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.



Ablaze Over Furniture Fires

Fire-safety scientists are outraged by a chemist’s campaign against flame-retardant chemicals

by William G. Schulz
October 29, 2012 | APPEARED IN VOLUME 90, ISSUE 44

CLARIFICATION: This story was updated on Oct. 30, 2012, to clarify the following: C&EN stated that the Polyurethane Foam Association (PFA) provided grant money to Arlene Blum; PFA maintains that it only ever gave Blum a one-time speaker fee, not grant money.
In an interview, C&EN’s William Schulz asks California State Sen. Mark Leno (D) about consulting the chemical industry on the issue flame-retardant chemicals.
Credit: C&EN


Ablaze Over Furniture Fires

The modest home of Arlene Blum in the hills of Berkeley, Calif., has no overstuffed sofas or chairs. That’s because this mountaineer and chemist is deeply concerned about her exposure—and yours—to the flame-retardant chemicals that are nearly always present in the polyurethane foam innards and other parts of such furniture. Flame retardants are necessary additives if foam and furniture makers’ products are to meet California’s strict fire-safety laws for upholstered furniture.

Credit: William Schulz/C&EN
A California Department of Consumer Affairs technician shows upholstered furniture tested to meet the state’s fire protection standards.
Credit: William Schulz/C&EN
A California Department of Consumer Affairs technician shows upholstered furniture tested to meet the state’s fire protection standards.

The chemicals won’t be needed if Blum has her way. As founder and executive director of the Green Science Policy Institute, she operates a nerve center in Berkeley dedicated to the eradication of many flame retardants in consumer products. She is credited in part with a move now under way to overhaul sections of California’s fire-safety regulations for upholstered furniture. That revision, if it takes effect next summer in its current form, would mostly eliminate the need for flame retardants in residential home furnishing products.

Fire-safety experts and other people who oppose Blum’s campaign—and who have called for a more reasoned analysis of the fire-safety science on flame retardants—say she brushes them aside as disqualified to provide expert opinion because of current, previous, or inconsequential chemical industry ties. They say Blum and her supporters shout them down at public workshops or refuse to engage in dialogue over the issues and the science. Worse, they say, Blum is promoting false information about fire safety from flame retardants via a high-visibility media campaign that stokes fear of chemicals that have been used for decades and have saved the lives of tens of thousands of people in the U.S. who have been the victims of home fires.

To some people, Blum’s concerns seem especially alarmist and speculative because of materials she writes and distributes. For example, they cite a 2008 Los Angeles Times editorial in which she ponders whether exposure to flame retardants brought on the hyperthyroidism that killed her cat, Midnight. A veterinarian found that Midnight’s blood contained high levels of flame-retardant chemicals.

Blum and other scientists insist that a growing body of evidence indicates that the brominated flame-retardant chemicals used in upholstered furniture may, in some cases, be endocrine disrupters or have neurological and other health effects that make them unacceptable for use in everyday objects like sofas and chairs. Such flame retardants, she says, tend to accumulate in tissues and have been detected in the blood of adults and children.

“There are some 3,700 peer-reviewed papers on flame-retardant chemicals’ toxicity,” Blum says emphatically. Children especially, she says, should have very limited—if any—exposure to compounds that might damage their physical and intellectual development or leave them more vulnerable to other chronic health problems.

Industry scientists and other researchers quickly counter that exposure does not equal risk. In a new book, “Flame Retardancy in 2012,” Florida Institute of Technology professor Gordon L. Nelson and coworkers write: “Analytical chemistry now allows testing at very low levels. Reports of flame retardants at levels of parts per billion have been reported for household dust and blood.” Nelson is considered one of the world’s leading authorities on flame retardants and fire safety.

The book’s coauthors go on to say that 70 or more chemicals are frequently found in human blood, “thus, knowing one class and not looking for or knowing the levels of the others can give misleading human health attribution. Interestingly halogen flame retardants receive considerable criticism, yet some 5,000 different organo­halogens are found in nature.”

Still, Blum’s questioning is not just idle speculation. Producers of pentabromodiphenyl ether (pentaBDE) withdrew the product from sale in 2004 because of negative risk assessment data developed in Europe. For similar reasons, U.S. suppliers of decaBDE have pledged to the Environmental Protection Agency that they will phase out decaBDE by the end of the year, except for some military and transportation applications. And EPA is moving to restrict the flame retardant hexabromocyclododecane (HBCD) as well as other members of the polyBDE class of flame-retardant compounds because of the very health and environmental concerns Blum and supporters have outlined. Other flame-retardant chemicals may come under review by EPA, which is seeking to refine the nation’s chemical control law (see box on page 36).

Blum says the case against flame retardants goes beyond their potential toxicity. She argues the chemicals do not work as advertised, an assertion she has made repeatedly in several review papers she has coauthored (Rev. Environ. Health2010,25, 261) and in many news media interviews she has given in recent months.

When asked if there could be differing interpretations of what is a large and complex body of data on the efficacy of flame retardants, Blum snaps, “I am not a fire scientist. This is such a broad field. There are about 10 disciplines involved.”

Blum says that other ways to prevent deaths from upholstered-furniture fires—improved building codes, reduced cigarette smoking, and increased use of sprinkler systems, for example—undermine the case for using flame-retardant chemicals.

