This paper will be a window into a world that most of us synthetic organic chemists haven’t explored very much: the deliberate optimization of what are delicately called “energetic materials”, aka explosives. Now, a lot of us have had things go up on us, that’s for sure, but most of the time it’s a surprise. The folks who do this sort of work for a living, though – such as the Army Research Lab at Aberdeen Proving Ground or the Naval Weapons Research Center at China Lake – are in the business of removing the element of surprise completely.
Which makes a lot of sense. A really useful explosive or propellant would be completely safe to handle, with no risk of setting it off through heating, impact, friction, electric sparks, or any other nonspecific physical trigger. It would not deteriorate on storage, not react readily with other common formulation materials (or with the walls of its containers), and have a sharp melting point at a useful temperature that would not cause problems under ordinary conditions. It would not be volatile (no sneaky evaporation or sublimation). But what it would do is release very large amounts of energy under very specific conditions (and no others!) and in a clean, controlled, reproducible fashion. Its properties would be finely fitted to the job: the ideal behavior you want from a propellant, for example, is different from what you want in a high explosive, and something that’s really good at one of those jobs is likely to be suboptimal at the other (no one launches rockets with dynamite, for example). You definitely don’t want them switching roles on you unexpectedly!
The ideal explosive doesn’t quite exist, although there are some pretty good ones. But each of those have their own properties which make them better at some things than others. For example, good old trinitrotoluene (TNT) is the classic “melt-castable” explosive: yep, you can indeed melt TNT (at 81C) and cast it into various useful shapes, although I don’t recommend trying this at home. Liquid TNT also dissolves some of the other common explosives (or can be formulated with them), leading to all sorts of blends and compositions. One of TNT’s most notable properties, actually, which can be a bug or a feature depending on the situation (usually the latter) is how relatively hard it is to detonate. You really need another explosion to do the job, from something more lively. On the other hand, the commonly used military explosive RDX (more energetic than TNT) melts at 204C, but starts to decompose well before that and is thus really unsuitable as a stand-alone melt-castable substance. Roughly 60/40 RDX/TNT (with some other additives), known as “Composition B” is a better melting mix, and there are plenty of others in the same vein. RDX is also not so easy to set off – unconfined by itself at room temperature it just burns rather than explodes. Soldiers have actually been poisoned by using C4 explosive mix (RDX with various plasticizers and binders) as campfire fuel for cooking, so don’t do that. I am told that you can even shoot pistols into the stuff, although that is also not a recommended way to spend a rainy afternoon. But you can see why vast quantities of both of these agents have been used over the decades, and continue to be used today. There are plenty of compounds that pack more explosive power than either RDX or TNT, but the problem with most of them is that they just don’t have the combination of properties that makes those two so useful. Prominent among these deficiencies is the problem of being too damn touchy, to use a nontechnical phrase.
Outside of the structural and physical properties like melting points, each explosive has of course its own characteristics when it, well, explodes. RDX, for example, has good brisance, a word that you just don’t hear much outside of the explosives literature. It means, roughly, “shattering power”, and it’s proportional to the detonation pressure a given explosive can achieve. Detonation velocity is important, too. If you’re working on propellants, you’ll similarly be looking at specific impulse, which is related to exhaust velocity. These features are in turn related to the intrinsic amount of chemical energy that a structure has in it to be released, and to things like density and crystalline form, efficiency of combustion, and others, and these all come together in ways that are very difficult to predict. TNT, as an example, has a pretty high activation energy for decomposition in the gas phase, but that barrier is much lower in the solid form because there are bimolecular decomposition mechanisms that can’t happen with isolated gas-phase molecules. There are algorithms that can give you some idea of these things, but the only way to really see how a new energetic compound performs in all the categories is to make some and test it. The similarities to the multi-parameter optimization that we have to do in drug design (and the uncertainties about what you’ll get, in the absence of hard data) are notable.
And that brings us back to the paper linked to above. It’s looking at an underexplored class of compounds, nitroazetidines (with a core four-membered nitrogen-containing ring). As the paper notes, trinitroazetidine (TNAZ, N-nitrated and bis-nitrated on the opposite carbon of the ring) was described in 1983, and is considerably more energetic than TNT while also being melt-castable. Unfortunately, it also has a high vapor pressure (too symmetrical a molecule, it looks like to me – it probably sublimes pretty easily), and that has taken it out of contention for real-world use. The biggest barrier to progress in the area since TNAZ has been lack of good synthetic routes, which is what this team (from the Army Research Lab and the University of Michigan) is trying to remedy.
As you’ll see from the preprint, they’ve got a photochemical route to the four-membered ring from dihydroisoxazole starting materials (a cyclic oxime ether, really), followed by N-O cleavage to leave you with usefully substituted azetidines, which are then nitrated. That last phrase is surely underselling things: nitration is one of those reactions that leaves you thoughtfully paging through a pile of literature references, because there are a lot of ways to do it. That, according to one of my old Laws of the Lab, is a sure sign that there is no really general method available – what you’re seeing is the residue of people having to invent new ones because the existing reactions failed. Some compounds just don’t stand up to the classic routes – and no wonder, since we’re talking fuming nitric acid, for starters. Others are just resistant to being nitrated under standard conditions, and need reactions with somewhat different mechanisms to push things through. With these compounds, for example, good ol’ nitric acid just wasn’t suitable, and a lot of the nitrations are done with acetyl nitrate instead.
This is relatively unexplored territory, the various substituted azetidine cores give a whole range of properties depending on the ring substitution patterns and relative stereochemistries (a fruitful area of exploration in recent years). This team discovered some really promising melt-castable solids (with melting points up in the 80C range), while related materials turned out to be liquids with very low freezing points. That makes them candidates as energetic propellant plasticizers, and two of them look like they’re worthy of being followed up on in that category. Meanwhile, there are of course other in-between compounds that fall into the neither-explosive-fish-nor-detonating-fowl category, low-melting solids that would be a pain to handle and work with. But the best candidates have both good physical properties and good stability to handling (in between RDX and PETN, another commonly handled explosive). And this is combined with higher detonation pressures and velocities than TNT for the solids, and high specific impulse for the liquids. And of course you don’t have to nitrate these cores. They’re interesting little heterocyclic building blocks all by themselves, and now they have published routes to them, so even those of us who aren’t blowing things up or accelerating them into the sky can find some uses for them!
