Analyzing Harran’s defense in the #SheriSangji case

As most blog readers have undoubtedly heard, at the July 27 court hearing for the University of California and UC Los Angeles chemistry professor Patrick Harran, the district attorney dropped the charges against UC in exchange for UC acknowledging responsibility for the laboratory conditions that led to Sheharbano (Sheri) Sangji’s death, following an explicit lab safety program, and establishing a memorial scholarship in Sangji’s name.

Harran’s attorneys, meanwhile, are trying to get the charges against him dropped by attacking the credibility of California Department of Occupational Safety & Health (Cal/OSHA) investigator Brian Baudendistel. A defense motion filed on July 26 alleges that Baudendistel was involved a 1985 murder. Baudendistel would have been 16 at the time of the crime. In a follow-up story on C&EN Online this week, Michael Torrice reported on California juvenile records law and how likely it is that the allegations will derail the case, if the person convicted was indeed the same Baudendistel.

The defense motion also claims that a 2009 Cal/OSHA report written by Baudendistel “mischaracterizes or ignores outright testimony and other evidence…that tends to prove that Professor Harran did not violate any health or safety regulation, much less that he did so ‘willfully.'” After the hearing on the 27th, Harran’s attorney handed out not just the motion but many pages of supporting material. Included in that package were selected transcript pages (pdfs) of the interviews conducted by Baudendistel to prepare his report. (The report and its supporting material are exempt from public records laws in California because they are part of pending litigation.)

Below, I explore how well that supporting material bolsters the motion’s assertions. Anything in bold is a point from the defense motion. Typos were likely copied verbatim from source material (e.g. “Baudendistal,” “alco-lithiums,” “in side”), but it is entirely possible that I introduced a few of my own. For a refresher on key factors in the incident, here’s a blog post summary and an extensive story.

Investigator Baudendistel specifically affirmatively declares that he does not understand chemical scientific literature. (O’Leary interview, page 14)
The motion references discussion between Baudendistel and Sangji’s undergraduate research adviser, Pomona College chemistry professor Daniel O’Leary, about Sangji’s two papers published in Organic Letters and the Journal of the American Chemical Society. Safety Zone readers, what say you? Do you think it is necessary for a Cal/OSHA investigator to understand that type of literature to understand the incident that injured Sangji?

The very technical bulletin upon which Investigator Baudendistel relies [to assess Sangji’s experimental technique] specifically states: “Some chemists still believe that very specialized equipment and complicated techniques are required for handling air-sensitive reagents. This is often not the case.” (Sigma Aldrich Technical Bulletin AL-134, version from March 1997)
The techniques for handling air-sensitive reagents probably are reasonably straightforward for many chemists. That said, the bulletin recommends several things that Sangji didn’t do, in particular use a 1 ft- to 2 ft-long needle on her syringe or, even better, cannulate volumes greater than 50 mL. And, of course, just because a technique may be straightforward, that doesn’t mean that using it with a particular reagent isn’t inherently dangerous.

Graduate Student Researcher Andrew Roberts, who accompanied UCLA EH&S during its inspection of Professor Harran’s lab and had received UCLA’s laboratory safety training, testified that (Roberts interview, page 127):
– Lab coats were available in a stock room in Young Hall at UCLA.
– There were lab coats available to “anyone working in Professor Harran’s lab[.]”

But Roberts’s transcript also contains these statements, which directly relate to the charge for failing to require that employees wear work-appropriate clothing and personal protective equipment (California Code of Regulations, Title 8, Section 3383(b)):

Baudendistal: Okay. During the training that you received in September, 2008, was that issued discussed? Regarding the use of a lab coat.
Roberts: It was – it was discussed, but it wasn’t discussed whether or not it was required or not.

