Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
3,3’-Diamino-5,5’-bis(1H-1,2,4-triazole) and Derivatives
3,3’-Diamino-5,5’-bis(1H-1,2,4-triazole) is the product of aminoguanidine condensing with oxalic acid followed by cyclisation. I've ran a few
experiments; I'll detail them here and attach the related published accounts.
It's of interest as some of its derivatives have high calculated performance, approximating that of HMX and being an aromatic amine, it lends itself
to diazotisation and subsequent reactions (C-nitro via NaNO2 or C-azide via NaN3) or to direct nitration with mixed acids to yield the nitrimine.
These are acidic and many salts can be derived from them, most of which have low solubility so are just precipitated from its sodium salt.
I've run the aminoguanidine - oxalic acid condensation, it's quite trivial with the poorly soluble products precipitating from aqueous solution. I've
also done a direct nitration of the diamine to dinitrimine, and formed its potassium salt.
3,3’-Dinitrimino-5,5’-bis(1H-1,2,4-triazole) is a pure white fine powder, it can be recrystalised from boiling water to small needles but its
solubility even in boiling water is very low. It forms as a dihydrate, which is lost at about 160C. Its energetic properties are odd, if I had to
compare it to anything it would be TMDD (yes, the puffy smokey peroxide). It's quite sensitive under the hammer but is quite feeble puffing off with a
mild pop. On ignition it gives a violent puff, ejecting yellow fluff residue into the air.
The potassium salt is less energetic than the free acid, smouldering through with a large lilac flame characteristic of potassium, it has no primary
explosive qualities.
The oxidation to the highest performing diol designated "MAD-X1" is in the attached patent. It's essentially TKX-50 with two of the N's replaced with
C-NO2.
A couple speculatory reactions are in the scheme attached. it may be possible to chlorinate the sodium salt. Chloramines of this nature are generally
much more sensitive than the parent compound but less stable. It might also be possible to drop a diiodo compound from its diazonium salt and oxidise
to iodyl with oxone (a pool chem). This would most likely have primary and initiatory properties if possible.
Experimental
3,3’-Diamino-5,5’-bis(1H-1,2,4-triazole)
A mixture of aminoguanidine bicarbonate (54.0 g, 0.4 mol) and oxalic acid dihydrate (25.2 g, 0.2 mol) was slowly treated with 70 mL of 32%
hydrochloric acid. After carbon dioxide evolution ceased, the mixture was heated to 75°C using a hotplate set at 125°C with stirring at 500 rpm. The
solution gradually transitioned from orange to colorless before solidifying into a thick white paste after approximately 20 minutes.
The paste was broken up with 100 mL of water and vacuum filtered, though filtration proved challenging due to the fine particle size clogging the
filter. The still damp and acidic filter cake was subsequently suspended in 260 mL of water and solubilized by adding 18.0 g of sodium hydroxide to
reach pH 14, resulting in a clear brownish-gray solution.
The solution (400 mL) was heated to boiling with continuous stirring to prevent bumping, and concentrated to approximately 300 mL, at which point a
white precipitate had formed. Acidification to pH 4 was achieved by adding 80 mL of 20% acetic acid resulting in a further thickening of the mixture.
The resulting suspension was vacuum filtered and washed through on the filter with an additional 250 mL of water, yielding slightly buff-colored
3,3'-diamino-5,5'-bis(1H-1,2,4-triazole). Yield 22g.
3,3’-Dinitrimino-5,5’-bis(1H-1,2,4-triazole) (DNABT)
10g of 3,3'-diamino-5,5'-bis(1H-1,2,4-triazole) was dissolved into 100mL of 98% sulphuric acid and cooled to 0C. 100mL of 68% nitric acid was now
added dropwise maintaining a temperature between 0-15C. The mixture was stirred in its cooling ice bath for 30 minutes then removed from the bath and
allowed to warm to room temperature over the next hour under constant stirring. Dumping the now clear-yellow solution into ice water precipitates a
fine white precipitate that was filtered and dried. Yield was about 10g, it wasn't weighed as I kept picking at and playing with it.
[Edited on 3-5-2025 by Axt]
Attachment: bis(aminotriazole) and derivatives.pdf (896kB)
This file has been downloaded 219 times
Attachment: 3,3'-DINITRO-5.5'-BIS-TRIAZOLE-1,1'-DIOL - US9481656.pdf (544kB)
This file has been downloaded 185 times
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
That looks very interesting. I'm especially looking forward to examining the MAD-X1. Performance levels between HMX and CL-20, but sensitivity less
than TNT sounds very enticing. The question is how it responds to initiation in small diameters, and or course if it is hygroscopic...
