Tuesday, October 25, 2011

Academic chemical safety: a discussion with STEM_Wonk

Shudder. (Credit: USCSB)
Dear STEM_Wonk:

How are you? I hope the week is treating you well. Thanks for participating in this little discussion on academic chemical safety. 

So I'm sure you had a chance to watch the US Chemical Safety Board's video on academic chemical safety. I'm a big fan of their videos, even if they totally creep me out. It's kind of awful to watch videos of these animated figures ignoring obvious safety issues and subsequently dying. I have a difficult time not yelling out "Don't do it!" (like I was watching a horror film or something.) Substituting the drawings make them no less gripping; the closeup on the hand of the TTU student (before the compound blows up) is pretty awful. 

What are your thoughts about the incidents and the chemical community's response to them? It seems like I spent a fair bit of time writing about the Sheri Sangji case, all those two to three years ago. It was such a raw red shock at the time -- I can see it in the posts that I was writing. You're not supposed to die from this is still what I remember thinking at the time. I still agree with most of the things that I said: I think Professor Harran bears a significant share of the responsibility, as well as her labmates.

 The Wetterhahn incident, I confess, I don't have much to say about -- it just seems like a terrible tragedy and something that couldn't have been avoided (assuming that the video's version of events was accurate.) I personally wasn't aware that they didn't know the appropriate level of PPE for dimethyl mercury. To a bench chemist, that's pretty scary. 

...and then there's the Texas Tech incident. I think I've said my piece about it, but I was surprised to hear what the Chemical Safety Board had to say. I wanted to get the transcript right, so I went back and listened to exactly what the CSB investigator (now there's an alternative chemistry job!) said (at 13:30):
With these academic incidents, people like to focus on the immediate actions of the individuals involved and try to poke holes and with hindsight and assert some sort of blame on the individual involved and what we have to recognize is that there are bigger systems at play here that can influence safety...
For what it's worth, I think that's true. People love to poke holes (I do!) and there are safety systems that should try stop these things from happening. But as one of the "people" involved in analyzing chemical safety incidents and armchair quarterbacking them, I think there's real worth in doing so. The classic question that chemical workers should ask themselves when confronted with these situations is "What would I have done?" I think it's only natural, healthy and instructive to think through these issues from a bench chemist perspective.

So I'll end with this bit of over-the-top provocation: the lesson to be learned from the TTU incident is "don't do that." Don't make 100 times the amount of an unknown shock-sensitive compound. Don't perform dangerous manipulations without the appropriate safety gear. Don't expose your students to danger by allowing them to be dangerously ignorant of energetic materials safety. Don't allow your institution to be exposed to legal risk from ignorant PIs and insufficiently trained students.

Is there any other lesson to be learned from TTU? Probably. What policy would you implement to stop this from happening again? Can you even do that? I think it might be possible.

[Now to deal with a random leaking pipe somewhere in the lab.]

Cheers, Chemjobber

Note to readers: STEM_Wonk and I will be having a back-and-forth this week on academic chemical safety. Tomorrow's installment with be on her blog. 


  1. About the Sanjy incident: that could have happened to me or other grad students in my grad school group. We used tBuLi quite a bit and we were cautious but if something happened during transfer (say a needle got clogged, very easy when using an organolithium) it could happen that you have a syringe full of extremely pyrophoric material. I wonder if her syringe & syringe barrel were mis-matched (again easy to do in a lab) and that why the barrel came out during transfer. That video was very sobering. We didn't often wear lab coats - sometimes we did and usually when transferring organolithiums but Ms. Sanjy did nothing outrageous.

    I have to admit that video gave me a different perspective on the TTU incident. I had read that this was a rogue grad student who knew very well that he was wrong. It sounds like that was not the case at all. I am not at all surprised that there was no written procedures and training for students using energetic materials. There really should be.

    As far as Prof. Wetterhan - it sounds like nobody did anything wrong, they just didn't realize the danger. However, who made the decision to use methylmercury as a NMR reference compound for Hg NMR. It's so toxic. I hope they've changed this since then.

