Chemistry matters. Join us to get the news you need.

If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



How To Make Chemistry Classroom Demonstrations And Experiments Safer

Fires that injure students prompt calls for safety assessments of demonstrations or experiments, plus teacher training

by Jyllian Kemsley
November 18, 2015 | APPEARED IN VOLUME 93, ISSUE 46

Credit: AP
Firefighters investigate Biery’s chemistry classroom at Western Reserve Academy after the 2006 fire that injured her and others.

“One of these days, someone is going to die” from injuries sustained in a chemistry classroom fire, Calais Weber Biery says. “I almost did.”


National Fire Protection Association: “Standard 45

National Research Council: “Prudent Practices in the Laboratory

American Chemical Society: “Chemical Safety in the Classroom

American Chemical Society: “Safety Data Sheets: Information that Could Save Your Life

Flinn Scientific: Videos

American Association of Chemistry Teachers: Webinars 

Download: PDF of this article

In 2006, when Biery was 15, her chemistry teacher set up a “rainbow” flame test demonstration of several metal salts that the teacher dissolved in methanol then set alight. One of the flames appeared to die down, so the teacher grabbed the stock bottle of methanol to add more. The classroom erupted with fire.

Biery suffered third- and fourth-degree burns over 48% of her body. “I had the rest of my childhood taken away,” she says. She spent two-and-a-half months in the hospital. After returning to school, she spent every break undergoing medical treatments. By her junior year of college, she decided that she’d had enough. Her remaining scars prompt stares or questions from strangers when she wears shorts or a bathing suit, she says.

Less obvious than the scars on Biery’s skin is her posttraumatic stress disorder. Although her symptoms have eased with time, the smell of something burning or the sound of people screaming still triggers anxiety, and she has what she calls “fire dreams” every few months.

What disturbs Biery just as often is that she’s not alone: Similar alcohol-fueled fires from educational demos occur with alarming frequency, injuring at least 72 people since 2011, according to 18 incidents tracked by C&EN, the U.S. Chemical Safety & Hazard Investigation Board, and the blog Chemjobber. Most recently, five students and a teacher were injured in an alcohol-fueled flame demo gone awry at W.T. Woodson High School in Virginia on Oct. 30, according to local media reports.

Despite her experience and concern for others, Biery agrees with safety and teaching experts that demonstrations and hands-on experiments by students are essential for science education. Biery herself returned to school to take Advanced Placement biology, followed by general and organic chemistry in college. But she thinks that chemistry teachers must take a step back to assess what they’re doing and how, and that states and school districts should provide teachers with training and equipment to do hands-on science safely.

Biery is not alone. “Kids need to experience phenomena if they’re going to be engaged and develop understanding,” says Joseph S. Krajcik, a former high school chemistry teacher who is now a science education professor at Michigan State University. “But I don’t think instruction should be driven by ‘splashiness,’ either,” Krajcik adds. Instead, teachers must stay focused on their educational goals and be clear about how a particular demo or experiment will meet those aims, then determine the safest way to achieve the objectives.

Credit: Jyllian Kemsley/Will Ludwig/NFPA/C&EN
In response to recent accidents in the classroom, here is a guide for performing experiments or demonstrations involving open flames, fire, or the use of flammable, reactive, toxic, or corrosive chemicals.

Download a PDF of this article and graphic here.

To help with finding the safest way, the National Fire Protection Association (NFPA) updated its standard 45 on “Fire Protection for Laboratories Using Chemicals” this year to add a chapter on educational and instructional laboratory operations, including specific guidance for conducting demonstrations. NFPA 45 also covers topics such as laboratory design, fire protection, fume hoods, personal protective equipment (PPE), and chemical storage and handling.

When an NFPA committee convened to update the standard, “it was really clear to us in light of the number of student injuries from demos with flammable materials that we needed to include requirements to protect students,” says committee member Barbara Foster, who is director of laboratory safety for the chemistry department at West Virginia University.

The committee wasn’t just concerned about flame demos, adds committee chair Andrew Minister, who is chief fire protection engineer at Pacific Northwest National Laboratory. “Whoosh bottle” demos in which alcohol vapor is ignited also frequently go wrong, as do demos involving oxidizing sugar with potassium chlorate or potassium nitrate.

