You Should've Known Azides Are Dangerous!

An interesting topic came up the other day in a Ph.D. seminar.  A student from a competing group had mentioned their recently published methodology paper, which was directly related to their natural product synthesis.  The methodology in question involved use of a very strong reducing agent to effectively obtain a variety of synthons for use in said total synthesis.  The project seemed routine methodology: vary the temperature, reductant loading, solvent, analyze dr by ¹H-NMR, repeat.

During the Q+A, a well-respected professor in the front of the room raised his hand.  “I see in your optimized results, you use methylene chloride as your solvent.  Have you measured the calometric outcomes of these reactions?”  “No.”  “Interesting, that’s something you may want to investigate.  Methylene chloride does some strange chemistry when it’s used as a solvent in reduction reactions.  There have been numerous explosions attributed to using a strong reductant in DCM.”

Later on that day, I spent a while searching for DCM explosions involving strong reducing agents.  While I couldn’t really find a clear-cut examples (with the exception of some rare Norrish chemistry) I eventually stumbled on to several reported explosions where methylene chloride was the solvent in the installation of an azide.  According to Peet
and Weintraub, “The culprit in these detonations is undoubtedly (the formation of) diazidomethane, which is unstable in high concentrations in solution.”  More digging uncovered that a drop of diazidomethane is enough to set off a reasonably catastrophic explosion.

In a paper that I would classify as some of the balliest experiments in the 20^th century, Hassner and co-workers studied mono-, di- and triazidomethanes ultimately publishing ¹J_CH values for each compound (J. Org. Chem. 1990, 55, 2304-2306).  Their synthesis, which involved “occasional shaking” of the reaction vessel over the course of 45 days, included GC to analyze for purity.  Why does that just sound like a bad idea?

More recently, Conrow and Dean published their account of inadvertent diazidomethane formation, which caused the destruction of a 20 L rotovap at Alcon a few years ago.  Synthetically, they were attempting to install an azide from the corresponding alcohol (simple mesylation followed by S_N2 displacement).  Investigation of the incident concluded that residual DCM from the sulfonyl ester formation was present in the reaction flask when the sodium azide and DMF were introduced.  Their paper is concluded by urging readers to “raise and reinforce awareness of this insidious hazard.”

If you want to get a little more daring, there’s always tetraazidomethane which was recently synthesized by Banert and co-workers in Germany (Angew. Chem. Int. Ed. 2007, 46, 1168–1171).  They propose that it might be a useful, high-energy material in real world applications (instinct would suggest that the US military is already playing around with the idea).

And now for the crux of my post.  Shouldn’t there be rigorous training about hazardous materials?  As synthetic chemists, we work with them all the time, yet I’m certain that most probably couldn’t tell you why HMPA is dubbed “liquid cancer.”  I just know that it is, and “You should avoid it like the plague.”  A former (remarkably careless) colleague used to use sodium azide on a daily basis.  Can you imagine what would’ve happened if they got a wild hare up their backside and decided to use DCM as a solvent to see what would happen?

I like to learn, and through the course of my research for this post, I stumbled onto Derek Lowe’s blog entitled “In the Pipeline.”  For those of you not familiar, Lowe occasionally runs “Things I Won’t Work With” posts, which describe the nastiest of the nasty compounds that he claims he tends to avoid.  In my mind, this begs a very important question.  Why isn’t there a reference textbook like this?  I am deeply troubled by the notion that good laboratory sense is not learned but rather seemingly taken in through osmosis. 

There are some things in lab that should be trial and error: learning how to run an efficient column, vacuum line technique, chemical inventory.  Lab safety should not be one of them.  There should be no room for error when it comes to the chemicals we use on a daily basis.  I recall sitting through hours of training when I worked in the biotech industry and for an analytical firm when I was an undergrad.  I think I get one hour of training every year at my University, and we inherently do much more dangerous things than I could’ve ever done in front of a GC or gowned up in a laminar flow hood. 

Perhaps my university is the only one with this problem, which if that’s the case then you can ignore my rant.  I am curious to hear from other Universities and businesses.  Namely, what sort of training do you receive?  How often do you get a refresher?  Do I have a point to the training/resource or are you ready to throw the BS card at me?

P.S. I have officially moved hoods due to the “incident” that happened last week.  Apparently the University wants to claim the hood as “lost,” and I’m not to go anywhere near it until maintenance can test its electrical system (the wheels of bureaucracy are slowly turning).  We did, however, come across a 2” puddle of water in our reductants cabinet (which contains sodium metal), and that discovery caused a fit of panic.  Nevertheless, we’re moving on.