  2. UI, I can pretty much agree with all of your assessments.

    I disagree with Wetterhahn's incident being lumped in with the others. Sangji's and Brown's incidents show lack of attention to safety in one way or another (though I will agree that Sangji's is more understandable and just about every grad student I know could have been caught in that very same situation). Wetterhahn did everything correctly, including reading the MSDS and following its instructions. Unfortunately, the MSDS failed her and that's why it's such a joke when EH&S tells you that you should always look at the MSDS before working with any new chemical (are any of them not irritants that should be flushed copiously with water? Look up the MSDS for sand to see just how cut & paste these things get). Only after her death did her colleagues conduct the tests that determined the penetrating power of dimethyl mercury.

    It's an insult to her to be lumped in with some dummy who thought it would be safe to grind an explosive material in a mortar & pestle just because it was suspended in hexane.

  3. Wetterhanh incident is unusual only because PI herself paid with her life. With 99.9% of PI's we'd be talking about a dead postdoc or two. As to no mistakes... what was tenured professor, an acting dean of Arts&Sciences, someone who needed help opening an ampule doing in the lab in the first place?
    An interesting side question - how do you support NMR tubes when depositing a sample? In your hand? Put it in a flask?

  4. I completely agree with Unstable Isotope and Dr. Zoidberg that Sangji's accident really could happen in any lab (even with good training). Since that accident, my lab now requires a lab coat (along with goggle, gloves) to be worn when working with pyrophoric reagents.

  5. We handled tBuLi in my lab routinely in the early 90s too. No labcoats, and I don't particularly recall wearing gloves either. One of our main preps involved cannula transfer of an entire 800 mL bottle of the pentane solution, taking it to dryness under vacuum, and then vacuum distilling heptane on to run the overnight reflux. It was considered a rite of passage for the new grad students to partipate in this prep. I shudder to think back on it, really.

  6. Anon 10:53, I don't see how being a tenured professor and dean of A&S disqualifies you from working in the lab, unless you're just wondering why someone of that stature was still in there when they didn't have to be. My point was that she followed procedures carefully but the procedures had incorrect/incomplete information. The others didn't follow proper procedures.

    STEM_Wonk; I've seen a surge in PPE policing these days too. But when you look at the coats that are provided for workers, guess what material they're made out of? Would you guess the same one that provided fuel to burn Sheri Sangji? That would be stupid, right? Most of these places are unfortunately just trying to put a band-aid over the safety issue rather than change the culture.

  7. Dr. Zoidberg, spilling a chemical on yourself is mistake, running an experiment in a way that increases the odds of an accident is a sign of poor form. Think about how you deposit a drop an the back of your hand and then tell me there was nothing wrong with the procedure. As to the lack of info in MSDS - in the lab you always handle all compounds as a deadly poison.
    I'll put it like that - a person who can not open an ampule in most cases is not qualified to handle its content. That, by the way, is pretty much what happened to Ms. Sangji as well.

  8. Dr. Zoidberg, Sheri Sangji was wearing a nylon pullover sweater with a loose weave. Both the composition and the fluffyness of the material contributed to it burning as fiercely as it did. Labcoats made of nylon tend are not only less susceptible to combustion, due to their tighter weave, but are also able to be removed quickly.

    Wow, @5:31, it's easy to be harsh to the dead. I can't say that the chemists I know who handle all compounds as deadly poison get much done. I for one have no problem spilling DI water (a chemical) on myself.
    That being said, Ms. Sangji was properly trained to handle tBuLi on small scales. The issue was that she was not trained to handle larger scale reactions (notably on the use of cannula transfers). There is precious little mention of proper precautions and methods in the scale-up of reactions in most safety courses, or most laboratory manuals.

  9. "nylon pullover sweater with a loose weave."

    Uh, how could you possibly know that? Also, do they make nylon sweaters?