Wednesday, December 28, 2011

UC Regents and Prof. Patrick Harran to face 3 felony charges in death of Sheri Sangji

Recap:  On December 29, 2008, Sheharbano 'Sheri' Sangji was working as a research technician in the laboratory of Professor Patrick Harran at UCLA. She was severely burned during the transfer of a 1.7M solution of tert-butyllithium (in pentane) using a 60 mL syringe and a 1.5 inch needle; the procedure called for 159.5 mL of tBuLi. She was intending to generate vinyllithium for the addition of a vinyl group to a ketone to generate a tertiary alcohol. The syringe came apart and the tBuLi (naturally) ignited.

She was wearing a polyester sweater, which caught on fire. Her labmates used a lab coat and water from the sink to put out the flames. She was transported to the hospital, where she died of injuries from her burns on January 16, 2009.

The news: Yesterday, the LA Times reported that Professor Harran and the UC Regents are both charged by the LA County District Attorney's office with 3 felony counts of "Willful Violation of Occupational Health and Safety Standard Causing the Death of An Employee." The 3 counts have to do with (in Paul's formulation) "failing to correct unsafe work conditions in a timely manner, to require clothing appropriate for the work being done and to provide proper chemical safety training." Harran faces a maximum 4.5 years in jail, while UCLA faces a maximum $4.5 million in fines if found guilty.

The reaction of the charged: Professor Harran is out of town and will surrender himself when he returns, according to his attorney. Harran's bail is set at $20,000.

UCLA has come out swinging in their press release. An excerpt:
Following a meeting with the district attorney in October 2010, UCLA had not been contacted by the district attorney or received any requests for documents or interviews until being notified about pending charges two days before Christmas. The district attorney's decision to file charges today is truly baffling and directly contradicts the findings of the state agency responsible for evaluating workplace safety.

Since the accident, UCLA has dramatically increased the number of laboratory inspections and established even more rigorous safety standards. UCLA's recently created Center for Laboratory Safety has become a leader in the field, and other universities and research organizations look to the center as a comprehensive resource on lab safety.

The facts provide absolutely no basis for the appalling allegation of criminal conduct, and UCLA is confident an impartial jury would agree.
I find it ironic that neither the names of Professor Harran nor Ms. Sangji are included in their press release. As I expressed in an e-mail at the time of the founding of the center, "I find most of UCLA's changes (the high-profile chancellor-level surprise inspections, the lab safety center) to be the worst sort of top-down cosmetic change; it will irritate bench-level workers, alienate PIs and promote the feeling that most post-accident safety changes are about restoring the image of the institution rather than promoting the safety of the individual worker."

I think this press release suggests that I might have been right. Harran's sense that "the administration and staff are scrambling to protect their own [hides]" is pretty much dead on as well.

The public reaction: Online comments on the articles are pretty much a mixed bag of idiocy, overreaction to over-litigiousness and appeal to authority. On the other hand, I was struck by the comment of a certain UC-Irvine professor (scroll down):
I think that's fairly reasonable, although we're going to re-litigate again and again and again how experienced Sheri Sangji was. There was this JACS paper and some analytical/computation work; I would consider her considerably more experienced than your average 1st year graduate student. That said, it doesn't speak directly to her ability to manipulate pyrophorics.

My speculation: Just like Paul, I can't imagine that Professor Harran will actually serve any jail time for this; I presume that this is an opening gambit between the DA's office, the UC Regents and Harran to extract  some sort of plea bargain. I predict fines for the UC system, some administrative changes and community service for Harran.

The repercussions for the academic synthetic community are large and severe. Professor Patrick Harran is not a small name in synthetic chemistry; he's well known for the quality of his work, especially for a relatively young researcher. I believe the level of unpreparedness and the lack of safety in his laboratory was no more than one standard deviation away from the median synthetic lab in academia.

Because of that, this case cannot be swept under the rug or forgotten. Pictures of Professor Harran and Ms. Sangji will grace Powerpoint slides of safety offices throughout the US academic science community from now on; I predict that "You don't want to be another UCLA" will become a refrain from top to bottom.

