Click here for free access to our latest coronavirus/COVID-19 research, commentary, and news.

Support nonprofit science journalism

Science’s extensive COVID-19 coverage is free to all readers. To support our nonprofit science journalism, please make a tax-deductible gift today.

Lab Safety Requires Training and Commitment

Credit: National Science Foundation and Argonne National Lab

“What can go wrong? What can I do to minimize risks of an experiment? What do I do if something does go wrong?” Researchers need to ask themselves these questions whenever they begin a new experiment, advises James Kaufman, president and CEO of the Laboratory Safety Institute, a provider of safety training and other services to academia and industry. “The frequency of academic research lab accidents is 10 to 50 times greater than in industrial labs. So there is a lot of room for improving safety.” Whereas the university can and should take very seriously the safety of their students, postdocs, and other workers, it would be foolish for those in the lab to depend on others to take care of things, Kaufman suggests. Keeping researchers safe is ultimately the responsibility of researchers themselves.

Before you start

Researchers, of course, often have other things on their minds--like research. Furthermore, they usually they lack specific training in the difficult task of keeping themselves and their laboratories safe--especially early in their careers. It's natural for young scientists and science students to assume that the institutions where they're training will take care of them, but that's a potentially dangerous assumption. So how should people in the laboratory go about rising to the challenge of keeping themselves and their colleagues safe?

The first step should be to learn their institution’s policies and practices, says Michele Johnson of the University of Utah Environmental Health and Safety Department. Johnson proposes three questions of her own that researchers need to answer before the work in the lab gets started:

  • What safety training do I need?

  • How do I get it?

  • Who do I go to if I have safety concerns?

This last question is important because it is the key to finding answers to the first two. On some campuses, principal investigators (who rarely have rigorous safety training) have the responsibility for ensuring a safe working environment. On other campuses it is the facilities manager. Why does it matter to researchers who's in charge? Ralph Allen, Director of the University of Virginia’s Office of Environmental Health and Safety puts it this way: "They need to protect themselves since the lab is a place where things are inherently dangerous. So it’s important to know where to go for safety information and know whom to consult.”

In the lab

“Nine times out of ten, lab accidents are caused by operator error,” observes Johnson--and operator error, in turn, is often due to operator fatigue, inattention, or haste. Other common causes of laboratory accidents are improper use of equipment, the use of the wrong tool for the job, and poor equipment maintenance.

At the University of Virginia, organic chemistry students were storing samples they prepared in an unlabelled refrigerator that was not explosion proof. When the refrigerator exploded, the doors to the main lab were blown to the other side of the room where they hit an apparatus used to purify solvents.

There's a happy ending. The solvent-purification setup was made of metal; similar setups, in laboratories across America, are made of glass. The metal purification system was damaged, but if it had been glass a fire probably would have resulted, causing far more extensive damage.

No one knows who chose that old refrigerator, but many people over the years--PI's, administrators, countless students, postdocs, and other lab staffers--could have recognized the threat it posed and switched it out. But until this accident, no one did.

After the accident, the university traded out all their old refrigerators for explosion-proof models and replaced glass solvent-purification systems with metal ones. If faculty members lacked funds to replace their unsafe equipment, the university paid the tab.It took a near-catastrophe to identify the problem, but the UVA is now a safer place for the workers in its laboratories.

Recognizing such risks isn't easy, but it is essential if labs are to be safe places to work. “Researchers themselves must maintain labs from a safety perspective,” advises Jim Kapin, 2006 chair of the American Chemical Society Division of Chemical Health and Safety and senior staff scientist with Advanced Chemical Safety, a firm that conducts academic research lab safety inspections and training programs.

In a laboratory at Ohio State University, solvent bottles fell from the top shelf of a four-shelf flammable-solvent storage cabinet because the clips used to secure the shelf failed. The clips had been modified improperly, probably in the distant past by a researcher who no one remembers. An accident had been stalking the lab ever since.

When it finally pounced a spill resulted, and the supply of spill-control agent proved inadequate to absorb the spill. Solvent vapors forced the students to retreat and the building was evacuated. An explosion and a fire followed, destroying the laboratory. Fortunately, researchers suffered only minor injuries, but the damage was extensive.

This incident demonstrates how the choices scientists make affect not just their own safety but also the safety of their colleagues, for years to come. It also shows how important--and how hard--it can be for researchers to identify unsafe conditions in their laboratory. Improperly modified shelf supports would be hard for anyone to spot in a routine inspection, but an attentive staffer might have noticed, when placing a bottle on the shelf, that the supports seemed weak or unsteady. If she had, the problem might have been solved before the accident occurred. Laboratory workers and managers--and safety inspectors--made another serious mistake, this one far more routine and easier to spot: they failed to make sure the laboratory's spill-kit was adequate for the volume of spills that were likely to occur.