Baudendistal: Okay. Did he give you, did Dr. Harran give you, any specific instructions relative to the use of the lab coast.
Roberts: Uhm – no. Not that I remember.
Baudendistal: Okay. Where there occasions when you were working in the lab that you didn’t wear a lab coat?
Roberts: Yes.
Baudendistal: How often did you not wear a lab coat?
Roberts: Uhm – pretty much every day I didn’t wear a lab coat.
Baudendistal: Was that – was that a common practice, within the lab you were working in?
Roberts: Uh – yes.

UCLA Chemical Safety Officer Wheatley testified that (Wheatley interview, page 219 and 221):
– Lab coats were available in Professor Harran’s lab during his October 2008 inspection.
– At the time of the accident, the UCLA laboratory safety manual “suggest[ed] wearing a lab coat in the lab.”

See comment on previous point. Also from the Wheatley interview:

Baudendistal: But you, you estimate that about 20% of the personnel inside the [UCLA labs in general]…
Wheatley: Yeah, that’s just roughly…
Baudendistal: …not wearing…
Wheatley: Yeah.
Baudendistal: …a lab coat, correct?
Wheatley: Yeah.

Baudendistal: Do you know if there was any direct policy by UCLA that lab coats were worn while personnel are in the lab?
Wheatley: No.
Baudendistal: At that time, anyway.
Wheatley: No. Other than what it says in the lab safety manual, how it suggests wearing a lab coat in the lab.
Baudendistal: So is it accurate to say that there was no, there was no rule that a lab coat be worn in side the lab?
Wheatley: Yeah, I don’t there there’s any rule, like major policy by the university.
Baudendistal: Was it your understanding that it was a discretionary matter?
Wheatley: Yeah.
Baudendistal: And discretionary between who?
Wheatley: Discretionary between whoever’s working in the lab. It could be the PI, who makes it discretionary for the whole lab group, or the workers themselves.
Baudendistal: But it was not, there was no rule that lab coats were required to be worn while in the lab.
Wheatley: No.

Professor Daniel O’Leary, Ms. Sangji’s chemistry thesis advisor from Pomona College, stated that:
– Ms. Sangji received laboratory safety training at Pomona College at least three times, including a video regarding fires (although “not necessarily” a laboratory fire). (O’Leary interview, page 3)

From the transcript:

O’Leary: Every summer there’s like a safety orientation meeting that the students were obligated to participate in… Well, it – my memory’s pretty fuzzy. I know that he had, he had some, some sections in there on, actually on fires. Because I know he showed a video of a, you know, how fast a fire can accelerate and get out of control. It wasn’t necessarily a lab fire. It was I think it was something that had been staged in a dorm room or something like that. He talked about kind of knowing your, you know, chemists work in hoods, right? In fume hoods with pull-down sashes. And he talked, he talked a lot about making sure that the indicators, the air flow indicators were, were properly working. And not to have your hood too crowded to obstruct air flow. He also talked about anger management issues in the workplace. So, it was kind of a broad spectrum thing. It never was, you know, specific to pyrophorics or, or any particular class of compound.

– Ms. Sangji’s undergraduate chemistry work would have exposed her to flammable solvents. (O’Leary interview, page 4-5)
But never pyrophoric reagents: “She wouldn’t have had any, any experience in my lab working with something as, as nasty as t-butyl lithium,” O’Leary says. The available evidence indicates that Sangji handled tert-butyllithium only once before the day of the fire.

– Typically, when people are working with alco-lithiums, “they’re either going to be working with a syringe or…a cannula transfer…” (O’Leary interview, page 6)
– There is a “mix of opinion” in the chemistry community regarding whether to use a glass or polypropylene syringe to transfer alco-lithiums. (O’Leary interview, page 6)

Correct, both in terms of what O’Leary says and in what I’ve heard in the three years I’ve been reporting on the incident. But O’Leary also notes that he would have used a syringe for transfers around 10 mL and a cannula for more than 40 mL. He also discusses why polypropylene syringes can be difficult to use, because some organic solvents can make them swell so they don’t move smoothly.