Finally, I'm on record saying that (this is the full quote): "I think that it will take a professor (Patrick Harran?) being railroaded to really engender change on the part of academia overall, i.e. some professor being punished (and probably unfairly) that will *really* get the academic chemistry community to seriously address the issue from the professor level down." We're about to find out, and a lot sooner than I expected.


  1. Wonder when the DA is up for re-election.

  2. 2012, looks like:

  3. I am considering a job change from industry to academia, and this case has been weighing on me for some time. The multiple hours of rigorous (and mandatory) safety training I have undergone in industry dwarfs the yearly chemical safety lecture I received in graduate school. But, as a professor, how do you go about instituting and reinforcing a positive safety culture in your lab? For instance, with industry, you have a very effective "stick" (firing) for instituting safety policies. There is not a very comparable one that can maintain your status as a professor in the same way. If you "fire" too many students, it would likely be viewed as poor performance on your part.

    There has to be a way to balance the low friction environment of academia with the enormous barriers of industry (writing SOP/SOC's for *everything*, multiple signature levels, etc.) Does anyone have suggestions on forcing safety on graduate students?

  4. Why aren't they charging the plastic syringe manufacturer while they're at it?

  5. Do it the way the Navy does it. Train and drill a lot.

    Once every six months, walk into your lab of young kiddos on a Thursday afternoon and announce that Jill's hood just blew the hell up and she's lying on the floor with a sep funnel in her arm. What do you do?

  6. 2012? This is pure politics, nothing more.

  7. I'm just as cynical as the next guy, but what possible political gain could come from this? This is an A3 story -- I doubt it will hit the news very much at all.

  8. Taking on the entire UC system in an election year? It's a low-risk, high-reward gamble to position himself as a crusader for workers' rights or some such nonsense.

    Plus, in an election cycle with rampant Tea-Party anti-intellectualism, arresting a professor could score him some extra votes.

  9. @CJ - I can't speak for anyone but my own experience in grad school / pdoc, but I can say this - the safety training was never adequate for all the various dangers we might encounter.

    To be fair, the Safety / Facilities department is usually woefully understaffed; even at major universities, it's not uncommon to have 250 grad students and 1-2 safety officers to watch over all of them.

    That said, safety training was a once-a-year affair, for about 40 minutes. It was usually Powerpoint, with lots of "don't do this!" type pictures, and a heavy focus on waste disposal, biosafety, and sharps. Many times, chemists would roll their eyes when told to look at MSDS copies for health and safety emergencies (they're notoriously vague) and didn't receive information on what to do for real-life lab accidents such as poisonings or explosions.

    In addition, many safety officials in schools are there to conduct OSHA inspections and check waste labels, and themselves are not chemists, nor have most spent time working in synthetic labs.

  10. That's the problem, so much of the "safety" stuff is useless garbage, the parts that are relevant, like how to handle tBuLi, are either not covered (because the bureaucrats in charge never spent a day in a lab) or are buried in with the chaff and ignored along with it.

  11. as CJ says:
    "Do it the way the Navy does it. Train and drill a lot.

    Once every six months, walk into your lab of young kiddos on a Thursday afternoon and announce that Jill's hood just blew the hell up and she's lying on the floor with a sep funnel in her arm. What do you do?"

    I think that is appropriate.
    what is so difficult about running these drills every now and then? of course you cannot anticipate every thing that can go wrong.

    but as @See Arr Oh says, the safety training really is never adequate for what we can encounter every time we run an experiment. but it really is also training us to be smart, to just begin to anticipate what could go wrong and how to quickly counter anything that does go wrong. Everywhere else, not just the NAVY, has contingency plans. why not academic labs?

  12. ...the safety training was never adequate for all the various dangers we might encounter.

    To be fair, the Safety / Facilities department is usually woefully understaffed; even at major universities, it's not uncommon to have 250 grad students and 1-2 safety officers to watch over all of them.

    To be fair, OSHA regs and recommendations place the responsibility on laboratory supervisors and personnel for these very reasons. Hazards vary substantially between labs, and it's quite unreasonable to expect a departmental/institutional training to cover them all (not to mention by the time you got through a fraction of it, most people would be asleep...).

    At my institution, our health & safety department provides templates for chemical hygiene and fire safety plans, which are to be customized by lab-appointed chemical hygiene and fire safety officers. Guess who got nominated in mine?

