LOS ANGELES — Sheri Sangji is on fire.
The 23-year-old research associate, a Pomona College graduate raised in Pakistan, has accidentally pulled the plunger out of a syringe while conducting an experiment in the Molecular Sciences Building at UCLA. The syringe contains a solution that combusts upon contact with air.
The solution spills onto Sangji’s hands and torso, and she is instantly aflame. She isn’t wearing a lab coat; no one told her she has to. Her synthetic rubber gloves provide no protection as the fire burns through her hands to the tendons. She inhales toxic, superheated gases given off by her burning polyester sweater, a process that accelerates as she runs and screams.
It’s December 29, 2008, mid-afternoon. The UCLA campus is mostly quiet for the holidays, but chemistry professor Patrick Harran’s team is working. Harran is in his office, one floor up from Room 4221, where at his direction Sheharbano “Sheri” Sangji has been trying to produce a chemical that holds promise as an appetite suppressant. She is unsupervised.
Two postdoctoral fellows from China are nearby when Sangji catches fire. One runs upstairs to summon Harran, the other tries to smother the fire with his lab coat. He doesn’t think to put Sangji under an emergency shower a few feet away. By now, deep burns cover almost half her body.
Harran finds Sangji “sitting on the floor,” her clothes “either caked to her or burned off,” he later tells an investigator.
After 18 days, on January 16, 2009, Sangji succumbs to her wounds at the Grossman Burn Center in Sherman Oaks, Calif.
Harran and the University of California’s Board of Regents will be prosecuted for the fire in Room 4221. Harran will be the first American university professor to be accused of a felony in connection with the death of a worker. Poor lab safety practices at UCLA will be brought to light, and researchers around the world will take notice.
“Sheri was a young girl who was working in a laboratory in one of the largest and most prestigious universities in the world,” says Sangji’s older sister, Naveen, a surgical resident in Boston. “There should be no safer place for someone to go to work. Instead, she never got to come back home.”
***
Sangji’s death and the prosecution of Harran and the UC regents have had far-reaching effects. Faculty members, department heads and deans at research institutions have followed the developments with consternation: Might they, too, be criminally liable if something happened in one of their labs? A federal investigation revealed that there had been at least 120 lab accidents at universities between 2001 and 2011.
On Friday, the criminal case against the regents was dropped after they agreed to adopt a lengthy list of safety measures and establish a $500,000 scholarship in Sangji's name. The case against Harran, who faces up to 4 1/2 years in jail, continues. His arraignment was postponed until September 5.
Naveen Sangji wants to see Harran behind bars. “If this were a regular person out on the street who got drunk and killed someone,” she says, “he would be going to jail.”
At the time of Sheri Sangji’s death, California’s Division of Occupational Safety and Health, Cal/OSHA, already had begun an inquiry into the accident at UCLA. That May the university was cited for four violations; it paid a $31,875 fine.
In December 2009, Cal/OSHA’s Bureau of Investigations, which looks into all worker fatalities in the state, recommended that Patrick Harran and UCLA be charged with involuntary manslaughter and felony labor code violations. “Dr. Harran,” investigator Brian Baudendistel concluded in a 95-page report, “permitted Victim Sangji to work in a manner that knowingly caused her to be exposed to a serious and foreseeable risk of serious injury or death.”
Harran, 42, did not respond to interview requests from the Center for Public Integrity and the Center for Investigative Reporting. In a 2009 statement to the Los Angeles Times, he called Sangji’s death a “tragic accident” and explained, “Sheri was an experienced chemist and published researcher who exuded confidence and had performed this experiment before in my lab. Sheri had previous experience handling pyrophorics, chemicals that burn upon exposure to air, even before she arrived at UCLA. … However, it seems evident, based on mistakes investigators tell us were made that day, I underestimated her understanding of the care necessary when working with such materials.”
UCLA officials declined interview requests, pointing to a written statement issued by university Chancellor Gene Block in January. “Sheri Sangji’s death was strongly felt by everyone at UCLA, and we were deeply saddened by the loss of a member of our community,” Block wrote. “I made a pledge then that we would go above and beyond existing policies and regulations to become a model of campus safety. And we have.”