NFPA has no legal authority itself, but some jurisdictions adopt its standards as regulatory or legal requirements. West Virginia incorporated the updated NFPA 45 into its fire code in June. New York City—where a “rainbow” demo gone wrong injured two students in January 2014—is doing similarly, with fire code updates to take effect in January 2016. But even where NFPA 45 isn’t adopted, it still represents good professional practice.

The American Chemical Society Committee on Chemical Safety recommends that educators discontinue flame experiments involving flammable solvents on open benches. Instead, it recommends soaking sticks in salt solutions, then burning them in a Bunsen burner, as demonstrated in this video.
Credit: ACS

For educational and instructional laboratories, NFPA says that instructors “shall be trained and knowledgeable in fire safety procedures, emergency plans, the hazards present in the lab, the appropriate use of PPE, and how to properly conduct a hazard risk assessment.” For instructor-led demonstrations or experiments done by students, instructors must review hazards, assess risks, provide safety briefings to students, and use or provide appropriate PPE.

NFPA also says that bulk quantities of chemicals must be locked in a separate room or cabinet. Before students arrive, instructors should dispense only the minimum amounts necessary for the activity. Demos involving open flames or flammable, reactive, toxic, or corrosive chemicals should be done in a fume hood if possible. If a fume hood is not available, instructors should place an impact-resistant barrier between the experiment and students, or ensure that students are more than 10 feet (~3 meters) away.

The NFPA guidance largely echoes recommendations issued last year by the Chemical Safety Board in a safety bulletin, “Key Lessons for Preventing Incidents from Flammable Chemicals in Educational Demonstrations.”

Beyond the NFPA guidance, teachers need to practice demos or experiments ahead of time, recommends Donald J. DeCoste, a chemistry professor at the University of Illinois, Urbana-Champaign, and adviser to the school’s master’s degree in the teaching of chemistry program. Even if an activity is something that an instructor has done before, “if you think it needs to be bigger, or you change some variable, then you need to think about that and practice it on your own before doing it in front of students, because you’re just not sure how it’s going to go,” he says.

And as schools embrace more inquiry-based learning in which students may construct their own experiments, teachers need to stay closely involved. “You don’t just send students into the storeroom and say, ‘Go,’ ” DeCoste says. If students are mixing various solutions, for example, then teachers need to be certain in advance that all combinations are safe. If there are safety concerns, then students can plan out their procedures, but the teacher must check the procedures before experiments begin.

But to accomplish all of this, teachers need better lab safety training and resources. The training needs to start before they’re in a classroom on their own, says Ken Roy, director of environmental health and safety for Glastonbury Public Schools in Connecticut and chief science safety compliance adviser to the National Science Teachers Association. “With some exceptions, universities and colleges don’t seem to be preparing teachers to do what they need to do safety-wise,” Roy says. Then, once teachers are in the classroom, they also need ongoing in-service training, and states and school districts need to allocate the resources for that—as well as for equipment such as eyewashes, fume hoods, impact-resistant barriers, and PPE.

“There are a lot of chemistry teachers who are very safety conscious and very careful,” comments Deanna M. Cullen, a chemistry teacher at Whitehall High School in Michigan and an associate editor of the Journal of Chemical Education. “At the same time, I’ve been teaching for more than 20 years, and nobody’s really ever checked on my safety. Early on, following my mentors, I did things that I didn’t realize weren’t okay. I feel like I’m lucky that something bad didn’t happen in my classroom.”



This article has been sent to the following recipient:

Joseph A. Castellano (November 18, 2015 8:43 PM)
I have done flame test demos in an 8th grade classroom for the past 10 years without incident. I use a technique that is similar to that recommended by ACS. I make a small slurry of the salt in water contained in a 25 ml test tube. I use a wooden stick with a small loop of wire attached at one end, dip the loop into the slurry and place the loop into the flame from a small propane tank. The students are always at a distance of about 10 feet.
Bruce Sahli (November 18, 2015 9:19 PM)
I was wondering if analysis has been done on the data available from these accidents that are being tracked to look for common root cause factors, for instance, are the instructors inexperienced, and also has there been any number crunching as to costs to the schools for the accidents, etc., and also what are the economic demographics, e.g. does this type of thing happen mostly in low income school districts, in schools lacking in safety equipment and money to spend on safety training?
Ian Seiler (November 20, 2015 11:17 AM)
In all the reports I have read, especially the reports released by investigations from the CSB, the root cause usually traces backs to complacence and a phenomenon of normalization of deviance. This is usually because the established safety protocols are communicated verbally and over time it becomes "hazard analysis by telephone".