    I'm building our MSDS collection & SOPs from scratch. Essentially as chemical hygiene officer, I'm supposed to identify hazards and develop SOPs. It's a substantial amount of work and goes rather slowly when one's trying to run experiments, write fellowship proposals, and read papers along the way. But our safety office has also been very good about addressing questions and providing additional information when requested.

    I think part of the culture in academia is to view regulations of any sort (safety, animal research, human subjects...) as burdensome and time-wasting. Some of it may seem ridiculous on the surface, but we should realize that it's intended to improve our environment not weigh us down.

  13. When a lab in my PhD department had a pentane still blow up they sent someone around afterward to check on all the labs and their various levels of vigilance towards safety regs. I remember being admonished for not writing an 'open date' on a large canister of petroleum ether. So while I was happy to have someone come through, as it gave me the leverage I needed with the boss to have a lot of old reagents thrown out, to echo See Arr Oh's point I'm not terribly sure that there would be much positive influence having an extra one or two of chemistry-clueless safety people running around a university laboratory.

    I think a much more effective change might be to introduce a mandatory, semester long graduate synthesis lab class that would tackle the nuances of transferring pyrophoric materials among a number of other advanced techniques.

  14. @Anon 9:25 - I'd gladly take that course, even now!
    @biochembelle - It's good to see your Safety Dept. is so responsive and forward-looking. That's much better for you guys in the long run, and worth your few hours to start it all up.

  15. If he wanted to do chemistry with such dangerous chemicals, he should had hired a postdoc. A person professionally trained to do such jobs, an expert who has used before such chemicals and has a proven record, a person who has deeply understood the risks that he is undertaking with doing such job. Of course this person has to be paid real money and he is not free. You pay for his experience and expertise. Wouldn't you hire a fluorine chemistry expert to do work with fluorine gas? Some chemicals are very serious business, and one of them is t-BuLi. This was totally irresponsible in his part.

  16. @Anon9:53 - Way to re-post your exact comment from Chemistry Blog, grammatical errors and all. (sigh)

  17. See Arr Oh: Yes I re-posted. Is there a rule that says I can't do it? As for the grammatical errors, yes I am not a native speaker and writer of English. Is this a crime?

  18. @Arr Oh-Enough with your racism about the grammatical errors. In my academic lab, 80% of the people are from foreign countries and are not masters of the English language. Does this mean that they should not be allowed to express an opinion? Again, stop being a racist.

  19. A9:53: Your comments are welcome here; your accusations against other commenters are not.

  20. OK I am sorry. Please erase my last comment. Respectfully.

  21. SuppInfo of her JACS has gram-scale preparation of aryl grignards. Not t-BuLi, but still shows previous non-fatal use of pyrophirics. It'll be interesting to see how much tBuLi appears in her notebook...

  22. I apologize for calling a racist somebody that I don't know. This was a mistake. Chemjobber please delete my comment, this was not what I meant to write.

  23. @Anon10:07 - First, allow me to apologize, it was a petty accusation. However, levying a charge of racist thought or behavior is unwarranted; I know plenty of native English speakers with poor grammar.

    But, since I have you on the line, I'll address your comment as well - Experience: a postdoc is not a master of all types of chemistry, and the same incident could have happened to him or her. Money: a postdoc makes less money than what Ms. Sangji was paid. Fluorine: few people use fluorine gas in an academic setting, so it's an inept metaphor.

    Finally, the culture of academic safety and PI responsibility are our foci here, not the dangers of any specific chemical, although I'll grant you that tBuLi is nasty stuff. Each day, synthetic chemists work with NaOH, KH, DCM, and thousands of other potentially toxic or lethal compounds. The issue is "where the buck stops" with ensuring safe use.

  24. A10:30a: Do you think ArMgBr is equivalent to tBuLi in pyrophoricity? (Is that a word?) I don't think so.

    Her lab notebook entries are here:

  25. @CJ, @Anon10:30 - I've heard several anecdotal reports that 10M nBuLi is even tougher to handle than tBuLi. I, personally, was always most afraid of Me3Al.

