SMX (1,4-dinitrato-2,3-dinitro-2,3-bis(nitratomethyl)butane) is the highest density nitrate ester known and as far as I'm aware the highest performing
castable explosive known (defined as <100C mp), both in brisance and power due to its perfect oxygen balance (0% to CO2). It's detonation velocity
and pressure are comparable to HMX, but I've never seen measured properties apart from its density of 1.917g/cm3. Calculated properties give it
9200m/s and 40GPa det pressure.
I ran across a far more convenient route to SMX, the intermediate tris(hydroxymethly)nitromethane (THMN) is made via a much cleaner reaction of
nitromethane, paraformaldehyde with a touch on KOH in ethyl acetate which produces very nice white needle crystals free of the orange resin which
forms when done in aqueous solution. I've repeated this reaction, it is very good.
The next reactions run very smoothly as well, The dehydrative condensation of THMN to acetone using P2O5, surprisingly results in a nice white
crystalline powder when the resulting gum is dissolved in water.
The 3rd step, the oxidative coupling of the ketal protected compound is smooth as well, using NaOH and sodium persulphate, its precipitates freely
from aqueous solution, the published reaction time I believe may be excessive it seems to drop out quite quickly. However, recrystallisation from
ethanol was necessary to free it from an insoluble higher melting point product. It forms nice glittering plates from ethanol tested for correct mp.
of 131C.
The only real stumbling block is the nitration, you have two choices, direct nitration of the ketal protected intermediate or acid hydrolysis to free
the tetramethylol product before nitration. The later, being way less convenient but supposedly results in higher yields overall. I went with direct
nitration, first using fuming HNO3 as in the literature then with KNO3/H2SO4. The fuming HNO3 attempt was a bit of a mess as I didn't recrystallise
the intermediate which was shown to be less than pure. I won't comment much on it, but a small yield was produced that was difficult to free from an
oily product, it was very explosive though.
H2SO4/KNO3 nitration went very smoothly, there is practically no exothermic nature to it. Added 5 g recrystallised dimeric ketal protected
intermediate to 25 g KNO3 in 60 mL H2SO4 in a few portions at 5C but I think I could have just thrown it all in at once. The glittering crystals after
stirring for a while turned into a milky suspension. No further cooling was used, ambient was 6C and reaction temp never went above 10C. Dunked it in
200 mL water with 200 g of ice, a white powderous precipitate formed (using just HNO3 resulted initially in an oil). Filtered, washed with bicarb.
Recrystallisation was done, 1 g of product quickly dissolved into 10 g of hot isopropanol, cooling produced what looked like a coagulated resinous
mass but this quickly turned into a fine white crystalline precipitate that was filtered and dried. Yield was about 4g from the 5g of precursor.
Testing the melting point, just by placing some on the centre of a thermostat hotplate with temp readout gave 75C mp, this is bad as it should be 85C.
But testing purified ETN gave a 55C reading whereas it should be 61C. It thinking the hotplate just heats up too fast in the initial stages and the
probe cannot keep up. This method is fairly consistant within a few degrees at higher temps. It was spot on at reading the 131C mp of the
intermediate, but at that range the temp it is increasing far slower.
Testing the presumed SMX, it deflagrates much the same as PETN, with no real residue but leaves a tarnished stain on Al foil. It seems about as impact
sensitive as PETN, as it should based on the literature. It propagates detonation easier under the blow of a ball peen hammer, what I mean by this is
that PETN will snap and spread the loose surrounds about, whereas SMX is more likely to transmit its detonation into it giving a much louder and
forceful bang. On the hotplate mentioned earlier, it melts at "75C", starts to bubble and decompose at 145C, vigorous bubbling and red fumes at 165C
and ignites at 187C.
Some pictures. A hammer blow where it propagates into the loose powder for a powerful detonation. The crystals from isopropanol were very small,
microscope struggles here but were quite spherical in nature.