These complacency issues can usually be resolved by having written protocols and a requirement to review the written protocols every time the demonstration is performed.

To answer your questions, nearly every incident comes down to a lack of effective safety training and lack of written protocols.
Jyllian (November 20, 2015 1:38 PM)
Not as far as I'm aware.
Kathryn Jensen (November 21, 2015 10:18 AM)
I have the students do flame tests using 0.2M aqueous solutions of the chloride salts and q-tips with a bunsen burner. They can see the colors and enjoy the connection to fireworks and color in nature.
Lynne Mattingly (November 23, 2015 7:43 AM)
Where can a chemistry teacher purchase impact resistant barrier for lab benches.
Jyllian (November 24, 2015 8:02 PM)
Search for "lab safety shield." I see them at Flinn Scientific, Cole-Parmer, VWR, and elsewhere. Make sure you check the dimensions to get what you need.
Gerald Ceasar (November 28, 2015 2:31 PM)
I live here in the Washington DC area where the latest accident with the Rainbow experiment was done. From reading the reports which were extensively reported in the Washington Post, the teacher involved violated all rules of chemical safety good sense. She lit methanol with dissolved metal salts with an open Bunsen burner. The methanol was splashed onto an open desk for further dramatic effects. Things really went bad when she tried to continue the show by adding further methanol from a bottle that ignited injuring her and several students. Anyone with good chemical sense would never bring an open flame to a volatile organic solvent like methanol. As a footnote to this terrible accident, the principal of this school district banned all use of Bunsen burners in all science demonstrations. She didn't realize that it was use of volatile organic solvents that was the real problem.
Gerald Ceasar,Emeritus ACS member
In reading the reports of others who have showed the different flames from dissolved metal salts, all the salts were dissolved in water not volatile methyl alcohol.This was the same way it was done in first year college qualitative analysis.Not as showy but much safer
Ann Marie Pfohl (December 4, 2015 6:39 PM)
As a member of a county level hazardous material response team I had an opportunity to take a chemistry of hazardous materials class supported by the local community college. The class included a lab as well as classroom lectures. Some of the labs took place in the college chemistry lab while the more hazardous experiments took place at a fire training facility. In addition to the facilities the level of PPE included the wear of structural firefighting gear or hazardous material response equipment. The determination of the required PPE was dependent on the hazards. The class was interesting and informative.
Roger Barth (March 10, 2016 10:16 AM)
Schools are used to forms. It may help to encourage school districts to require a form for demos that go beyond water, food coloring, and salt in which the teacher has to specify the flammability class (methanol is class IA, the most flammable class), toxicity information, information about accidents that have occurred with the material and other information to assure that the teacher has some recent information. If the teachers involved in these accidents had to look up the properties of methanol to fill out the form, they may have had second thoughts about running the demo.
Ivone Gomez (May 17, 2017 12:08 PM)
I am a chemistry student for master degree and I am wondering if it is possible to do virtual labs instead of these dangerous procedires that are only for demsontration purposes. it will save time, money, and the most important the life of our students.
Jyllian Kemsley (May 22, 2017 12:59 PM)
In some cases, yes, a virtual lab or video might suffice. But live, hands-on demos and experiments provide a different learning experience. To quote from the story:
"Despite her experience and concern for others, Biery agrees with safety and teaching experts that demonstrations and hands-on experiments by students are essential for science education. Biery herself returned to school to take Advanced Placement biology, followed by general and organic chemistry in college. But she thinks that chemistry teachers must take a step back to assess what they’re doing and how, and that states and school districts should provide teachers with training and equipment to do hands-on science safely."

Leave A Comment

*Required to comment