In fact, Blum is now working in overdrive to support a revision of California Technical Bulletin 117 (TB 117), as the Cal­ifornia fire-safety standard for upholstered furniture is known. The potential impact of a revision—ordered in June by California Gov. Edmund G. Brown Jr.—is hard to overstate. Because the federal government, for nearly 40 years, has failed to produce a national fire protection standard for the polyurethane foam and upholstered furniture industry (see box on page 32), TB 117 has become the de facto national standard. If California substantially alters or weakens its fire-safety upholstery standard—and many fire-safety scientists who spoke with C&EN believe that state officials, including Gov. Brown, are trying to do just that—then the same standards will apply nationwide.

The current regulation is spelled out in a highly detailed document that defines the fire-resistance performance standards for a wide variety of furniture fillings—natural and man-made—as well as upholstery fabrics. The regulation prescribes the exact test procedures to be used for each type of material or furniture component, as well as the testing equipment to be used. Each furniture component must pass two tests: One examines how quickly the material burns when ignited by a smoldering cigarette and the other by an open flame.

Credit: William Schulz/C&EN
Credit: William Schulz/C&EN

The cigarette smolder test was developed because smoking materials are the leading cause of upholstered-furniture fires. In 2009, the most recent year for which data are available, 27% of the 5,600 fires in the U.S. that began with upholstered furniture were ignited by cigarettes and other smoking materials, according to data gathered by the National Fire Protection Association. Candles, lighters, and matches were the sources of another 20% of these fires, and this is the statistic fire-safety experts cite as the reason to have an open-flame test.

Other sources of upholstered-furniture fires include space heaters and ember or ash from fireplaces or wood stoves.

The fear that California is weakening its standard—and thus compromising public safety—comes from a draft revision of TB 117 put forth by the California Department of Consumer Affairs earlier this year for discussion by stakeholders, including firms that make flame-retardant compounds. The draft revision of TB 117 sets furniture performance benchmarks only for the fire hazard from a standardized cigarette smolder ignition test. There would be no open-flame challenge to furniture components, as is required by today’s TB 117 and considered an absolute necessity by many independent fire-safety experts who spoke with C&EN.

Today’s fabrics and other barrier materials surrounding the foam, Blum and proponents of a smolder-only revision say, are enough to prevent the chief cause of fires started by upholstered furniture—an accidentally dropped cigarette. Foam and furniture manufacturers could meet the new standard, Blum says, without having to use flame retardants at all.

Credit: William Schulz/C&EN
Chemist Carrie Cathalifaud of California’s Department of Consumer Affairs shows a strip of fabric tested to comply with the state’s strict fire protection standards for home furnishings.
Credit: William Schulz/C&EN
Chemist Carrie Cathalifaud of California’s Department of Consumer Affairs shows a strip of fabric tested to comply with the state’s strict fire protection standards for home furnishings.

Blum’s insistence that the smolder standard is enough and that flame-retardant chemicals are toxic and ineffective has gained widespread support and publicity—much of it generated by Blum herself. Her conclusions and assertions about flame-retardant chemicals, along with those of experts who agree with her, have appeared in several high-profile media outlets, including the Chicago Tribune, the New York Times, and the Diane Rehm Show on public radio. California politicians, like State Sen. Mark Leno (D), credit her with their awareness of the flame-retardants issue and the health and environmental problems flame retardants allegedly cause.

“I see no proof that TB 117 has helped,” Leno says. Before Brown issued the directive to revise TB 117, Leno introduced five bills in the California State Legislature to deregulate the use of flame retardants so that consumers could choose whether to buy furniture treated with the chemicals. Leno says he is convinced that the chemicals do not retard fires.

When asked about seeking information from experts on the other side of the argument, Leno becomes visibly angry. He claims that strong-arm tactics by industry lobbyists and dishonest experts working on industry’s behalf shot down all of his efforts toward measured reform of the state’s fire-safety laws. “They lied to us,” he says, and in the process have discredited themselves, especially through the now-defunct and heavily criticized industry group Citizens for Fire Safety. In such a sour atmosphere, Leno says, the governor had no alternative but to issue a directive to reform TB 117 and lower exposures to flame-retardant chemicals.

“Toxic flame retardants are found in everything from high chairs to couches, and a growing body of evidence suggests that these chemicals harm human health and the environment,” Brown said in a statement after issuing his order. “The guidelines in place now—Technical Bulletin 117 for flammability standards—will be updated to reflect modern manufacturing methods that can lower the use of harmful chemicals,” he said. The governor’s office did not respond to a request for an interview with C&EN.

A growing chorus of independent fire-safety scientists, materials chemists, and chemical industry experts is pushing back against Blum’s anti-flame-retardant-chemicals campaign in the hope that they can preserve and perhaps even strengthen TB 117. They are openly critical of Blum, her scientific credentials, and her assertions regarding the safety and efficacy of flame-retardant compounds. They say Blum’s campaign ignores or downplays the danger of residential fires and the fire hazard presented by nontreated upholstered furniture and polyurethane foam. Many contend that she is simply conducting another in a long line of misguided crusades against “toxic chemicals.”