  26. So who here's done the Et2Zn syringe flamethrower?

    [sheepishly, shamefully raises hand]

  27. @CJ - (This may not be a very apropos discussion line, considering the comment thread we're in!)
    We used to test "pyrophoric behavior" by setting Kimwipes in the back of the hood, and dripping 0.5 mL of solution onto them to see what happened.

    Good first impression about how reactive they'll be.

  28. You know, I did that in grad school, too.

    But what's funny is that, according to AkzoNobel, who handles these things on tugboat scale, that's a standard test of pyrophoricity. (See page 3 for the "paper char test".)

  29. Scale-up + nasty reagents + enthusiasm + work done in hurry + poor technique is how typically lab accidents happen. I think a limited previous research experience is more dangerous than having no experience: it lends false self-confidence. (I had many close calls in the lab, often working alone)

    In retrospect, someone should have looked over Ms. Sangji and warn her that giant syringes are terrible for air-free work and that a much better choice is using a graduated addition funnel with a septa and a canula transfer under positive pressure.

  30. Couldn't some out-of-work chemist make a mint by selling S.O.P.'s for common or not so common chemical reactions reactions. If they did their marketing right, they could sell them to individual groups. 1 industrially-vetted SOP could go to 10 groups in a single department. The industrially-vetted part is important for keeping the board of trustees happy. I seriously think that there is enough scare out there right now to keep a couple people gainfully employed for the next year or two.

  31. It is a fact that in academia there is a significant discrepancy between the safety culture that is and the safety culture that should be, especially compared to industry. And unfortunately, this case will be made an example of.

    However, Sangji's death arises from two main issues as far as I can see:
    1. Not wearing a lab coat

    2. (probably most importantly)Improper use of a sure-seal type bottle.

    I want to address #2. Regardless, of the chemical inside a sure-seal bottle, the protocol is virtually the same. Whether it be a bottle of anhydrous ethyl acetate or a bottle of tBuLi, positive argon pressure is required. By all accounts, Sangji did not observe this requirement.

    So why does someone accept a paid research position (ca. 45K/yr) and not no how to properly use a sure-seal bottle? Accepting that position is an unspoken contract between employee and employer that you have some basic technical skills. If she was uncertain of how to use a sure-seal bottle she should have asked someone and not proceeded, that is her responsiblity.

    If I took a job at a wharehouse and one day the forman asked me to use the forklift and I said, "Sure," would he be responsible when I crashed it because I don't no how to drive thing. Wouldn't I have a responsiblity of saying, "Oh, hey I don't know how to use that."

    If I went and bought a circular saw from Home Depot and then cut my hand off because I had never used one before and KNOWINGLY was not sure what I was doing, would everyone be screaming bloody murder that Home Depot should have trained me to use that saw. NO, everyone would say, "Why were you using the saw if you had no idea what you were doing.

    Again, regardless of the chemical in the bottle, she apparently didn't know how to use a sure-seal bottle. If she was unsure of how to use that equipment she needed to ask. That is what an adult with a job does.

    Additionally, it was her responsibilty to observe PPE regualtions and wear a labcoat. Again, that responsiblity was shrugged.

    The point is:
    What happened was tragic and was sad and I feel for her family and her loss. But posting picutures of her in a cap and gown with a lay around her neck and writing ad naseum of how she was a good student and planning to go to law school will not change the simple reality that everyone should realize.

    She made a huge mistake. And she died.

  32. Not equivalent, but still shows experience in handling pyrophorics 2 years before the incident. Notebook shows the same rxn on scale 2 months prior.

    I did my undergrad in a zinc lab, and learned proper precautions pretty damn fast. Nobody worked alone, and there were plenty of (controllable) fires.

    I find it hard to believe that an tech with at least two years' experience was inadequately prepared to run this reaction for at least the second time.

    Simplest explanation for me is still that the DA is engaged in publicity-whoring.

  33. The real issue is how do you generate a safe work environment where the labour force is heavily casualized, poorly paid, under pressure to work long hours and get results with minimal expenditure on correct equipment and additional staff but literally performing life threatening procedures everyday?

    The academic system is broken,.it delivers huge amounts of funding to a few celebrity researchers who are free to throw laboratory staff under the bus as they see fit. Safety is just one aspect that suffers in this system.