I did some work on SMX in my final thesis at the university (though I called it NDBD at the time). I tested the different nitration systems I had
access to and found that HNO3/Ac2O, mixed acids and N2O5 in HNO3 (via P2O5) all gave a decent yield (77-85 % crude) when nitrating. I only nitrated
the deprotected substrate (and the deprotection was a little annoying, IIRC). I did Raman spectra of the synthesized compounds and they corresponded
nicely with the predicted spectrum, clearly showing the absense of -O-H (so indicating full nitration). greenlight - 6-11-2025 at 10:15
Really impressive work!
Would be very interested to see how it goes in a melting point apparatus where the temperature ramps much more gradually. Maybe there is still a
small amount of impurity in there forming a eutectic and lowering the melting point slightly if it isn't the hotplate.
Is it possible that the crystal structure/type forms hotspots more efficiently resulting in enhanced shockwave propagation in the hammer test. That
coupled with the higher crystal density would be beneficial. Mush - 6-11-2025 at 16:17
Experiments and simulations on interactions between 2,3-bis(hydroxymethyl)-2,3-dinitro-1,4-butanediol tetranitrate (DNTN) with some energetic
components and inert materials
doi.org/10.1016/j.fpc.2021.06.001
3.3.1. The impact and friction sensitivities
There are many uses for knowing the sensitivity of energetic materials to impact and friction stimuli, which can propose the evaluation of energetic
materials’ practical application over time. The sensitivity of the systems between DNTN with some energetic materials was confirmed, and the results
listed in Table 4. From the results depicted in Table 4, it can be seen that the tested mixtures are more sensitive to friction stimuli for which the
values for most of the mixtures are 100 % under the investigated condition, while the impact sensitivity is lower than that of pure DNTN sample. From
the impact sensitivity viewpoint, the DNTN in combination with some energetic materials are sensitive except for (5), (7) and (11). The RDX, HMX and
CL-20 in systems (5), (7) and (11) are very sensible, and it makes the system more sensitive. Binary systems (9), (13) and (15) are of great
compatibility and low sensitivity. Contrasting the sensitivity results with compatibility experiments data that there was no consequential trend
between the sensitivity and compatibility. Systems (17) and (19) are incompatible while their impact and friction sensitivities are very low. Binary
system (11) displays a moderate compatibility, while its impact and friction sensitivity is very high. However, if the energetic materials can react
with each other when they are mixed, they must be incompatible and very sensitive.
Melting Point 86.8 °C Melts before decomposition.
Decomposition Temp. 185.1 °C (exothermic peak)
Temperature at which rapid decomposition begins under standard test conditions.
Density (Theoretical) ~1.853 g/cm³ Varies in mixtures, e.g., 1.697–1.752 g/cm³ in DNTN-based propellants.
Impact Sensitivity Moderate (e.g., 10 cm H50 value for pure DNTN) Generally lower than CL-20, but can be sensitive in some mixtures.
Friction Sensitivity High (e.g., 100% reaction under tested conditions for pure DNTN) More sensitive to friction than impact stimuli.Axt - 6-11-2025 at 23:27
Everyone seems to designate a different name to it, NEST-1 and Hisk75 are a couple more, but SMX seems to have stuck the most. I think its hype died
down after its thermal instability became apparent, so like the vast majority of discoveries it's unlikely to ever find use in industry.
I assume you were using the patent route Microtek? Theres a lot of hurdles in route for the amateur. The simple condensation of NIBG with acetone
using P2O5 was the biggest key to opening it up. The "dimerization" via NaOH and persulphate confuses me, why is it heated with NaOH for an hour
before the addition of persulphate?
If I was to do it again I'd probably use a larger excess of KNO3/H2SO4, I used a 3x excess but it's likely that the protecting fragment is also being
subjected to nitration, possibly to nitroform so this should be accounted for.
Yeh Greenlight. It's totally plausible that it's the crystal form and size that allow ease of propagation, but I don't know of any specific study on
that. It does seem likely there is one somewhere though. Comparing critical diameter to crystal size/morphology.
Yes, I used dimethoxypropane for the protection step. Very clean and easy reaction. I used way too much time trying out different ways to prepare
dialkoxypropane from acetone, alcohol and different dehydrating agents. In the end, I just bought it.