Credit: William Schulz/C&EN
A student at Florida Institute of Technology places a sample in a calorimeter for flame-retardant testing.
Credit: William Schulz/C&EN
A student at Florida Institute of Technology places a sample in a calorimeter for flame-retardant testing.

“What’s the evidence that flame-retardant chemicals don’t work?” asks Florida Institute of Technology’s Nelson, who is also an American Chemical Society past-president. Since the start of his career in the early 1970s, he has researched and developed flame-retardant technologies to protect against fire in a variety of plastics, electronics, consumer products, and military applications. He says Blum’s claim that flame-retardant chemicals “don’t work” contradicts and is offensive to the thousands of scientists whose decades of work reported in thousands of peer-reviewed papers demonstrate otherwise. “Flame-retardant materials work when properly used,” Nelson says emphatically.

One project that showed this definitively—and that is often cited by Nelson and many other experts—is “Fire Hazard Comparisons of Fire-Retarded and Non-Fire-Retarded Products,” a 1988 publication from the National Bureau of Standards, which is now the National Institute of Standards & Technology (NIST). The study addressed the question: Is overall fire hazard reduced when comparing fire-retarded to non-fire-retarded items that are otherwise substantially identical products? Agency scientists devised tests to burn treated and nontreated items of furniture in special test chambers, or calorimeters, to find out.

These tests didn’t replicate or examine the efficacy of the TB 117 standard. The study aimed to test the overall performance of flame retardants and was meant to reflect a situation in which furniture was a secondary source, or fuel, for spreading fire. As such, the tests examined the chemicals’ ability to retard fires from ignition sources of about 30 kW compared with the 83–295-kW ignition sources used in TB 117 tests.


The agency’s tests, the report says, made clear that when furniture is treated with flame retardants, less of it is consumed by a fire. The agency also found that heat release from treated materials is 25% that from nontreated material and that the quantity and toxicity of gases produced in a fire were much less for treated products. To the last point, it is often claimed—incorrectly, as many later tests have also shown—that flame-retardant additives contribute to more, and more toxic, smoke than nontreated products.

Credit: Courtesy of FIT
Credit: Courtesy of FIT

In all, the agency’s report concluded that “fire retardant additives did decrease the overall fire hazard of their host products.”

“This work determined that escape time for building occupants can rise significantly with the use of fire retardants,” says NIST Senior Research Scientist Richard G. Gann, who organized the project. “Fire retardants can decrease the amount of a flammable product—such as foam used to cushion furniture—that burns and can slow the rate of that burning.”

“Flame-retardant chemicals are effective. There’s no doubt about it,” says Matthew S. Blais, director of the Fire Technology Research Laboratory at Southwest Research Institute (SWRI), a nonprofit, independent testing laboratory in San Antonio. Recent fire-testing results at SWRI facilities for the National Institute of Justice, part of the Department of Justice, back up his assertions.

The details are in a 2012 report, “Reducing Uncertainty of Quantifying the Burning Rate of Upholstered Furniture,” which offers guidelines for how best to estimate the burning rate of upholstered furniture and quantify the uncertainty of the predictions. Such information is useful to law enforcement in arson investigations, for example.

The SWRI team conducted a series of 79 full-scale fire tests using upholstered-furniture mock-ups made from foam, fabrics, and other materials available off-the-shelf in the San Antonio area. The idea was to test materials in common use for furniture and furniture components nationwide. Next, the team made mock-up cushions constructed with either flame-retardant-treated or non-flame-retardant-treated cotton, as well as one of six padding materials—including TB 117-compliant polyurethane foam. The team was trying to determine heat release rates for the various types of furniture materials tested, which can be useful in documenting how individual fires may have started and spread.

Urethane-foam-filled furnishings have the potential to contribute tremendous energy to a fire. When not protected with flame retardants, Blais says, these materials burn rapidly, eventually yielding enough heat that everything else in a room ignites. The study revealed that untreated foams needed to burn for as few as 200 seconds to reach this so-called flashover point.

The tests proved that adding a cover treated with flame retardants over the foam adds a layer of defense that delays transition to flashover to almost 800 seconds from initiation, Blais says. And the additional use of TB 117-rated urethane foams prevented sustained burning, he adds. “The impact of adding flame retardant to the covering material and urethane foams adds defense in depth to the furnishing that may save lives. Ten minutes is a lot of escape time.”

With the results of these and other fire tests in mind, Joel Tenney, director of advocacy for ICL Industrial Products, a manufacturer of flame-retardant compounds, says: “In California, there has been a rush to say, ‘Let’s avoid chemicals,’ rather than have a high fire-safety standard. Going away from the open-flame challenge is the wrong direction.”

“We think the weight of the evidence supports that flame retardants are effective,” says Jackson Morrill, director of the North American Flame Retardant Alliance, which is part of the chemical industry association American Chemistry Council. “We believe in a fact-based, science-based dialogue. To say flame retardants don’t work—that’s a massive misperception.” He says the dialogue on flame retardants must include discussions of fire safety.

Blum and her supporters, however, cite a 1997 Consumer Product Safety Commission study of TB 117-compliant chairs, “Upholstered Furniture Flammability: Fires Ignited by Small Open Flames and Cigarettes.” It reports that the TB 117 standard “would not, if federally mandated, ensure a substantial reduction in the risk of small open-flame ignition of finished articles of furniture.”

