You know, normally when you start combining interesting or reactive functional groups in the same molecule, you end up with something that’s worse than before. Would I pick up a flask containing a compound that has both a perchloryl ester and a geminal di-azide? I would not, and neither should you, should someone ever be foolhardy enough to prepare such a thing. Those are only for lobbing at hostile monsters in video games. So I was interested to see a recent paper on perfluoroalkyl peroxides. Changing the electronics of that group would surely have an effect, but of what kind? (And perfluoroalkyls want all the electron density, so you know that something is going to change).
Well, to my surprise, the result is the friendliest batch of alkyl peroxides that you’re ever going to see. There was some foreshadowing, though. It turns out that the only member of this compound class that had really been characterized in the literature to date was the bis-trifluoromethylperoxide, which dates back to 1933. It was found to be weirdly stable, as in up to 200 degrees C. Do not, under any circumstances, try heating up a regular dialkyl peroxide in such a manner – you’re not going to get very far, although the pieces of the reaction vessel will.
Preparing these things is worse than handling them once they’re made, by far. As this paper shows, the previous syntheses relied on rather spirited reagents like chlorine trifluoride, and when your prep calls for that one, it’s time to re-examine the life choices that have brought you to that point. That’s what this Berlin group did, clearly, coming up with a route that involves some generally unappealing chemistry, such the treatment of silver wool with elemental fluorine gas at 100 degrees. OK, you don’t start at that temperature, no matter how much of a buckaroo you are – they introduce “small portions” of fluorine to the silver in a sealed metal apparatus at RT “until the pressure remains constant”, that is to say, until it stops whooping and hollering in there, and then they raise the temperature. You also have to prepare hypofluorites of the alkyl groups, which you do from the alcohol and cesium fluoride, introducing (yet again) elemental fluorine in portions until all the overpressures stop spiking, and then do trap-to-trap distillation at -150C. Treating the hypofluorite with that silver fluoride/silver wool, followed by another trap-to-trap purification, gives you the desired perfluoroalkyl peroxide.
None of this am I lining up to perform. Straight fluorine has long been on my “Things I Won’t Work With” list, although if I had to choose, I would go for that over using chlorine trifluoride, albeit with profound unhappiness and thoughts of chucking the whole business and opening a taco truck instead.But at the end of all this, the compounds you get out are, as mentioned, surprisingly tame. The bis(perfluoro-t-butyl) peroxide, for example, melts at about 18C, so at most room temperatures it’s colorless liquid that just sits there, bizarrely. It’s insensitive to shock, insensitive to friction, and you can even get a boiling point on it (99C), albeit with decomposition.
You cannot say any of that about the plain bis-alkylperoxides, that’s for damned sure, except for the decomposition part, which unfortunately can happen with fearsome speed when you so much as look at them funny. So what’s going on? The paper notes that the O-O bond energy isn’t that different in the two sorts of compounds. What changes, though, is the energetics of the alkoxy radicals that form when that bond cleaves. The reactions of the perfluoroalkoxy species are slow and endothermic; the reactions of the regular alkyls, definitely not. The paper goes into a number of structural and computational details – there have been some longstanding arguments about peroxide conformations and bond angles versus theoretical predictions, and these compounds have some bearing on that.
So there you have it: if you’re willing to walk through the valley of elemental fluorine, you can arrive at these odd compounds. It’s be interesting to see what sort of reactivity they have as selective oxidants. Who knows, they may turn into familiar reagents if they do something useful, or they may just remain as chemical curiosities. They’ll always be that!