Baudendistel referred the Harran case to the Los Angeles County district attorney’s office for prosecution, as is Cal/OSHA’s practice when it believes it has evidence of gross employer misconduct. The DA filed a felony complaint in December 2011, focusing on the labor code violations. Chemists and safety consultants were stunned.
“The district attorney got the attention of every research institution in the United States,” says Harry Elston, editor of the Journal of Chemical Health and Safety.
‘A scientist’s scientist’
Sheri Sangji was raised in Karachi, Pakistan, and graduated from Pomona College in California in May 2008. A superior student and athlete, she earned a degree in chemistry but had no plans to enter the field. “She was a very dynamic person with lots of interests, a lot of spark,” says her sister Naveen, 29, also a Pomona graduate. “She was interested in the environment, women’s rights, minorities’ rights.”
Sheri took a job with a pharmaceutical company in Pasadena, hoping to save money for law school. She was intrigued by an ad Harran placed for a research associate at UCLA. The idea of moving to Los Angeles and working for a “rising star” in organic chemistry appealed to her, Naveen says.
The job interview took place in September 2008. Harran was impressed. Sangji “was very familiar with analytical instrumentation of the type that I really wanted her to focus on, which was great,” he told investigator Baudendistel. “I asked her if she was comfortable with general techniques and properties of organic chemistry. And I asked her if she worked with air-sensitive materials …. Just how generally comfortable she was in the laboratory. That’s what we spent most of our time on, and she left. And, you know, I loved her. I thought she was fabulous.”
Sangji began work in the UCLA Molecular Sciences Building on October 13. Four days later, Harran watched her perform a small-scale experiment using tert-Butyllithium solution, a chemical its manufacturer, Sigma-Aldrich, describes as follows: “Reacts violently with water. Contact with water liberates extremely flammable gases. Spontaneously flammable in air. Causes burns.”
Sangji did a “great job” on the experiment, Harran told Baudendistel, and had knowledge of chemistry beyond her years. “She had published in top peer-reviewed journals with very well-known researchers. … She stood out.” Harran acknowledged, however, that Sangji did not receive “generalized safety training. I believe my assistant told me that it was not offered for her category per se, although we were going to follow up on that.” He also said that no fire-resistant clothing was available to lab employees at the time of the accident.
Harran had come to UCLA as a tenured professor the previous July, having been recruited from the University of Texas Southwestern Medical Center in Dallas, where he’d spent nearly 11 years and won a number of honors, including the AstraZeneca Excellence in Chemistry Award and the Pfizer Award for Creativity in Organic Synthesis.
“He was literally the first chemist we succeeded in hiring,” says Steven McKnight, chairman of the biochemistry department at UT Southwestern. “He was articulate and personable and easy to communicate with, unlike many of the candidates. Most of the equivalent scientists didn’t have the fearlessness or fortitude to go to a department that had no history in chemistry.”
Harran “built a very strong laboratory and proceeded to make a number of really nice discoveries in the field of synthetic chemistry,” McKnight says. Harran and a colleague, Xiaodong Wang, developed a chemical that causes cancer cells to kill themselves. They published a paper on the breakthrough in Science magazine.
A graduate of Skidmore College and Yale, Harran was “a scientist’s scientist,” McKnight says. “He really wanted to dig in and make discoveries of consequence. When he went to UCLA it was a heartbreaker.”
In a 2006 interview with the Proceedings of the National Academy of Sciences, Wang, who has since returned to his native China, offered a glimpse of life at UT Southwestern. “Nothing is ever good enough, and you’re only as good as your last paper, which I think is great,” Wang was quoted as saying.
A violent reaction
UCLA pursued Harran aggressively, offering him a budget of $3.2 million to set up a state-of-the-art organic chemistry lab on the fifth floor of the Molecular Sciences Building. He and his team were given temporary space on the fourth floor while renovations were made upstairs.
On October 30, 2008, UCLA chemical safety officer Michael Wheatley conducted an annual inspection of the fourth-floor labs. Wheatley found a number of deficiencies, one particularly relevant to events that would soon unfold: “Eye protection, nitrile [synthetic rubber] gloves and lab coats were not worn by laboratory personnel.”
In an email on November 5, Wheatley asked Harran when they could meet to discuss the findings. “Is it possible to wait until we get settled on the 5th floor?” Harran replied a week later. “That would make for a better meeting — our labs on 4 are overcrowded and disorganized. I wasn’t planning to be in temporary space for this long.” Wheatley agreed to the postponement.