– The method of transferring alco-lithiums depends on the type of research being done, and that “[u]sually the traditions [of using a syringe or cannula method] just get carried along.” (O’Leary interview, page 11) Incredibly, the BOI Report omits this statement from Professor O’Leary, yet excoriates Professor Harran for making nearly the same precise statement.
I read O’Leary’s statement differently. In context, it seems that he was focusing on groups that use glove boxes versus groups that do reactions in the hood:

O’Leary: And it’s really a matter of – I don’t want to use the word “style” loosely here but it’s really a matter of how the group typically has always been doing it, you know. Some groups are equipped with glove boxes and they might prefer to have their alkyl lithium reactions run in a glove box. Other groups may just traditionally always do it with syringe or with cannula transfers. Usually the traditions just get carried along.

And here is the relevant passage on Harran from the BOI report (quote is from page 15; the relevant interview excerpt is on report page 75):

However, Dr. Harran later admitted that the Aldridge AL 134 bulletin was utilized as a “general reference” only and that training relative to the handling of t-Butyllithium was based on “knowledge” passed down from one generation of researcher to another.

Dr. Harran confirmed that he did not review the procedures outlined in the AL-134 Bulletin with Victim Sangji, nor did he inquire whether she [Sangji] was aware of the procedures outlined in Technical Bulleting 134. Dr. Harran also admitted that he never discussed with Victim Sangji the risks associated with the use of t-butyllithium.

This is likely in part the basis for the charge for lack of training, in addition to the fact that Sangji received no general lab safety training at UCLA. Employers are legally required to ensure that employees are trained on “the measures employees can take to protect themselves from these hazards, including specific procedures the employer has implemented to protect employees from exposure to hazardous chemicals, such as appropriate work practices, emergency procedures, and personal protective equipment to be used” (California Code of Regulations, Title 8, Section 5191(f)(4)). It’s not hard to see how proper training can easily go downhill with oral knowledge transfer and no quality control, and that’s likely at least part of the reason the UC court agreement mandates written standard operating procedures.

– People reading Ms. Sangji’s publications may conclude that she “had a massive amount of synthetic [chemistry] experience” (O’Leary interview, page 14)
– Ms. Sangji had “a pretty significant amount of experience in kind of traditional synthetic organic [chemical] manipulation” (O’Leary interview, page 15)

I can’t tell whether the defense is using the first quote to present Sangji as experienced or to imply that Harran was overly influenced by her resume. If the point is to say that Sangji was an experienced synthetic chemist, the quote is taken entirely out of context:

O’Leary: In fact both of these papers were accounts of analytical chemistry that she did in the lab. So really she had no – her contributions to these papers was purely taking measurements of compounds, not doing any synthesis. And they’re collaborative papers. Like the first paper, the 2006 paper, we wrote with a group at Boston University. And the 2005 paper was written in collaboration with a group over at Irvine. Now, both of those groups are kind of synthetic groups. So some people might be looking at her papers and they might be concluding that she had a massive amount of synthetic experience that went into those papers. But her contributions again were purely from a measurement standpoint.

The second quote is accurate. And O’Leary says this earlier in the interview (page 2):

O’Leary: For the first summer, she did mainly analytical chemistry, which means she really didn’t do any traditional, organic chemistry. She was really taking measurements on compounds. And um that work actually is what constitutes her two published papers that a lot of people point to. … And the next two summers, she did mainly peptide synthesis work … That entails kind of traditional, synthetic, organic manipulations – running reactions, working up reactions, isolating products, purifying products, more analytical chemistry to kind of prove the identify of the products.

But, again, none of this involved pyrophoric reagents.

Hui Ding, a post-doctoral researcher in Professor Harran’s temporary lab, stated that Professor Hurley was “oftentimes” working with Ms. Sangji, because “she’s working on some compounds that Professor Hurley is working on.” (Ding interview, page 344)
– Indeed, the BOI report misstates Dr. Ding’s testimony, as the Report states that “Dr. Ding stated that Paul Hurley would occasionally work with Victim Sangji, as they were working on related research. (BOI report, page 53)
– Given Professor Harran’s statements that Dr. Hurley was responsible for training Ms. Sangji (BOI report, page 53), contrasted against that the Report’s conclusion that “it is clear that Victim Sangji was not properly trained, if at all…”, the Report’s failure to accurately state Dr. Ding’s testimony is nothing short of incredulous.

