Process Wednesday: data comes from the weirdest places

There sure seem to be a lot of articles about working with hydrazoic acid safely in Organic Process Research and Development recently; all to the good! In the latest one (González-Bobes et al. [1]), the authors perform an "enantioselective palladium-catalyzed desymmetrization of a meso-bis-ester using TMS-N₃. They do quite a bit of work towards making sure that there's not too much hydrazoic acid in the headspace -- after all, it's both terribly toxic and explosive. From the article:

To maintain safety during downstream processing, a high pH and/or near complete reaction conversion must be achieved. Since the process required an acidification for intermediate stabilization (approximate pH 3.5), our control strategy relies on achieveing high reaction conversions (low residual free azide.) Therefore, the next step in the quantitative analysis becames the determination of the accetable residual free azide level after the palladium-catalyzed desymmetrization reaction. To accomplish this, we first interpreted data from the PUREX process, a nuclear fuel reprocessing method which forms HN3 as a byproduct. Using this data, we established a gas phase limit of 0.625 vol % which is based on the enriched condensate LEL in equilibrium wit the solution and gas phases. 

As a relative youngster, I'm not really aware of the PUREX process, so I thought I would take a look at it. It's pretty remarkable stuff, being able to extract out uranium and plutonium from spent nuclear fuel with tributylphosphate. Here's an excerpt from a random ORNL paper from 1977 on using hydrazine (the source of the hydrazoic acid) in the PUREX process:

In order to separate uranium and plutonium from each other in PUREX processing of nuclear reactor fuelds, advantage is taken of the fact that uranium (VI) is readily extracted from nitric acid solutions into the tributylphosphate-diluent organic phase while plutonium (III) is relatively inextractable. The uranium and plutonium are usually coextracted from the nitric acid fuel dissolver solution as uranium (VI) and plutonium (IV) and are then differentially stripped. The plutonium is reduced to plutonium (III) which strips into a dilute nitric acid solution leaving the uranium (VI) in the organic phase to be subsequently stripped with water. In some flowsheets, the plutonium is reduced prior to the uranium extraction in order to achieve the desired separation.  

Plutonium (III) is not stable in nitric acid and, without the addition of a holding reductant, rapid and complete reoxidation to plutonium (IV) may occur. This can lead to reextraction of the plutonium during reductive stripping and plutonium recycled within the countercurrent contacting apparatus... Holding reductants are routinely added to solutions of plutonium (III) to destroy the nitrous acid which is continously formed by radiolytic decomposition of nitric acid. Very few reagents have proven useful as holding reductants... 

Hydrazine rapidly reactions with nitrous acid, and the favorable kinetics of this reaction make hydrazine a practical holding reductant. With excess hydrazine present, the condition prevailing when hydrazine is used as a holding reductant, the fast reaction  

N₂H₄ + HNO₂ -> HN₃ + 2H₂O 

occurs which stoichiometrically yields hydrazoic acid, HN₃. Hydrazoic acid is relatively stable in nitric acid solutions. 

There's something amusing about a situation in which rocket fuel gets added to nuclear waste to generate a "relatively stable" explosive waste product. My (figurative) hat's off to Kelmers and Browning for an education in nuclear fuel reprocessing, and to González-Bobes et al. for leading me there.

[1] González-Bobes, F.; Kopp, N.; Li, L.; Deerberg, J.; Sharma, P.; Leung, S.; Davies, M.; Bush, J.; Hamm, J.; Hrytsak. M. "Scale-up of Azide Chemistry: A Case Study." Org. Process. Res. Dev. ASAP. DOI: 10.1021/op3002646