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
Interestingly if malonic acid is used in place of oxalic acid the final dinitrimine product is (unintuitively) even more sensitive, only 1J impact
sensitivity.
I've been looking for reference to other carboxylic acid products, namely tartaric acid. While acidic nitrate esters aren't generally a target
molecule for practical purposes it would still make for an interesting experiment. I couldn't find anything specific to tartaric acid but glycolic
acid is shown here to have the expected analogous structure https://ijisrt.com/assets/upload/files/IJISRT23DEC1125.pdf .
Proposed tartaric acid derivative is attached.
[Edited on 5-5-2025 by Axt]
Attachment: Malonic acid + aminoguanidine to a triazole.pdf (329kB)
This file has been downloaded 175 times
Attachment: DNBTM primary (malonic acid + aminoguanidine).pdf (571kB)
This file has been downloaded 158 times
|
|
|
Elemental Phosphorus
Hazard to Others
Posts: 193
Registered: 11-11-2016
Location: Is everything
Member Is Offline
Mood: No Mood
|
|
I'm also curious about the carbohydrazide salt. 9.3+ km/s is crazy high. But if the malonic acid product is more sensitive, I'd expect a dinitrate
ester to be an order of magnitude more sensitive.
As for the greater sensitivity of the malonic acid product compared to the oxalic acid one, I wonder if it has to do with decreased rigidity of the
molecule. I can't pretend to know much about this, but in my experience with non-energetics rigidity often trumps enthalpy in terms of stability. And
according to the XRD structures it seems like the methylene-bridged malonic acid derivative is rotated and the oxalic derivative is flat.
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
The properties of the carbohydrazide salt were taken from the attachment.
Wang, R., Xu, H., Guo, Y., Sa, R., & Shreeve, J. M. (2010). Bis[3-(5-nitroimino-1,2,4-triazolate)]-Based Energetic Salts: Synthesis and Promising
Properties of a New Family of High-Density Insensitive Materials. Journal of the American Chemical Society, 132(34), 11904–11905.
doi:10.1021/ja1055033
Its melting point is the lowest of them all and highest (calc.) performance of them all. You could probably attempt a cast of the carbohydrazide at
least on a small scale. mp. 162°C, decomp. 222°C.
The ammonium salt isn't far behind on performance, but it is a "corrected" value since it forms as a hydrate.
Attachment: Salts of Bis[3-(5-nitroimino-1,2,4-triazolate)].pdf (121kB)
This file has been downloaded 185 times
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
OK, so I synthesized diaminobistriazole (DABT) and then went for the dinitro derivative (DNBT) via Sandmeyer reaction as described in the Klapötke
paper and patent.
As expected, the reaction evolves a lot of gas, both N2 and NOx. Coupled with the low solubility and very small particle size of the DABT, a
persistent foam is formed which threatens to overflow the beaker. If you decide to do this experiment, I would suggest using a container with
considerable excess volume. I will continue with workup and oxone oxidation (though the "stir at 40C for 60 hours" bit is rather annoying).
On a related note, can any of you offer a reason for why the very large excess of nitrite (they use 20 equivalents) is destroyed with sulfuric acid
before the DNBT is filtered off? It seems very wasteful and also generates large amounts of gas and therefore foam and so requires a lot of time.
|
|
|
dettoo456
Hazard to Others
Posts: 289
Registered: 12-9-2021
Member Is Offline
|
|
I don’t know if it’d achieve the same result, but maybe try substituting the H2SO4 for a weaker acid like Citric or AcOH. And scaling the
equivalents down to at the least 2-5mol probably shouldn’t hurt either. If using this approach, longer rxn time, low temp, and slow additions would
be necessary.
In the preparation for TNX (1,3,5-trinitroso-1,3,5-triazine), Citric acid is actually preferred due to the buffering/limiting effect it provides,
compared to the H2SO4 that immediately consumes all NaNO2. The reactions are of course different, and it might in any case not work for you, but
it’s the only thing I can think of that might help.
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
Yes, tinkering with the reaction is desirable to make a viable "energetic". If the large excess of nitrite is necessary for some reason, it might be
possible to just reuse the mother liquor for the next batch and simply add enough fresh nitrite to make up for what was actually consumed. For now,
I'm just trying to replicate the procedure from the paper though.
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
How did it go?
