resolving power of your NMR the easier it will be to detect differences.

Don’t just limit yourself to 1H NMR, you can also try 13C and 14N. Other spectroscopic techniques may be in order though. Some reading is always advised. See “Determination of the HMX and RDX content in synthesized energetic material by HPLC, FT-MIR, and FT-NIR spectroscopies” at www.scielo.br/scielo.php?script=sci_arttext&pid=S0100-40422004000400005 at Quimica Nova. See also Methods for the Determination of Absolute Configurations by NMR at qoweb.usc.es/RIGUERS/nmr.HTML which has some references to read up on the techniques.

[darkfury]

I have tried the synthesis of RDX by the E-method (p-formaldehyde, acetic anhydride and AN).
USE
My experience was that the AN wouldn't dissolve in the acid. I used megalomania’s synthesis (scaled down): 21 g AN in 50 mL of acid. After a lot of stirring and heating to 90°C it still wouldn’t dissolve.
PARAGRAPH
The AN was dried in an oven (120°C, 1 hour), acetic anhydride was 98% reagent grade. I have checked Urbanski’s description, and he makes no mention of the AN being difficult to dissolve …
BREAKS

[Diabolique]

Quicksilver's post (thanks for the pdf, I've been looking for that article) is the Bachmann method. Urbanski gives the following on it:

117 parts of 98% nitric acid is mixed with 508 parts acetic anhydride at 5* to 15* C. This is heated to 70-75* C. 114 parts of ammonium nitrate is mixed with 192 parts of hexamine dinitrate (hexamethylenetetramine dinitrate). This is added to the nitric acid - acetic anhydride in small portions with constant mixing, keeping the temperature constant at 73*-78* C - cooling needed. After all ingredients are combined, keep at this temperature for 15 more minutes, then cooled to 60* C. Filter out the product and wash with acetic acid, then wash with water. Yield is said to be 191-202 parts.

Hexamine dinitrate is prepared by adding 65 parts of 70% nitric acid to a solution of 40 parts hexamine in 70 parts water at 15* C. Cool to 5*C and filter the product. Yield stated to be ~95%.

High temperatures favor products other than RDX, so the reaction should be strongly cooled to allow addition of the components more quickly.

The reaction using Hexamine dinitrate, ammonium nitrate, nitric acid and acetic anhydride may be more efficient and easier/safer - no hot acids.

High grade acetic anhydride is hard to obtain, it is a controled (tracked) substance as it is used to convert morphine into heroin. FGAN is the same due to its use in ANFO.