The first point is correct, at least as far as the cited page is concerned. I don’t have the rest of the transcript. As for the second point, in the two transcript pages provided, Ding really does not give much information about Hurley’s interactions with Sangji or Sangji’s training. The defense did not choose to include anything from Paul Hurley’s transcript, which would probably have better information. Also, the motion itself fails to accurately state Hurley’s title–as far as I’m aware, he was not and is not “Professor.”

William Peck, former manager of UCLA EH&S, stated that t-BuLi was a “pretty common” compound in the organic chemistry section. (Peck interview, page 403)

Wei Feng Chen, a post-doctoral researcher in Professor Harran’s temporary lab, stated that there were lab coats available to personnel in storage on the third floor and there were funds to obtain lab coats as well. (Chen interview, page 310)
But see earlier points about whether people thought lab coats were optional or required.

Author: Jyllian Kemsley

Share This Post On


  1. Nice work, Jyllian — this is really great.

  2. I really hope Harran spends a few months in jail. This is not really about him, but rather because it would scare the bejeezus out of every other chemistry PI out there and cause a lot of them to act far more responsibly than Harran did.

    It is obvious that Sangji did not know how to properly transfer large quantities of pyrophoric reagents, had not received even remotely proper training, did not have proper equipment, and did not have and was not required to use proper PPE. If I were to do the same reaction in my corporate lab, there would be multiple rounds of safety reviews long before I was allowed to move around 50 ml of t-butyl Li, and I am many times more experienced than Sangji was. And of course, proper PPE is mandatory just to set foot in our labs, let alone work with dangerous reagents.

    Harran was deeply irresponsible and deserves to pay a price. Just because far too many other chemistry profs do the same thing is not an excuse.

  3. @Chad: really, a few months in jail? For what, not promoting a certain safety culture? That’s not criminal. I do think that PIs should institute a safety culture in their lab (maybe I’m lucky that my PI has). In the end, accidents happen when users are careless. It was tragic what happened here but ultimately safety lies with the primary user.

  4. I’m really sick of seeing the old Aldrich bulletin on the use of alkyllithiums and the like quoted as the authoritative text on the safe handling of such materials. I don’t think Aldrich ever intended it to be used for that purpose. It reads like it’s purpose is to protect the reagent, not the user. I’ve worked with these materials since the early ’80’s and while I like the Aldrich Sure-seal bottles and the like, I always threw that bulletin in the trash and instructed those reporting to me myself.

  5. @Ken–what do you do differently from the Aldrich bulletin and why?

  6. Jyllian – too much to go into in a blog like this, but here are a few comments. First, if you haven’t done so, take a look at the current Aldrich technical bulletin: It is clearly not a written from a personnel safety perspective. Search it for the word safety and you will find it appears only 3 times and never in the context of personnel safety. Likewise, there is no mention of gloves, lab coats, training, or performing work in a hood. The word fire appears only once in the entire document.
    Other particulars: The glass on glass Micro-Mate syringes shown on page 6 are woefully inadequate. They leak and decomposing reagent very frequently causes the barrel/plunger to freeze. The disposable polypro syringes shown on the same page are the only ones that should be used.
    No mention is made of the hazards of syringing liquids, especially pyrophoric liquids, when they are under positive nitrogen pressure and/or in volatile solvents (tert-butyllithium is still sold in pentane). As soon as a filled syringe + needle under such positive pressure is removed from a septum it will spew some material out of the needle as the pressure vents. Unless the needle is cleared prior to removal, and the syringe is held properly, a substantial quantity of material can and will shoot out the end of the syringe.
    The more pyrophoric it is the best it is not to syringe it at all but to use a cannula. Experiments with something like tBuLi should use the full contents of small bottles. Tare them before cannulating, then reweigh. Then adjust the amounts of other reagents accordingly. Aldrich is selling tBuLi in 25 mL bottles now, so this should be a viable option in many cases.
    Best practice, if at all possible – measure out the material in a dry box.

