A year and 2 months have passed since the regents of the University of California (UC) system reached an unprecedented settlement with the district attorney of Los Angeles County in the criminal case against the university arising from the death of 23-year-old lab assistant Sheri Sangji from burns suffered in a University of California, Los Angeles (UCLA) chemistry laboratory. Safety experts predicted at the time that when the vast and prestigious 10-campus system agreed to "acknowledge and accept responsibility for the conditions" surrounding the fire that mortally injured Sangji, the precedent would inspire (or terrify) other institutions into tightening the lax safety standards that have long prevailed in many college and university labs.

Fourteen months on, it's hard to measure how much conditions have changed on the nation's hundreds of campuses—but, happily, there are some hopeful signs. There have been real improvements in safety practices within the UC system. The criminal case against Patrick Harran, the UCLA professor who headed the lab where Sangji worked, appears headed toward legal resolution. A new report from the American Chemical Society (ACS) provides concrete and detailed guidance to academic institutions on upgrading their safety practices. And, while most efforts concentrate on culture change and training, at least one researcher is taking a material approach, using nanotechnology to reduce common lab hazards.

Workplace safety pays

As the UC settlement noted, Sangji's injuries happened because she lacked the training, equipment, and attire needed to work safely with a pyrophoric substance. In exchange for dropping felony charges, the pact required the university to make prompt and sweeping changes in those three areas.

The UC system has worked vigorously to meet the terms of the settlement, reports Jyllian Kemsley of Chemical & Engineering News. A university policy document says that rules requiring safety training for everyone working in the system's labs and scientific stockrooms now apply "to workers regardless of their title; e.g., student, graduate student, postdoctoral scholar, faculty or other academic appointee, staff, visitor, volunteer, etc."

UC has purchased more than $4 million worth of personal protective equipment (PPE) for its researchers, including 115,000 lab coats, which the institution will regularly launder at a cost of half a million dollars a year. Students must now show up at lab courses equipped with their own appropriate PPE. UC's bulk lab coat order allowed them to get a bargain price, Kemsley notes. It also "solve[d] one PPE problem plaguing female researchers: a lack of small-sized women's coats from usual lab supply sources. UC custom ordered them." The new regimen appears to lower the chances that any member of a UC lab will again undertake a dangerous task as poorly trained, clad, and equipped as Sangji was that day in 2008.

UC can afford the laundry and PPE bills despite today's stringent budgets, Kemsley writes, because system-wide improvements in workplace safety have cut injuries by 3400 a year and workers' compensation payments by $50 million a year compared to 2004—a fact that should persuade other universities to boost their safety standards.

Legal responsibility

Scientists across the country disagree about whether the law should hold lab chiefs personally and legally responsible for the consequences of poor safety practices. Lawyers, however, note that under California's penal code and tough labor laws, an employer's failure to provide a worker required training and equipment when one has the resources to do so constitutes a willful violation, and a fatal outcome makes any such failure felonious. The courts, furthermore, have determined that the charges the district attorney has brought against Patrick Harran—who is of course presumed innocent until proven otherwise—are sufficiently serious to require a trial to establish his innocence or guilt. Harran is the first American academic to face felony charges related to a lab death. As we prepare this column, a pre-trial status conference is scheduled, after many delays, to take place on 3 October.

[UPDATE: The conference took place as scheduled. At such conferences, lawyers inform the judge of preparations for trial and any pending plea negotiations. The only apparent action was to set a new status conference for 20 November.]]

The core of safety

Identifying and Evaluating Hazards in Research Laboratories, the new ACS report issued in September at the organization's national meeting in Indianapolis, is the society's response to a request the U.S. Chemical Safety Board made in its landmark 2011 report on academic lab safety. The board asked ACS to provide "good practice guidance," ACS President Marinda Li Wu said at a press conference introducing the ACS document. The report also builds on a guide the society issued in 2012, Creating Safety Cultures in Academic Institutions. With these materials, Wu continued, "research laboratories throughout academia will be well equipped with the proper guidance to ensure safer learning and working environments for their students and faculty."