On December 29, a Monday, Sheri Sangji reported for work in Room 4221. Harran wanted her to replicate the chemical reaction she’d performed on October 17, but on a scale three times larger. Around 3 p.m., Sangji was using a 60-millileter plastic syringe with a 2-inch needle to transfer tert-Butyllithium from a 100-ml bottle to a glass flask.
The needle was too short; Sigma-Aldrich recommends using one at least a foot long. This, investigator Baudendistel theorized, forced Sangji to tilt the bottle of tert-Butyllithium or lay it on its side, awkwardly withdrawing the liquid with one hand while holding the bottle with the other. Had the needle been long enough, she could have clamped the bottle, upright, to the workbench, a less risky procedure. Safer still would have been the “cannula transfer” method, in which a liquid is pushed by an inert gas like nitrogen from one container to another through a tube.
Sangji inadvertently pulled out the plunger of the syringe, spilling the solution and triggering a flash fire. Had she been wearing a fire-resistant lab coat, her burns might have been less severe. In fact, she was wearing no lab coat, not even a cotton one.
At the time there was no university policy requiring such protection. “That policy has been put in place since the accident,” Harran told Baudendistel. Requisition forms from the UCLA Department of Chemistry and Biochemistry show that fire-resistant lab coats were, in fact, ordered, at a cost of $45.05 each.
In an interview with a deputy UCLA fire marshal, Harran described what he saw in Room 4221 before the paramedics arrived.
“Sheri was, you know, she was in shock … she was shaking. I asked her what happened. She didn’t tell me much. She just said there was a fire, and she just kept asking, ‘Where are they, where are they, where are they?’ … She wanted water on her arms, and she was holding her hands out like this, and the skin was separating. It was awful.”
Sangji was taken to Ronald Reagan UCLA Medical Center. Harran finished the experiment she had started — at the request of the Los Angeles and UCLA fire departments, which feared another conflagration, he said. Shortly after 4 p.m. Pacific time, Naveen Sangji’s cellphone rang in Boston. Then a medical student at Harvard, she recognized Sheri’s number and assumed her sister was calling to tell her about another law school acceptance letter. They had been coming regularly.
It was a hospital social worker, using Sheri’s phone. “She told me Sheri had been in an accident and described what happened,” Naveen says. “As a medical student I could understand the gravity of what she was saying.” She caught a flight to Los Angeles early the next morning.
Naveen went straight from the airport to the Grossman Burn Center, to which Sheri had been transferred. “Her arms were suspended from the ceiling to keep them in a certain position, all wrapped with bandages,” Naveen says. “The only part of her that I could see was her face, which was unwrapped.”
Naveen encountered Harran at the burn center on New Year’s Eve. “He came to the hospital and spoke with me and my uncle, who is a structural engineer,” Naveen says. “[Harran] explained some of the details of the experiment Sheri was doing that day. We obviously asked him questions about why she was doing this experiment, this dangerous experiment, without supervision. My uncle, because of his engineering background, asked specifically about training and about why she wasn’t given fire-resistant protective equipment before doing this experiment. Harran refused to answer.”
Sheri’s friends began a vigil. “I had thought she would survive,” says Aakash Kishore, a lab assistant in the UCLA psychology department at the time and now a graduate student. “I remember reassuring Naveen.” Kishore and others showed up at the burn center almost every day during the 18 days Sheri was there. About two dozen of her friends and relatives were waiting in the parking lot the day she died. “Her dad came outside and let us know that she was gone,” Kishore says. “He looked very weak.”
Willful violations
In the months to follow, Naveen pressed UCLA officials for details on the accident. She found the responses wanting. The university, she felt, was trying to make it appear that Sheri was an experienced chemist, and that the fire was her fault. On June 17, 2009, replying to an email Naveen had sent two days earlier, Chancellor Block recalled “the elegant and successful way” Sheri had performed the tert-Butyllithium experiment the previous October.