Is the 20% H2SO4 / DABT "solution" described in the patent actually a suspension?
I'm guessing DNBT is a strong acid relative to nitrous acid, thus without the acidification with H2SO4 it would be present as it's sodium salt.
Presumably using more available HCl in place of H2SO4 wouldn't pose a problem. Or maybe (as suggested by your description) the sodium salt is
filterable, thus preserve the nitrite/sulphate filtrate and then acidification during recrystallisation to free the DNBT.
[Edited on 14-5-2025 by Axt]
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
Yes, DABT is very slightly soluble in 20% H2SO4, so it is a suspension rather than a solution. I ultrasonicated it to create a quite stable
suspension, though I'm not sure how much difference it made. The tough foam is quite annoying to deal with, but it may be more manageable with
mechanical stirring instead of magnetic.
Anyway, after acidification with H2SO4, the mix was cooled to 5C and filtered. The recovered solid was dissolved in boiling water (about 25 ml per
gram of utilized DABT was required). I didn't notice any insolubles so I didn't filter it, but simply cooled it very slowly to 5C. This gave needle
shaped crystals which I would describe as pale yellow rather than pale green. These crystals were filtered off and dried (1.89 g from 1.50 g DABT
which is about what Klapötke got, though some of it was dried at 110C since I needed to get the next step of the synthesis going. This may have
eliminated some water of crystallization).
0.50 g of the supposed DNBT was suspended in a solution of 7.5 g potassium acetate diluted to 30 ml, and heated to 40C. Then Oxone (11.7 g, finely
ground) was added in one portion and the mix was stirred at 40C for 65 hours. Then 15 ml of 50% H2SO4 was added in one portion. This caused the colour
to go from orange to pale yellow. The mix was stirred at room temp for 5 minutes, and then extracted with 4x10 ml ethyl acetate. The combined extracts
were filtered and the pale yellow solution is drying over MgSO4 at time of writing.
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
I ran a test, without acidifying the diazotisation liquor and I will say the result was a little disturbing.
As per your description Microtek it foamed up quite a lot in the final stages of the DABT/H2SO4 addition to the nitrite. And was thick with a
buff-coloured precipitate. I filtered this directly without acidification, washed through with water then boiled the wet cake.
The recrystallisation liquor turned orange and was filtered, which only captured a very small amount of solids even though the liquid stayed turbid.
On cooling a crop of very fine salmon coloured powder formed that is difficult to filter, rapidly clogging it.
This salmon powder is a primary explosive as vehemic as any I have ever seen. I don't know what it is, but it is very sensitive to the point where I
wouldn't have expected it to be the sodium salt of DNBT, not that its properties are given in the literature, so I really can't discount it for sure.
One guess would be a zwitterionic diazo compound, where you have half substituted nitro and the other side a balancing diazonium group similar to
DDNP, structure attached. I cannot stress enough how dangerous this stuff is, some impregnated into absorbent paper and dried detonated when the paper
was torn.
On a separate note, chlorination of the sodium salt of the dinitrimine does not form the expected chloramine, it seems to break the ring and gas off.
The highly staining yellow product that forms is non-energetic.
[Edited on 15-5-2025 by Axt]
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
That is certainly good to know. It begs the question whether this kind of thing could be harnessed - a lot of amino substituted triazoles are easily
accessible. Maybe some of them have acceptable properties as primaries when transformed to the diazonium salt.
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
I did a small experiment: Since I figured diaminobistriazole (DABT) must be zwitterionic, and recalling the performance of 4-amino-1,2,4-triazolium
perchlorate, I reacted a small amount of DABT with two equivalents of 50% HClO4. A small, but noticable exotherm was observed, but the DABT does not
dissolve as such. On oven drying the mud-like substance at 110C, an off white powder was obtained. This powder burns energetically and detonates by
hammer blow on anvil. It doesn't seem excessively sensitive, but these are only preliminary tests.
I have weighed the remaining product and placed it in a room with moderate humidity to see how hygroscopic it is.
Edit:
Over the course of 24 hours it had gained 15% mass, and exhibited a tendency to clump. It wasn't deliquiscent, and this level of hygroscopicity may
possibly be manageable, but time will tell. I have left it for another 24 hours to see if the 15% is the plateau, or if it will absorb more.
Additionally, I nitrated DABT to experiment with the nitrimino derivative. I used WFNA in order to follow the paper as closely as possible. I noticed,
a fairly strong exotherm at first while adding the nitric, but after about 1 ml was added (I was nitrating 1.00 g, so the procedure called for 3 ml
HNO3), no further exotherm could be detected.