  34. I feel deeply uncomfortable about the PI being pursued with criminal charges for Sangvi's death, but, as the previous poster said, in the long term it could go some way towards redressing the imbalance of power between academics and their exploited grad students and post-docs.

  35. There's a real "useful bad idea" ring to threatening a PI with jail time. The precedent set, though...

  36. Anon @12:13pm

    "experience in handling pyrophorics"? Glassware catalogs are chock-full of specialized pieces designed for transfers just like the one she was performing - graduated schlenk tubes, addition funnels, yet she opted for a syringe, apparently one without even a luer lock. That hardly speaks of experience, or common sense. And comparison to aryl Grignards is flawed - unless you mix them with chlorinated solvents they are pretty tame.

  37. CJ, I don't think it's such a bad idea. Something's got to change, and if it takes slapping handcuffs on a tenured faculty member to make PI's realize that they are personally responsible for the safety of their students, then so be it.

  38. 2:17,

    If you look at the lab notebook pages posted by CJ, you'll see that she ran the t-BuLi rxn on large scale at least once without killing herself. This means that at a bare minimum, she either successfully quenched that tiny bit of excess t-BuLi left in the needle, or saw what if was capable of. There's a big stretch from lack of common sense or experience on her part to what the criminal charges claim.

  39. @2:35
    That she was able to do it once is pure luck - pulling that much t-BuLi into a syringe _is_ an accident waiting to happen. Another thing is - I was originally trained in an organometallic lab and about the first thing they teach you is to never use acetone bath with lithium reagents.

  40. 2:46,

    Really? I always used dry ice/acetone. For n-BuLi, I'd quench the excess in the bath (t-BuLi would get diluted in dry hexane), and I'd routinely use 20-mL plastics. Sash down as far as possible, coat/gloves on, etc. Sometimes the syringe would start to get warm near the plunger, which was a little disconcerting. I knew what I was doing (having previously spent 3 years in an organometallic lab with plenty of Na stills), and the PI rarely appeared in lab. Anything happened, I was fully aware whose fault it was.

    There's an assumption of risk every time you set foot in lab, and primary responsibility for not getting yourself killed begins with the guy running the rxns.

  41. It's tempting to generalize but the specifics of this case are in my view what is driving the D.A. The lab was cited for multiple very serious safety violations and given a hard deadline by which to remedy the situation; the deadline came and went without the lab making any serious effort to comply with the order; the lab was therefore cited again; and months still after that the technician dies directly because of the very safety violations the lab, twice notified, allowed to persist.

    It doesn't mean anyone is a villain. They were moving soon anyway, everyone knows what they're doing, nothing horrible has happened yet and so probably never will... all true, and everyone in all walks of life makes judgment calls like that every day that most often play out just fine. Still, they were warned, and warned again, and did nothing. So when this one didn't play out just fine and the technician turned out to have an unusually tenacious and effective family to boot, it's hard to argue that anyone is being railroaded. Dr. Harran and UCLA took a gamble and lost. Something very like it could have happened to most of us, but this happened to them.

    Also her, by the way.

  42. @3:21

    I, too, use plastics with Li reagents...but these plastics we are using have a luer lock... Unfortunately, those giant 60 mL syringes 1) are not luer lock, and 2) effing difficult to control the plunger. I HATE using them. Cannula is the way to go if you are going to transfer that much of a reagent.

    And as someone previously mentioned, she may have been in a hurry. If she has done the reaction previously, perhaps she was more comfortable... After 7 years in an organometallic lab, I am still scared sh!tless with pyrophorics, and that is ALL that I use.

    I'm wondering if she used that dumb short needle cause there were no clean long needles?! And with THAT short of a needle, she HAD to be tilting the bottle with that stupid syringe and uncontrollable plunger, minus positive inertness, a vacuum is pulled in the bottle.... YIKES.

    Such a tragedy. As a mentor and lab manager, each person who enters my lab needs to understand what he or she can and cannot do.

    This is our community, and we should all work to keep it safe...

  43. If I've got this right, she was using a plastic 60ml syringe fitted with a non Luer-Lok short needle for tert-BuLi, which she was pulling out of a sure-seal bottle without an argon inlet. Just thinking about this scene makes me sweat with nerves! In other words, she was doing EVERYTHING wrong! This level of ineptness would never have been permitted in properly-run lab.