Regarding the coupling reaction, I think the heating step with NaOH is to dissolve the substrate before the addition of persulfate.MineMan - 7-11-2025 at 03:06
AXT. You mention thermal instability, what would be a reasonable decomposition temp for an industry suitable melt cast? My understanding is a 100c
past casting temperature?Microtek - 7-11-2025 at 11:39
It's not so much the decomposition temperature that is the problem but rather that in 18 days at 20 C, 0.02 % of the material decomposes. In
comparison RDX takes a couple hundred years to do the same. Mush - 7-11-2025 at 16:41
Google's AI overview summed up very well what the problem is with SMX.
"The primary disadvantages of the explosive DNTN (2,3-bis(hydroxymethyl)-2,3-dinitro-1,4-butanediol tetranitrate) include potential
compatibility issues with other energetic materials and propellants, which can lead to instability, and a high sensitivity to
friction stimuli.
Disadvantages of DNTN
Compatibility Issues: DNTN exhibits poor to bad compatibility when mixed with certain energetic components and inert materials, such
as dinitrofuroxan (DNTF), N-guanylurea-dinitramide (GUDN), and N-butyl-N-(2-nitroxy-ethyl)nitramine (Bu-NENA). This incompatibility can result in a
significant decrease in the mixture's thermal decomposition temperature, making the combination hazardous and unsuitable for long-term storage or use
in practical applications.
High Friction Sensitivity: While its impact sensitivity can be low in certain mixtures, DNTN mixtures generally show high sensitivity
to friction stimuli, with most tested mixtures showing 100% sensitivity under standard test conditions. This necessitates careful handling and
specific manufacturing processes to ensure safety.
Need for Stabilizing Agents: Like other aliphatic nitrate esters, DNTN can be unstable under ambient conditions. Stabilizing agents
must be incorporated into energetic compositions to inhibit or slow down decomposition reactions.
Potential Toxicity: Although research often focuses on performance and physical properties, many explosive materials, including DNTN,
are known to have toxic and/or carcinogenic effects, posing potential health and environmental risks that require strict safety protocols and
remediation strategies.
In short, while DNTN is a high-energy material with good performance potential, its use is challenging due to specific compatibility issues with other
components in a mixture and its sensitivity to friction, requiring careful formulation and handling procedures."
[Edited on 8-11-2025 by Mush]Axt - 12-11-2025 at 12:36
I tried mixed 68% nitric acid and 98% sulphuric acid. This is much hotter than the fuming nitric or mixed nitrate salt methods. I'm guessing the
higher water content initiated deprotection easier (hydrolysis to acetone) and it accelerated into a runaway, nothing was recovered. Mixed at -5C,
temp lay dormant, but the crystals looked to form into resinous small beads. On allowing it to slowly warm to room temp it fired off into a red
volcano, this was all outside so I let it run its course. For the record 5 g dimer, 23 g 68% HNO3, 64 g 98% H2SO4, running this same protocol on
erythritol for example would have been a totally fine cool nitration.
Regarding the coupling reaction, I think the heating step with NaOH is to dissolve the substrate before the addition of persulfate.
Mines immediately dissolving in hot NaOH so heating for 1 hr I don't think is necessary. A bit like the 20 hr oxidation, it seems excessive based on
the observed precipitation but since its run at room temp it's not such a bother.
AXT. You mention thermal instability, what would be a reasonable decomposition temp for an industry suitable melt cast? My understanding is a 100c
past casting temperature?
I guess if an arbitrary figure was wanted 100C seems reasonable. It depends on the scale, you'll never heat a large quantity evenly enough, so you
need a larger safety margin. It's a bit easier on the smaller scale.
Axt - 25-11-2025 at 23:07
Just an update on the melting point, I placed a small portion of naphthalene next to the SMX and slowly raised the temperature 5C at a time, letting
it stabilise from 70C. Nap melted at 81C (lit. 78-81C), SMX melted at 84C (lit 85-86C). Considering the crudeness of this test I'm convinced it is
relatively pure SMX at this point, the higher decomposition and ignition temps match that which is published as well. It's possible that the earlier
75C test was holding onto residual recrystallisation solvent that has been lost in the past 2 weeks, as all these tests are of the same batch.