  7. Very good investigative report

    But really…All this over tBuLi? Mistakes happen. To even bother talking about the length of needle attached to a syringe for transfer shows the prosecution has no idea about chemistry.

    from a long time user of tBuLi, that uses easily 70 mLs at a time. Measured in plastic syringes.

    Like cdm says “It was tragic what happened here but ultimately safety lies with the primary user.”

  8. asdassdsa the attitude you have just exhibited cuts to the root of the problem that ultimately resulted in this death. Chemists become inured to the hazards which with they work because they A) are familiar with the reagent and B) have not had an accident themselves. This ultimately erodes the safety culture because proper respect is no longer given because “nothing bad has happened”. If proper safety culture had been present in this lab lab coats would have been worn, and she would not have worked with a pyrophoric next to a flammable solvent bottle.

    There is no doubt in my mind that Harran is guilty of negligence in how he ran his lab. The only question is was it criminal. That is up to the court.

  9. From the U.S. Chemical Safety Board Mission page: The CSB conducts root cause investigations of chemical accidents at fixed industrial facilities. Root causes are usually deficiencies in safety management systems, but can be any factor that would have prevented the accident if that factor had not occurred. Other accident causes often involve equipment failures, human errors, unforeseen chemical reactions or other hazards.

    At UCLA you had it all. A chemist was not wearing proper PPE and using an unsafe procedure to transfer t-butyllithium. Chemistry faculty who don’t ensure their subordinates have proper training or demand safe working practices. An EHS unit that was ineffective (It is suggested lab coats be worn). Regents who did nothing until disaster struck. The sad thing is that this was not uncharted territory. Private and government laboratories have had strong safety programs for many years. UCLA could have had one, too.

    Jillian, I agree with Chemjobber. This is a great post.

  10. In the May 8, 2012 “The Safety Zone”, Jyllian linked to safety videos by UCSD’s Dr. Haim Weizman. Faculty should watch “A Day in the Lab” to get an idea how they should mentor. There are also three videos on handling pyrophoric reagents.

  11. As someone who has instructed students to use pyrophorics, and written a SOP for pyrophoric liquids – I totally agree with Ken, the Aldrich bulletin is not the best authority, and it indeed focuses primarily on preserving the reagent rather than safety.

    We use plastic polypropylene syringes only, and a small positive pressure of argon from a line or from a balloon (less frequently). And we _never_ allow anyone to fill the syringe with the pressure from the line, nor do we allow transfers with syringes over 10 mL.

    There are so many ways a syringe could leak and spill the contents – pulling the plunger too far is just one of them – and fortunately this is something that can be very easily avoided by never filling the syringe too full. However, a more frequent source of leak is between the needle and the syringe. This could happen even when a Luer lock is used if the lock is faulty, improperly closed, or the needle is blocked. What we recommend is cooling the needle in a desiccator (if it has been dried in the oven) _before_ placing it to the syringe, and using only fresh syringes for highly pyrophoric compounds like t-BuLi, Et2Zn, and Me3Al. The connection between the syringe and the needle must be tightened prior to use_and_ the person doing the transfer must also hold on to the base of the needle while syringing out the material.

    Additionally, for pyrophorics, a small piece of glass tubing with septa at both ends is very helpful for needle protection.

    We also recommend that a second person directly oversees the transfer, and this is mandatory for more than 2 mL transfers.


  1. Shaking off the Jet Lag « Chemical Space - [...] been about the Sheri Sangji case, but some sterling work by ChemBark, ChemJobber and, especially, Jyllian Kemsley (surely that…