  44. She should not have been using a plastic 60mL syringe to begin with. She should have known to use a glass one or, preferably, use a cannula.

  45. They do make 60 mL syringes with Luer locks. You should have a nitrogen or argon ballast connected when withdrawing 60 mL from a 100 mL bottle, but you need to watch the gas pressure doesn't push the plunger out the back end. I worked with this stuff as an undergraduate, grad student, and postdoc. I never had a lab coat, and I generally didn't wear gloves. Nobody in the world wears a fireproof lab coat to transfer butyllithium or tert-butyllithium or trimethylaluminum or anything. The claim that one should have been provided is ridiculous. I'm sorry for the girl and her family, but the existing rate of accidents in academic labs is entirely acceptable, and one outlying incident is not a good reason to change the world. There will never be any such thing as zero accidents.

  46. I know I'm a little late to the party, but I wanted to get my two cents in.

    "Do it the way the Navy does it. Train and drill a lot."

    This. As a lifeguard of many years prior to my foray into chemistry, I can say that drills are the number one way to keep your skills sharp. We had drills every Friday morning 1 hr before opening the beach as well as random mornings on slow days. The idea is that you never know when an accident or incident will happen and you better be prepared if and when it does. These drills were often led by the head lifeguard and sometimes by the beach manager. Because staying sharp was instilled by the higher ups, junior guards maintained sharpness through supporting and challenging other peers. Although the causes are different, in lifeguarding and in chemistry, lives are at stake when accidents happen.
    During my tenure in grad school after my years lifeguarding, I was utterly surprised at the lack of training offered to incoming students by my advisor and peers’ advisors. This is an advisor issue and nothing more. Clearly, University officials ‘get it’. EHS and OSHA safety officers ‘get it’. Grad students understand the risks when working with hazardous chemicals, however the breakdown occurs at the advisor when it comes to proper training. Some advisors ‘get it’ but not all and therein lies the problem.

    There is personal responsibility when an accident occurs. Failure can occur either because safety wasn’t adhered to in the first place or it truly was an unavoidable accident but the proper protocols in dealing with the accident weren’t followed. At the beach, in either case, the responsible lifeguard takes blame, however this indicates a fault with the training of that guard. “Why didn’t they know how to deal with the problem? They have certifications, no?” Lifeguard certifications are pieces of paper, as are BS degrees, as Rychnovsky claims she therefore must have known what was she was doing. Certifications and BS degrees don’t guard lives and work with chemicals. People do. Blame lies in head guards and beach managers to make sure their junior guards carry out proper safety protocols prior to, during, and after an accident.
    I apologize for the novel, but I'd like to leave you with an honest, non-rhetorical question: how are incoming grad students supposed to learn proper technique? Magic? Other peers who they themselves don't know proper technique?

  47. The above poster has a good point - when I was a grad student, I found it bizarre that all of the vacuum pumps were carefully vented to the building exhaust system in a lab where working without goggles and dumping waste down the sink were acceptable practices. Looking back, it's because the university safety people "got it" and the breakdown was with my P.I.

  48. An undergraduate with under one year's post graduate lab experience is not an experienced chemist. A Post doc is an experienced chemist. A bench chemist with ten years work experience is an experienced chemist. A first year graduate student is not. Nor was this poor young woman. Inexperienced but enthusiastic young workers often don't even know WHAT they don't know, which is why they are supposed to be put under the guidance and watchful eye of someone who does. Which is also why inexperienced young RA's and students aren't paid the same as the PI, aren't given the same prestige as the PI, aren't given the decision making authority of the PI, and aren't given the responsibilities of the PI, and most certainly shouldn't take the lion's share of the blame when things go wrong. I find it appalling that anyone would blame this young woman for her own death. As a new tech at an educational institute, not a for-profit Fortune 500 company, she reasonably might have expected to learn and be trained in techniques that she had not done before, rather than being expected to hit the floor running and make money for the company.


looks like Blogger doesn't work with anonymous comments from Chrome browsers at the moment - works in Microsoft Edge, or from Chrome with a Blogger account - sorry! CJ 3/21/20