But for Blum it is the toxic potential of flame retardants that is of paramount concern. She first worked on the toxicity of flame retardants in 1977 when, as a postdoc in the laboratory of University of California, Berkeley, chemistry professor Bruce N. Ames, she coauthored a paper with him on flame retardants in children’s sleepwear (Science, DOI: 10.1126/science.831254). That work demonstrated the mutagenic and carcinogenic hazards of the flame retardant tris(2,3-dibromopropyl) phosphate and contributed to an eventual withdrawal of the chemical in children’s clothing.

Blum praises Ames for his courage to publish strong statements about the hazard to children they uncovered for this flame retardant. She says the experience left her with the impression that whenever strong scientific information is presented, the “correct” public policy decisions will follow.

Then Blum took a long sabbatical from chemistry to climb mountains. Her absence from the field often gives her critics pause when evaluating her chemistry expertise or knowledge of the flame-retardant chemical industry and practices. What’s more, several critics point out that Blum has had a sustained relationship with officials of the Polyurethane Foam Association, another industry group stakeholder that might see cost-saving benefits from a TB 117 revision. Critics also say that Blum has accepted grant money from that group, which Blum confirms. PFA, however, says it only ever gave Blum a one-time speaker fee, not grant money.

After her mountain-climbing career, which includes two books she wrote about her adventures, Blum returned to work on chemistry and health and environmental issues. In 2007, she published a letter in Science detailing her opposition to a variety of compounds related to or that have replaced the polyurethane-foam flame retardant pentaBDE (DOI: 10.1126/science.318.5848.194b). Soon after, furniture manufacturers began to rely on foam treated with Chemtura’s Firemaster products, including those that contain 2-ethylhexyl-2,3,4,5-tetrabromobenzoate (TBB) and bis(2-ethylhexyl)-3,4,5,6-tetrabromophthalate (TBPH).

Blum is proud of her activism and what she clearly sees as a battle between industry profits and public health. And she insists that she is not “antichemical.” “I love chemistry,” she says, and she believes in the utility and benefits that many different chemical compounds have contributed to daily life.

She says her experience in the current go-round fighting the use of flame retardants is different from her experience in the 1970s, in terms of the resistance she has encountered to changing the fire standards or getting rid of the chemicals. She sees a conspiracy in industry to keep the chemicals on the market despite knowledge of their hazards, and she claims to have anonymous sources working for flame-retardant manufacturers who can back up her allegations. It would be too risky to put them in touch with C&EN, she says, however, even as anonymous sources.

“Some people have been on a crusade against flame-retardant chemicals for years,” scoffs chemist Marcelo M. Hirschler of GBH International, a fire-safety consulting firm. On the basis of a poor or deliberately misleading understanding of chemistry, he says, they will seize on the possible toxicity of one problematic molecule—pentaBDE, for example—to damn whole classes of related chemical entities. As a chemist, he says, Blum should know that even slight variations in chemical structure can result in vastly different toxicology profiles for related chemicals.

Blum often cites the structural similarity of compounds related to pentaBDE, for example, to drive home her message about the potential toxicity of other brominated flame retardants. She is perhaps not wrong to do so, because even EPA uses structure-based predictive toxicology to prioritize the flame retardants it must examine in more detail for possible toxic effects.

Like many fire-safety experts, Hirschler says that, questions and concerns about flame retardants’ toxicity aside, tough fire-safety standards force innovation. Already, he points out, numerous research efforts are under way to develop new flame retardants that can be used in lower amounts or have greatly improved toxicity profiles (see page 39).

If the revision of TB 117 stays as is, Hirschler says, “it will be the first time in the U.S. that the law lowers safety. That never happens. We always go in the opposite direction.” He says both the polyurethane foam industry and the furniture industry, both of which typically absorb the cost of flame-retardant treatments without passing it along to consumers, have a stake in seeing the standard revised downward. Two-thirds of all furniture sold in the U.S. is TB 117-compliant, he says. If the standard is lowered, “why should furniture manufacturers make 117-compliant products? They won’t.”


Hirschler is not alone in his alarm and deep concern about possible changes to TB 117. Nearly every fire-safety, fire-test-standards, and flame-retardant-materials expert C&EN has spoken with over the past few months has expressed anger and frustration with Blum’s campaign and what they say is a foolish drive to weaken the TB 117 standard because of chemophobia. The California fire standards for upholstered furniture work, they say, and the state’s own fire statistics since the rule went into effect back them up.

Indeed, Gordon H. Damant, the chemist who in the 1970s wrote the original version of TB 117, says, “There was a significant decline in furniture fires and subsequent deaths over the first 10-year period” of using flame-retardant chemicals.

In a paper commemorating the 20th anniversary of California’s flammability standards for upholstered furniture, Damant wrote that, in 1974, the year prior to implementation of TB 117, “there were about 2,500 upholstered-furniture fire incidents in California. The number of furniture fires had declined about 50% by 1987, and by 1991, the last year for which a full year of data were available, furniture fires had dropped to about 800” (Proc. Int. Conf. Fire Safety 1994,19, 1).

Annually, upholstered furniture is the number-one cause of death and injuries from fire in the U.S., says Damant, who is now retired from California’s Department of Consumer Affairs and runs his own fire-safety consulting firm in Sacramento. Some 30,000 people in the U.S. have died in furniture fires since 1972, he says. People who want to get rid of flame retardants “want to ignore all that. They are more concerned about what might happen” in terms of potential human health effects from exposure to the chemicals.