Otherwise, the reaction went as Axt describes it. I haven't isolated the product yet, so yield remains to be seen.
[Edited on 16-5-2025 by Microtek]
[Edited on 16-5-2025 by Microtek]
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
Quote: Originally posted by dettoo456  |
In the preparation for TNX (1,3,5-trinitroso-1,3,5-triazine), Citric acid is actually preferred due to the buffering/limiting effect it provides.
|
Where is this derived from? I've never seen this before.
| Quote: Originally posted by Microtek | | Additionally, I nitrated DABT to experiment with the nitrimino derivative. I used WFNA in order to follow the paper as closely as possible.
|
There is actually 3 published procedures that I found, the other two used 70% and 68% nitric acid and actually notes higher yields (the nitration
seems as trivial as that of nitroguanidine itself). Although when adding 68% HNO3 to H2SO4 maintaining 0C as published is practically impossible using
just ice water, so I did let it heat up to 15C.
I'll give the perchlorate salt a go, it's interesting that a lot of these derivatives are so insoluble.
I was thinking benzoic acid may be interesting, if selective reduction is possible followed by diazotisation you'd end up with a
DDNP-nitriminotriazole. The nitriminotriazole moiety seems add a good bit of sensitivity. Theres no detailed procedure specific for
3-Phenyl-1H-1,2,4-triazole-5-amine but it's mentioned in the following. Seems as simple as heating aminoguanidine bicarbonate and benzoic acid in
toluene.
"The compounds mta and pta were synthesized by the condensation reaction between aminoguanidine bicarbonate (amgu) with acetic acid or benzoic acid
(Boechat, Pinheiro, Santos Filho, & Silva, 2011). These compounds were filtered off in air, washed with hexane and kept in desiccators prior to
use in complexation reactions" The Journal of Engineering and Exact Sciences – jCEC, Vol. 08 N. 01 (2022)
"A mixture of aminoguanidine bicarbonate (1) and the appropriate organic acid 2–4 (1 equivalent) was stirred at room temperature to complete release
of CO2 and then toluene (150 mL) was added. The reaction mixture was heated to reflux with a Dean-Stark apparatus under magnetic stirring for 20 h.
The resulting white solid was then cooled to RT, filtered, washed with toluene, and dried. Compounds 5–7 were used without purification." Boechat,
N. et. al. Molecules, 16(9), 8083.
|
|
|
dettoo456
Hazard to Others
Posts: 289
Registered: 12-9-2021
Member Is Offline
|
|
This article references the use of citric acid for nitrosations (at least in this case). I used their process for TNX awhile ago, and got a similarly
decent yield.
[Edited on 17-5-2025 by dettoo456]
Attachment: PEP201742126.pdf (1.3MB)
This file has been downloaded 267 times
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
I prepared what I believe to be the carbohydrazide salt and tested some of its properties:
In the chinese (+ Shreeve) paper they dissolve DNiBT in hot water and then add NaOH to deprotonate it. Then a solution of carbohydrazide sulfate is
added. Due to the very low solubility of DNiBT, I instead suspended about one gram of it in water, added a slight stoichiometric excess of solid NaOH,
heated to 80 C and then gradually added water until all was in solution (about 40 ml was required).
Carbohydrazide sulfate was found to be quite insoluble, so I used the hydrochloride instead. I dissolved about 1.5 molar equivalents of carbohydrazide
in the stoichiometric amount of 10% HCl and added it to the Na2DNiBT soln a little at a time. This caused a very fine precipitate to appear. It was
very similar in appearance to DNiBT, so I worried that lowering the pH (by adding the quite acidic carbohydrazide*2HCl) had simply caused DNiBT to
reprecipitate, so I added more NaOH to neutralize the soln. This didn't cause the precipitate to disappear.
The soln was allowed to cool to room temp, and was then vacuum filtered. The recovered filter cake was air dried overnight, and then for 30 minutes at
110 C in the oven. There was about one gram of pale yellow product. After drying, I tested the melting point, which should be 162C according to the
paper. I found that it decomposed slowly without melting when temp approached 300C.
This makes me wonder if what I have is actually something else, or if maybe the reported melting point is in fact an endotherm spike on a DSC which
has been interpreted as melting but might have been something else (loss of trapped water maybe).
Finally I tested detonability by hand pressing 600 mg in a 7 mm ID brass tube and initiated it with 50 mg uNAP --> 300 mg RDX.