Although Cal/OSHA had issued four citations to UCLA in May, Block wrote, “The campus believes … that many corrective measures ordered by our inspectors were taken before the tragic accident, though they were not properly documented.” Cal/OSHA, he noted, “found no willful violations of regulations or laws by UCLA personnel. Neither [chemistry department chair Al] Courey nor Dr. Harran were in the lab the day of the tragedy and did not have the opportunity to remind Sheri to put on her lab coat.”
In his interview with the deputy fire marshal, however, Harran — the lab’s principal investigator, or PI — admitted that his safety policies were less than rigid. Harran said he “never explicitly” told his senior employees, such as postdoctoral fellows, to make sure subordinates were wearing protective equipment. In the same interview, Harran said that he and the fellows erred by cleaning up potentially dangerous items in Room 4221 immediately after the accident, before investigators returned to gather evidence. “We shouldn’t have touched anything,” he said.
In November 2007 — 13 months before Sangji was hurt and eight months before Harran came to UCLA — a graduate chemistry student named Matthew Graf spilled a bottle of ethanol near an open flame; some of the alcohol splashed on his shirt and he caught fire. Graf wasn’t wearing a lab coat and sustained second-degree burns to his hands and torso. He spent a week at the Grossman Burn Center, the same place Sangji died, and underwent surgery to repair his hands. Cal/OSHA learned about the accident nearly two years after the fact and cited UCLA for failing to report it; the university is contesting the citation.
In Naveen Sangji’s view, the fine UCLA paid for her sister’s death was insufficient. She was relieved and gratified when Cal/OSHA’s Baudendistel issued his report in December 2009, recommending that Harran and UCLA be charged with felonies.
Baudendistel concluded that “the laboratory safety policies and practices utilized by UCLA prior to Victim Sangji’s death were so defective as to render the University’s required Chemical Hygiene Plan and Injury and Illness Prevention Program essentially non-existent.” There had been “a systemic breakdown of overall laboratory safety practices at UCLA,” he wrote.
Indeed, on Dec. 22, 2008, one week before Sangji was burned, another graduate chemistry student, Jonah Chung, was completing a reaction when “the reaction pot detonated, causing glass, hot oil, and chemicals to strike his face and torso,” Baudendistel wrote. Chung, who sustained burns to his torso, arms and face and cuts to his neck and forehead, “was not wearing a lab coat, gloves, nor appropriate eye protection … at the time of the incident.”
Baudendistel sent the Harran case to the Los Angeles County district attorney. This was not unprecedented: from 2001 through 2011, Cal/OSHA made 486 such referrals statewide, mostly in worker death cases; 174 resulted in criminal charges.
“You know, we have put owners of companies, supervisors, foremen in jail,” says Cal/OSHA chief Ellen Widess. “That is noticed. We’re definitely looking for these cases to make … an impression, leave nothing unspoken and unclear about the severity of the punishment that will be meted out.”
Still, Harran wasn’t a foreman on a trenching job or the owner of a roofing company. He was an award-winning chemistry professor with the backing of a powerful university.
It took two years. On Dec. 27, 2011, the DA filed a felony complaint against Harran and the UC regents. The allegation: “willful violation of an occupational safety and health standard causing the death of an employee.”
Hard questions
Chemists in academia and private industry already had been debating the Sangji case; bloggers and journal editors had written about it. The filing of the complaint took the discussion to another level.
Uncomfortable questions followed: Why were academic labs more dangerous than those in industry? Were some principal investigators so obsessed with publishing papers, securing grants and winning prizes that they’d lost sight of their responsibility to keep employees and students from being hurt?
“Each lab is like an island where the PI is king,” says Paul Bracher, a postdoctoral researcher in chemistry at Caltech who writes a blog called ChemBark. “He provides for the lab, brings in grants, decides how the money is spent. There are a lot of demands on their time, and the safety stuff a lot of times gets lost in the shuffle. I’ve never heard of anyone getting fired for being unsafe.”
Bracher — whose trachea was pierced by flying glass 12 years ago after a fellow undergraduate at New York University mishandled a reactive chemical — says he’s surprised UCLA has stood by Harran so steadfastly, given the evidence that’s come out. “When something like this happens, much like a drunk driver when someone loses their life, there should be consequences,” he says. “UCLA has doubled down. It sends an incredibly disconcerting message.”