As a career public servant, Damant is also angered by the accusations that people who want to uphold or even strengthen TB 117 are somehow tainted by industry money. “I have seen lots of cases where many children die in fire,” Damant says, his voice rising. “I don’t see any recognition of that from people who want to get rid of flame retardants. They are not willing to look at the facts.”



This article has been sent to the following recipient:

Tom Muir (October 29, 2012 5:08 PM)
Quite a few things left out in this story.

1. You need to get the data on smoking cessation to see how it corresponds with the declines in furniture fires you point out. More generally, declines in smoking line up quite well with declines in residential furniture fires.

2. Time to flashover is a wrong and misleading metric of fire safety in a residential context. Most people who die or are injured in fires do not burn to death, but are asphyxiated by smoke and gases from the smoulder.

3. It is also misleading to point to different chemical structures as a dodge of the general toxicity issue. THere may be large differences in properties in some cases, but across the span of about 26 orders of magnitude in differences of properties, the six or so orders seen in properties of different chemical FR classes is relatively small.

Regardless of differences in properties, most of the known problem FRs have very similar health related effects in all studies including in vitro, in vivo, and epidemiology. New ones also have been shown to be similarily toxic.

4. THere are still lots of fires and furniture fires, despite the omnipresence of billions of pounds of FRs. If they are so effective, why are there still so many fires?

5. Who is complaining about FRs in airplanes?
gnelson (October 30, 2012 2:31 PM)
Mr. Muir's comment shows the need to come together on a common understanding of the fire problem.

Two-thirds of civilian fire deaths in the US are post-flashover outside the room of origin. People are dying outside the room of origin from toxic gases. These are not smoldering fires. A large fire scenario maximizes the impact of carbon monoxide, which is the primary toxic gas in fires. We have measured 10-50,000 ppm CO in the plume coming out of a room in flashover. Thus eliminating flashover and delaying flashover is key.
Helen Sullivan (October 31, 2012 12:04 PM)
The C&EN editorial changes, shown in the online version above, combined with the clarification in red at the top of the article do not sufficiently correct the erroneous and misleading statements in this article relating to the Polyurethane Foam Association.

For the record, the Polyurethane Foam Association has never provided any compensation or grants to Arlene Blum or the Green Science Policy Institute. Arlene Blum was paid a one-time honorarium to speak at a PFA conference in 2006. For the record, this was made clear to the reporter during the interview.

Another point in the article that requires clarification relates to the statement the polyurethane foam industry might see “cost-saving benefits from a TB 117 revision.” There is no evidence to support that statement. Since flame retardants are often used as replacement for other raw materials in formulation, when FR components are removed, there is little likelihood of economic benefit. There is no known economic motivation for foam manufacturers who support revision of TB 117. TB 117 is an outdated standard that needs review and improvement. PFA does not have a position on flame retardants. They are an important resource for foam manufacturers to meet customer specifications for certain applications. PFA is on record in support of a smolder standard for California and for the United States through the CPSC standards-making process. Once a smolder standard is in place, work can focus on development of a small open flame test standard that might be considered as an addition. Developing a small open flame standard that meets combined safety, testing reliability, reproducibility, economic, and general feasibility objectives will be challenging. This may take a little time, but could be a worthwhile effort in the end.
Helen Sullivan, PFA Communications Counsel
gnelson (November 1, 2012 1:36 PM)
2009 NFPA data for "Home Structure Fires That Began with Upholstered Furniture" show the following heat sources: Smoking materials 27%; Candle, lighter, or match 20%; Operating equipment 24%; Ember or ash 10%. Home Structure fire deaths that began with upholstered furniture show the following heat sources: Smoking materials 49%; candle, lighter or match 15%; operating equipment 15%; ember or ash, 12%. To establish a standard that focuses only on smoldering cigarettes covers only 27% of the fires and 49% of the deaths. To focus on smoldering only is clearly a weaker standard. An open flame test is needed day one, not some time in the future.
Bob Luedeka (November 1, 2012 2:30 PM)
Gordon, thank you for your response.

Quoting from "2008–2010 Residential Fire Loss Estimates", U.S. National Estimates of Fires, Deaths, Injuries, and Property Losses from Unintentional Fires," U.S. Consumer Product Safety Commission, September 2012

"With respect to item first ignited, upholstered furniture was involved in the greatest number of fire deaths. From 2008 through 2010, an estimated annual average of 450 deaths was associated with these fires. This constitutes 19.3 percent of the estimated annual average of total deaths associated with residential structure fires for the same period. On average, during 2008 to 2010, mattress or bedding ignitions accounted for an annual average of 320 deaths, which is 14.0 percent of the average annual estimated number of total residential fire deaths.

With respect to heat source, smoking materials were the largest contributor to deaths, associated with an annual average of 500 deaths from 2008 to 2010. This accounts for 21.5 percent of the estimated annual average of total residential fire deaths. The estimated annual average number of deaths from candle fires is 70, which represents 3.2 percent of the average annual estimated total number of residential fire deaths during 2008 to 2010. There were an estimated 50 deaths from lighter fires (2.1 percent of the estimated annual average of the total number of residential fire deaths) while, on average, matches were responsible for 20 deaths, or 1.0 percent of total deaths annually."