At the core of the report's approach is the task of dealing with laboratory hazards before work is undertaken. "[H]azard identification, hazard evaluation, and hazard mitigation in laboratory operations are critical skills that need to be part of any laboratory worker's education," Identifying and Evaluating Hazards states, and this process needs to become deeply ingrained in the daily practice of research. "[I]ntegrating these concepts into research activities is a discipline researchers must establish to ensure a safe working environment for themselves and their colleagues."

Key to the guide's approach is the distinction it draws between hazard and risk. A hazard, it explains, is a "potential for harm. The term is often associated with an agent, condition, or activity … that if left uncontrolled, can result in an injury, illness, loss of property, or damage to the environment. Hazards are intrinsic properties of agents, conditions, or activities." Here is an example: Pyrophorics like the one Sangji was handling are hazards, unavoidably bursting into flame on contact with air. A risk, on the other hand, is "[t]he probability or likelihood that a consequence will occur." Whether a worker handling a pyrophoric will suffer harm depends largely on what is done to cope with the material's known dangers.

Understanding, controlling, or mitigating the hazards inherent in experiments is therefore essential to reducing or removing the risk of mishaps and injuries. Only by identifying and analyzing the hazards present in a lab and its work can leadership "determine what hazard controls need to be put into place to allow the work to be performed safely," the report continues. Specifically, "[t]he results of an assessment or evaluation of the risk of the hazards of a given experiment should guide the selection of risk management techniques and tools—elimination or substitution of materials; primary safety devices or engineering controls, such as chemical fume hoods; personal protective equipment (PPE); and specific procedures and processes."

Beyond that, the report adds, "[i]t is equally important that time be taken after the work is completed to reflect upon lessons learned—what went as predicted or designed, as well as those things that did not." In addition, "when the work to be performed changes, that change must be evaluated against the current hazards analysis to determine if the hazards analysis continues to be sufficient. If this is not done, the researcher could begin the task not fully armed with the knowledge and mitigations to do the work safely." And, the report emphasizes, "[a]s the content expert in matters related to the laboratory, the PI [principal investigator] is most able to provide guidance concerning what constitutes a hazard in the performance of an experiment or research plan."

Analyzing and identifying specific research hazards is a highly technical and detailed process that must be adapted to the particular circumstances of each institution and laboratory. Fortunately there's no need to start from scratch. "Numerous hazard analysis techniques are used throughout various industries and institutions." Each organization can therefore select those most appropriate to its circumstances.

Ending insidious hazards

Training and analysis approaches are getting a lot of attention, but chemist Allen Apblett of Oklahoma State University in Stillwater believes that new technologies can eliminate some common but "insidious" hazards. For example, "there are a lot of solvents that go bad just like butter goes rancid by … oxidation, and those solvents can become explosive," he explained in a press conference at the ACS meeting. Ideally, labs should either test their solvents for safety "every few months or just throw them away," he continued—but the former often doesn't happen and the latter is seen as "wasteful."

Apblett has designed technologies to solve this problem, which he says represents "a revolutionary approach to making solvents safe in the laboratory." The first is a color-changing test strip that distinguishes safe from deteriorating solvents. Even more convenient is a pellet intended to be placed permanently in solvent bottles that changes color to indicate dangerous peroxidation. "As soon as you pick it up, you know the solvent is safe," he said. A third technology solves the problem of storing and handling certain potentially explosive substances by encasing them in a mineral that is "completely stable. You can hammer on it, you can grind it, you can put a flame to it but it doesn't blow up."

Whether through teaching or technology, increased attention to safety is providing opportunities to lessen the dangers that lab workers face. But reports, policies, and inventions will only help reduce those risks if campuses everywhere make them part of everyday life in every lab.

Beryl Lieff Benderly writes from Washington, D.C.

10.1126/science.caredit.a1300217