Full detonation was achieved. I have yet to do a plate dent test to compare it with other energetics, and I haven't measured density yet either.
|
|
|
Axt
National Hazard
Posts: 946
Registered: 28-1-2003
Member Is Offline
Mood: No Mood
|
|
It's a shame about the melting point, I suspect due to its poor solubility achieving a high pressed density with a fine powder will be the problem.
Did you try touching off the free acid? as air dried it won't do much (dihydrate), I had to let it sit on a 175C hotplate for an hour or so before it
would pop into yellow fluff.
Throw some pictures up for clarity,
1st is the initial oxalic + aminoguanidine condensation. I did this before having the full information, so I used less HCl solution than the
literature and as a result solidified, although this didn't seem to affect the yield.
2nd is the cyclisation in NaOH, wasn't expecting it to turn brown-grey, not sure if that's a result of the drain cleaner NaOH or not. The cyclisation
went fine though and readily precipitates.
3rd is the nitration to the dinitrimine, it readily dissolves in the concentrated acids and precips on dilution. I used the round flask thinking I
could better cool it, but it was still impossible to keep near 0C. I'm sure a beaker would work just as well and be more convenient.
4th is the diazotisation of DABT, this is using the patent quantities. A 1L beaker is required to contain the foaming. I followed the patent apart
from not acidifying the mixture to destroy the nitrite at the end. the 5th pic shows the result of this.
5th is the recrystalisation of the diazo product, the turbid orange solution passes through the filter and deposits a salmon powder of cooling.
6th is the salmon powder impregnated in absorbent paper, it detonated on a light tap of a hammer and then when I attempted to tear off a bit more it
detonated from the simple act of tearing. I watered the rest into the lawn to dispose of it.
7th is the failed chlorination, it gasses and foamed off to a non-energetic yellow precip. The glass is still stained bright yellow.
[Edited on 19-5-2025 by Axt]
      
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
I examined the achievable loading density of the carbohydrazide salt of DNiBT. I loaded 300 mg into a 7 mm ID brass tube, mounted the tube in a
reinforcing fixture to avoid deformation under pressure, and really pressed it. I used a fairly large bench vice for the pressing, and adopted a
protocol with 30s dwell times after each pressing. I don't have a way to measure the exerted pressure but I cranked it quite hard, and would guess at
least a couple thousand Newtons.
The measured density was 1.83 g/cc corresponding to about 94% TMD. While this is reasonably good, I would say that is not very practical to apply this
amount of pressure, so one of my focus points will be to modify the morphology of the particles.
|
|
|
Boffis
International Hazard
Posts: 1910
Registered: 1-5-2011
Member Is Offline
Mood: No Mood
|
|
| Quote: Originally posted by dettoo456 | This article references the use of citric acid for nitrosations (at least in this case). I used their process for TNX awhile ago, and got a similarly
decent yield.
[Edited on 17-5-2025 by dettoo456] |
Dettoo; I read the paper you attached and I am a little surprised. I post the preparation of TNX on this forum years ago and I can routinely get
>30% yield based on hexamine and sometimes > 50%, using dilute sulphuric acid so I don't know why they claim <5% yields wth this acid.
|
|
|
dettoo456
Hazard to Others
Posts: 289
Registered: 12-9-2021
Member Is Offline
|
|
I only ever used Citric, so I cant attest to how great the yields are with any other acid (carboxylic or mineral). The preparation with Citric may
just be similar to the way in which Tetrazene is prepared with AcOH, even though mineral acids technically can be used instead.
|
|
|
Microtek
National Hazard
Posts: 972
Registered: 23-9-2002
Member Is Offline
Mood: No Mood
|
|
I have prepared what I believe to be Bis(3-(nitroimino)-1H-1,2,4-triazol-5-yl)ethane. I didn't have malonic acid but I did have some succinic, so I
just followed almost the same procedure as the one from malonic. I was hoping for a potential metal free primary, but this is definitely not it. It
may be very sensitive (I haven't done any sensitivity experiments yet), but it doesn't even DDT when heated in foil, it just flashes weakly. It is
very insoluble just like DABT and DNiBT. It might be worthwhile to try for some salts (CHZ or hydroxylamine or hydrazine), but I only prepared it at
the 10 mmol level (and only got a 42 % yield of the diamine after ring closure - I haven't dried most of the nitrated product yet).
|
|
|
|
![[*]](images/xpblue/default_icon.gif){kind=icon}