I think you will agree, that if we can raise the bar in regard to reducing the incidence of smolder-based ignition of furnishings, we've accomplished something important. And, because even one death is one death too many, creating a workable small-open flame standard may also be important, but what exists now is problematic. The component-based, small-open-flame requirement within the existing Calif. TB 117 testing standard needs careful review and possible revision.

Bob Luedeka
Executive Director
Polyurethane Foam Association
Vytenis Babrauskas (November 1, 2012 2:08 PM)
The Oct. 29, 2012 C&EN cover story Ablaze over Furniture Fires claims that “Fire-safety scientists are outraged by a chemist’s campaign against flame-retardant chemicals.” I am not only a fire-safety scientist, but the first-ever person to be awarded a Ph.D. degree in the field. And I am outraged at the halogenated- FR chemical makers’ campaign against Planet Earth. It is regrettable but true that the stewardship record of my profession has been poor. Substances such as asbestos, carbon tetrachloride fire extinguishers, PCB transformer oils, Halons, Freon refrigerants, and now a growing list of halogenated fire-retardant chemicals have been promoted as fire safety “solutions” despite the severe problems left in the wake of their use (it is not all FR chemicals that are deleterious, but primarily halogenated ones).

Of specific concern is that C&EN discussed the 1988 NIST study of which I was the lead author, misstated its implications, yet never contacted me. The specimens studied in that project were provided by industry as constituting the best commercially available materials. Consequently, they were highly FR treated and, essentially, failed to burn when ignited. The industry then has been trying to apply these findings to foams made to California’s 1975 TB117 standard. Such foams use a minimal (typically 4%) FR loading. This is categorically different from the kinds of materials I tested and it is erroneous to claim that my work shows the effectiveness of TB117 foams. Indeed, a couple of years earlier at NIST I specifically tested TB117 foams. That project demonstrated no fire safety benefit whatsoever from the low-concentration FR treatments used to meet the TB117 standard. The facts are clear: FR treatments can be highly effective against fire if used in large amounts, but are without benefit when used in low concentrations. In fact, Gordon Nelson, an industry spokesman C&EN interviewed, has published a paper showing that small amounts of FR loading increase the production of toxic chemicals (CO specifically) yet provide fire safety benefits that are trivial or non-existent.

Finally, the industry’s claim that removing the flaming-test provisions from TB117 will result in a “lowered” standard is untrue since the standard is already ineffective against flaming fires. My tests at NIST showed that, CPSC’s tests showed that, and furniture industry tests showed that. Even C&EN published a story recently (7/23/12) which concluded that “Flame retardants in furniture foam are not effective.” Contrary to the claims of Gordon Damant, the author of the TB117 standard, statistics do not
in any way show that TB117 was beneficial to fire safety in the State of California. California had an exceptionally good fire record before the current-era fire statistics started being available. But ever since that time, the statistics show that there was not any improvement in California’s performance vis-à-vis the rest of the nation.
William Schulz (November 1, 2012 3:22 PM)
As the story makes clear, C&EN did speak with the organizer of the 1988 NBS study, Richard Gann. In trying to contact Dr. Babrauskas, we were informed that he was out of the country and not available for comment until past our deadline. Gordon Nelson is an academic researcher--I interviewed him in person at his office and laboratories at the Florida Institute of Technology. He is a past president of the world's largest scientific society, the American Chemical Society, which to most people speaks volumes about his authority and very high level of professional reputation and integrity. The insistence by Babrauskas and others that he is an "industry spokesperson" has never been backed up with any evidence or facts. Moreover, I am not sure what Babrauskas thinks this label for Nelson would prove even if it was true. C&EN interviewed several "industry spokespersons" as credible sources for this story and all with a stake in this debate--most notably, perhaps, Polyurethane Foam Association Executive Director Bob Luedeka.
Gordon Nelson (November 2, 2012 2:07 PM)
Dr. Babrauskas continues to misrepresent the results from the paper in J. Fire Sciences [18, 430-455 (2002)]. That paper was studying polyurethane foams for transportation applications. Commercial FR loadings were hardly small. Flammability data included Cone Calorimetry data plus 12-second and 60 second Bunsen Burner vertical tests. Not all FRs work equally well. The best formulations showed over 50% reduction in Peak Heat Release Rates, and passed the Bunsen Burner tests, which are hardly trivial. Not surprisingly in the Cone there was an increase in CO with FR materials. CO generation is not material specific. In the real world CO is determined by the intensity of the fire. Big fires post flashover generate huge amounts of CO. The purpose of FR materials is to have fewer and smaller fires.

The actual California fire data should not be ignored. During the period 1974-1991 the California State Fire Marshall had its own fire incident database (CFIRS). In 1974 the year before implementation of the requirements there were 2500 upholstered furniture fire incidents in California. In 1987 the number of such fires had declined by 50% and by 1991, the last year of full data furniture fires had dropped to 800 or a 1974-1991 decline of over two-thirds. This was despite the fact that California was increasing in population at about 750,000 per year. Fire deaths in furniture fires from 1982-1991 declined from 60 to 10. In 1990 US CPSC estimated in the US 16,000 residential fires which originated in upholstered furniture resulting in 890 deaths. On a per capita basis California should have had 110 deaths rather than the 20 reported for 1990. In CFIRS data from 1997-2002 California had no deaths in the category "Upholstered Furniture and Structure Fires." During that period California had 226,000 fires (incomplete data), 50,000 structure fires, and 161 upholstered furniture fires in structures. That California has one of the best fire records in the US is no accident. To claim that TB117 has had no impact is bogus.