“The PI has to be actively involved in safety, as does the president of the university, the provost, the dean — everyone who supervises other people,” says James Kaufman, a former Dow Chemical researcher who runs the nonprofit Laboratory Safety Institute, a training organization near Boston. “The dog sled can’t go any faster than the lead dog.”
Following the Sangji accident, and another at Texas Tech University that badly injured a graduate chemistry student in January 2010, the U.S. Chemical Safety Board began an investigation of lab safety at academic institutions. In a report last fall, the board, which can make recommendations but can’t regulate, said it had documented 120 incidents at university labs since 2001 and identified “safety gaps” that threatened more than 110,000 graduate students and postdoctoral researchers in the U.S.
“Fiefdoms” in academia were partly to blame, the board found.
“At some academic institutions, PIs may view laboratory inspections by an outside entity as infringing upon their academic freedom.” At Texas Tech, “some PIs saw the notification of safety violations to the [department] Chair as ‘building a case’ against them, felt that the safety inspections inhibited their research, and considered recommended safety changes outside their control because they could not ‘babysit’ their students.”
The board recommended that the U.S. Occupational Safety and Health Administration — whose 1990 lab standard emphasizes the need to protect researchers from carcinogens and other health hazards — make clear that physical hazards also must be controlled. And it urged Texas Tech to revamp its lab safety program by documenting and acting on near-misses that could portend more serious accidents.
Board officials believe the message is getting out — not only to universities but also to grant-makers such as the Department of Homeland Security, which funded the work on explosive materials that led to the Texas Tech accident. “DHS changed its requirements after this incident,” says board investigator Cheryl MacKenzie, demanding that grantees’ labs undergo independent safety audits before funds are released.
The American Chemical Society, a professional association for chemists, assembled a task force after the Texas Tech blast and recently unveiled a draft report that recommends ways to change the “safety culture” in academia. The study, its authors wrote, was prompted by “devastating incidents in academic laboratories and observations, by many, that university and college graduates do not have strong safety skills.”
UCLA, for its part, has created a Center for Laboratory Safety which, Chancellor Block said in his January statement, will “identify and institute best practices in safety, going beyond the minimum requirements of outside agencies so that we can hold our laboratories to even higher standards. We also dramatically increased the number of lab inspections, strengthened our policy on the required use of personal protective equipment and developed a hazard-assessment tool that labs must update annually or whenever conditions change.”
The real-world impacts of these changes remain to be seen. “I think the university is trying,” says Rita Kern, a staff research associate in the UCLA Department of Medicine who sits on the health and safety committee of University Professional & Technical Employees — Communications Workers of America Local 9119, the union to which Sheri Sangji belonged at the time of her death. “Some things have changed, but it’s like turning a big boat in the middle of the ocean. It doesn’t turn very fast.”
Indeed, following inspections in August 2009 and February 2010 — eight and 14 months, respectively, after the fire that killed Sangji — Cal/OSHA cited UCLA for 16 lab safety violations, five classified as “serious” and one as “repeat serious.” The university paid a $36,690 fine.
Ryan Marcheschi, a postdoctoral fellow in the UCLA chemical engineering department who works with flammable and explosive compounds, says the university has “tightened up” on safety since the Sangji accident, though much of this has come in the form of increased paperwork.
When he learned that the criminal complaint had been filed against Harran, “I thought it was extreme,” Marcheschi says. “But then I thought, maybe that’s what’s needed to make policies change.”
***
Despite her grueling schedule as a resident at Massachusetts General Hospital, Naveen Sangji remains an advocate for her younger sister, now 3 ½ years departed. “My sister had her whole life ahead of her,” Naveen says. “She would be graduating from law school right now.”
Her parents live in Toronto. Her father, a small businessman, “has almost completely stopped socializing,” Naveen says. Her mother, a Montessori teacher, was “completely destroyed” by the accident and immerses herself in her work. They go to the cemetery on Sundays.
“Sheri was a very brave person,” Naveen says, a trait that became evident at the burn center.
“Before my parents came into the room, she asked me to cover her up with the sheets so that they wouldn’t be distressed at seeing all the bandages; her concern was for them. When my dad arrived, he put his hand on hers lightly through the sheet and she screamed because it was so painful. And we couldn’t touch her anywhere except her face. But her thoughts were, even as she lay critically injured, for other people.”
This story was produced in collaboration with the Center for Investigative Reporting.