Interdisciplinary risk

Interdisciplinary research labs may present more risks than labs doing work in established disciplines, says Kapin, because the range of risks is wider. “Chemists know how not to blow things up," he notes, "but often don’t know the biological hazards of working with laboratory animals.” Mixed environments also mean more opportunities for reagents, supplies, equipment, and researchers to mix in unpredictable ways, creating new risks. The only real solution in such environments is extra vigilance. “Constant vigilance and regular inspections offer the only hope for spotting these problems,” says Kaufman.

Safety glasses

Fortunately not all safety-related judgments are hard. One thing that's very easy to do--and that students, postdocs, and other lab works can take responsibility for--is always wear safety glasses in the lab. Risks to researchers’ eyes, notes Kaufmann, include “impact from objects such as broken glass, heat, dust, chemicals and optical radiation." So "researchers should use the appropriate safety glasses for the particular hazards associated with their experiment.” For more severe chemical splash hazards, Kaufman recommends wearing goggles and a face shield.

It's important to look beyond your few feet of bench; chemicals can splash, and glass shards can fly, a long way. "You may need safety glasses, not for your own work, but for someone else’s near you," says Allen. "So pay attention to what’s going on around you, not just your own experiments.” And always remember that safety glasses can protect you from risks you may not have even thought of--so wear them even when there isn't an obvious reason.

In 1987, the U.S. Centers for Disease Control and Prevention issued a bulletin recommending that researchers working with monkeys wear safety glasses. Ten years later, 22-year-old researcher Elizabeth Griffin was working at Emory University’s Yerkes Primate Center. She was not wearing eye protection and a macaque monkey’s urine contacted her eye. Ms. Griffin wiped her eye with a wet paper towel and flushed it 45 minutes later. It was too little, too late: she contracted the Herpes B virus and died within 2 months.

Three years later, the Coulston Foundation, which also used monkeys for disease research, was cited for lack of safety eyeglasses and other personal protective equipment. And even today, Kaufman observes, the use of eye protection “is not as high as one might reasonably expect.” The proper use of eye protection seems to depend, more than anything else, on the example set by the PI and senior lab personnel (see the related story).


Frequent and regular lab safety inspections are “the only hope for spotting problems” according to Kaufman. Johnson agrees that "external inspections every 6 months or so are essential"--but, she says, they are not sufficient. Johnson recommends that research groups do their own internal inspections once a month, and that responsibility for those inspections be placed on a postdoc or senior graduate student rather than a PI; many principal investigators, Johnson observes, may not have enough recent experience in the lab to recognize the hazards. And just as airplanes undergo a safety check before every flight, Kaufman notes, most laboratory equipment should be safety-checked before every use.


Aren't most researchers already aware of the risks they face in the laboratory? That, says Allen, “is a dangerous assumption.” Specific training is essential; in laboratory safety, common sense can only take you so far. It is important that all incoming students and postdocs undergo safety training that goes beyond which form to fill out and who to report an accident to. “New researchers need to consult with their principal investigator to determine what specific training they need to do their research and how to get this training,” advises Johnson. They need to become well versed in the specific risks the research they’ll be doing presents, and in industry-standard methods of mitigating these risks. Also, they need to know what to do whenever the most likely and predictable accidents occur.

But even this kind of training is insufficient. Researchers need to learn to improvise, and to evaluate their surroundings for potential hazards--like hazardous electrical wiring and weak shelf supports--beyond those listed in user manuals, training courses, and materials safety data sheets. Such hazards are best evaluated using common sense and an attentive eye.

“Researchers’ best bet to work safely is to educate themselves about the hazards of their laboratories and know who to go to for information that they can’t find themselves,” advises Kapin. "Take the time to find out what can go wrong," says Allen. Determine "what you can do to be prepared[,] and don’t assume anyone else will be looking out for you.”

Reliable sources of lab safety information

Besides your own university’s safety Web site and the references your principal investigator or lab safety officer recommends, the following are useful sources of reliable information:

Web sites


Handbook of Laboratory Health and Safety
R. Scott Stricoff, Douglas B. Walters
Wiley-Interscience; 2 edition (March 20, 1995)

Improving Safety in the Chemical Laboratory: A Practical Guide
Jay A. Young (Editor)
Wiley-Interscience; 2 edition (June 1, 1991)

Handbook of Chemical Health and Safety
Robert J. Alaimo (Editor)
ACS Handbooks (Hardcover)
American Chemical Society Publication (April 19, 2001)

OSHA Medical Radiation Safety Guidebook
Bruce Gordon, Daniel Farb
UniversityOfHealthCare (July 2005)

Safety Sense: A Laboratory Guide
Cold Spring Harbor Laboratory
Cold Spring Harbor Laboratory Press (September 1, 2001)

John K. Borchardt has a Ph.D. in chemistry. He is the author of the book Career Management for Scientists and Engineers.

Comments, suggestions? Please send your feedback to our editor.