In the US about 10 people still die each week as a result of residential upholstered furniture fires. NFPA 2009 data indicate that 27% of fires and 49% of deaths are due to smoking materials. Testing only resistance to smoldering cigarettes misses 73% of the fires and 51% of the deaths. That is not a fire safety standard.
Susan Kegley (November 6, 2012 4:22 AM)
To Rudy Baum, Editor, C&EN News:

Most of the people who read C&EN news are scientists. We like the facts. And it appears that we were not given the facts in this article on flame retardants. Instead, we were deluged with innuendo, character assassination, emotional representations, and statements that are clearly not true (See the letters to the Editor, above, for some of the statements that were found to be incorrect).

Here are a few facts we'd like to have:

1) Why were all of the people who support the continued use of halogenated flame retardants described as "experts" calling for a more "reasoned" analysis, and those who question the need for these chemicals that don't appear to work as flame retardants (see Babrabuskas's comment above) and have toxicity characteristics that should give us pause, framed as "visibly angry", "alarmist", "speculative", "going in the wrong direction," "foolish," having "chemophobia," with "conspiracies" and "allegations" "having poor or deliberately misleading understanding of chemistry." The article was rife with words that we don't usually see in C&EN News. Why, Mr. Baum, did you allow publication of this article? Where did C&EN's usually excellent reporting go, where good and evil are not attributed, but the writer presents the information known about the topic and lets the facts speak for themselves?

2) Can you please tell us a bit about Gordon Nelson's affiliation and/or sources of funding for his research? Does he take grants from the manufacturers of flame retardant chemicals? How about Damant? Hirschler? If you are going to try to make Dr. Blum look like she has a conflict of interest for accepting a "grant" from the foam industry (but not really---see the correction issued above), at least provide the same information about the other people you quote.

3) Schultz notes that the rate of furniture fires declined about 50% between 1974 and 1987. Can you tell us how the rates of smoking changed over that same time period?

4) It is ironic that this article cast aspersions on a technique organic chemists have been using for years to assess reactivity of similar molecules--comparative structure analysis. Indeed, we now have software that will conduct this analysis for us---and it is widely used in drug development and toxicology. Why, then does Hirschler believe that it is incorrect to use QSAR in the case of flame retardants?

Overall, this article was very carefully crafted to cast doubt on Blum's work, indeed on anyone who might even dare to think that the 10 pounds of halogenated flame retardants in most homes is not a great idea. But guess what? The audience you write for is not stupid, and will see the article for what it is---a hit piece. Please, Mr. Baum, keep these kinds of articles out of the ACS news magazine.

Susan Kegley, PhD
Principal and CEO
Pesticide Research Institute
Professor Miriam Diamond (November 16, 2012 10:47 AM)
I am astonished and dismayed by the lack of critical analysis and bias shown in the article “Fighting Fires” and “Ablaze over furniture fires” by William G. Schultz. The article appears to be a rebuttal to that in the Chicago Tribune rather than one aimed at informing science-based discussion. Instead of arguing the science behind the issue, you conducted a character assassination of Dr. Arlene Blum and other scientists who oppose the views expressed in the article. You cite Dr. Gordon L. Nelson without disclosure of his research funding sources. To say, as William Schultz has in response to another post, that Dr. Nelson’s credentials are proof enough of the scientific worth of his views, counters any rules or logic of “conflict of interest”. Instead, you raised “conflict of interest” allegations against Dr. Blum which were in error. Dr. Nelson asserts that Dr. Blum’s positions are false and “also an affront to the many scientists whose thousands of peer-reviewed papers prove conclusively that the chemicals do work and that they have save thousands of lives”. Your article is an affront to many scientists, including as myself, whose similarly numerous peer-reviewed papers document the local-to-global distribution of an increasing number of brominated flame retardants and cause-effect linkages between specific FRs and a range of adverse health effects in humans and the ecosystem.
At the last ACS meeting in San Diego (March 2012), I attended part of the Fire Safety Symposium chaired by Dr. Nelson. I was astounded by the “discussion” moderated by Dr. Nelson that quickly descended into angry argument after presentations made by two fire scientists who questioned aspects of the effectiveness of gas-phase additive flame retardants (such as the brominated flame retardants). There was no opportunity for scientific debate as the discussion from members of the audience, including Gordon Nelson and Marcello Hirschler, was rancorous and dismissive of points made in those presentations.
Several times the article refers to the ample evidence that flame retardants are effective. Schultz cited studies showing that adding flame retardants to products reduces their flammability. The central question is, however, whether the flammability standards such as California TB 117 can be credited with reducing fire deaths. Is there a cause-effect link between flammability standards such as TB 117 and reduced fire deaths? Perhaps such evidence exists, but neither your article nor my search of the literature found studies to support this view. I believe that we need an independently-conducted analysis of fire statistics to determine the role played by flammability standards before we can assume, as is written in the article, that weakening flammability standards will necessarily compromise public safety.
In closing, your article further polarizes an already highly polarized debate. The character assassination of Blum is unprofessional and a cheap tactic to undermine her and those of us (also a member of the ACS) who take a position counter to what you have argued. I expect balanced reporting and informative discussion from C&EN, not biased reporting.
Arlene Blum (November 25, 2012 10:23 PM)
We are glad C&EN decided to cover this important issue in depth and have been impressed in the past with its balanced and accurate coverage. However, we are very surprised by this article, which has multiple inaccuracies and does not present a balanced analysis of the complex issues involved:

1. This article states that the majority of fire scientists agree that California Technical Bulletin 117 (TB117) is an effective standard and that there is considerable data to this effect. However, numerous sources (1–4) including C&EN’s own previous coverage, not referenced in this article, find the TB117 standard is not effective: “Flame Retardants In Furniture Foam Are Not Effective, CPSC And UL Officials Tell Senate Subcommittee” July 23, 2012.
2. Babrauskas, et al 1988 (5) is cited in the Schulz piece as showing TB117 is effective. However Dr. Babrauskas has previously stated that these experiments are not applicable to TB117 furniture and that an interpretation to that effect is inaccurate and a distortion of his results.
3. The articles also imply that concerns about health and environmental effects of flame retardant chemicals are “alarmist,” “chemophobic,” “misguided,” and not based on scientific evidence. This is also not the case, as over 200 scientists have signed the San Antonio Statement, which reviews a body of data and states: “Therefore, these data support the following: Brominated and chlorinated flame retardants as classes of substances are a concern for persistence, bioaccumulation, long-range transport, and toxicity.” (6)
4. The issue of fire toxicity is dismissed when in fact there are many studies showing that, depending on the combustion conditions, products containing flame retardants may produce greater amounts of carbon monoxide, soot, smoke, and halogenated dioxins. (7–11)
5. For upholstered furniture fire deaths, NFPA data shows that those caused by smoking materials have significantly declined since 1980, while there is no trend for those caused by small open flames. As TB117 addresses fires started by small open flame ignitions, not smoking materials, it is seems flawed to attribute the decline in total furniture fire deaths in California to TB117. Neither NFPA nor CPSC statistical analysis mentions TB117 or flame retardants as contributors to the decline in fire deaths from smoking materials. (3,12)
6. Several of the scientists interviewed (Matthew Blais, Gordon Nelson and Marcelo Hirschler) have financial conflicts of interest which were not disclosed, as is standard practice. (13-15)

For the above reasons, we would like to voice our disappointment with quality of the reporting in these articles and register our strong disagreement with their conclusions.

Arlene Blum , Visiting Scholar, Chemistry, University of California, Berkeley
Executive Director, Green Science Policy Institute
Craig Criddle, Professor of Civil and Environmental Engineering, Stanford University
Miriam Diamond, Professor of Geography, and Chemical Engineering and Applied Chemistry, University of Toronto
Rachel Morello-Frosch, Associate Professor of Environmental Science, Policy and Management, UC Berkeley
Mary Turyk, Research Assistant Professor of Epidemiology, University of Illinois at Chicago
Alastair Isle, Assistant Professor of Environmental Science, Policy and Management, UC Berkeley
Lynn Huntsinger, Professor of Rangeland Ecology and Management, UC Berkeley
Susan D. Shaw, Director, Marine Environmental Research Institute
Michelle Douskey, Lecturer of Chemistry, UC Berkeley
Margarita C. Curras-Collazo, Associate Professor of Neuroscience, UC Riverside

1. Babrauskas V. Upholstered Furniture Heat Release Rates: Measurements and Estimation. Journal of Fire Sciences. 1983 Jan 1;1(1):9–32.
2. Talley TH. Phases 1&2, UFAC Small Open Flame Tests and Cigarette Ignition Tests. In: Annual AFMA Flammability Conf. 1995.
3. Medford RL. Upholstered Furniture Flammability: Fires Ignited by Small Open Flames and Cigarettes. 1997; U.S. Consumer Product Safety Commission
4. Mehta S. Upholstered Furniture Full Scale Chair Tests – Open Flame Ignition Results and Analysis. 2012; U.S. Consumer Product Safety Commission
5. Babrauskas V, Harris R, Gann RG, Levin B, Lee BT, Peacock RD. Fire hazard comparison of fire-retarded and non fire-retarded products. NBS Special Publication. 1988;SP749.
6. DiGangi J, Blum A, Bergman A, de Wit CA, Lucas D, Mortimer D, et al. San Antonio Statement on brominated and chlorinated flame retardants. Environmental health perspectives. 2010 Dec;118(12):A516–8.
7. Nelson GL, Sorathia U, Jayakody C, Myers D. Fire-Retardant Characteristics of Water-Blown Molded Flexible Polyurethane Foam Materials. Journal of Fire Sciences. 2000 Nov 1;18(6):430–55.
8. Babrauskas V. Plastics, part B, The effect of FR agents on polymer performance. In: Babrauskas V, Grayson SJ, editors. Heat release in fires. London: Elsevier Applied Science Publishers; 1992. p. 423–46.
9. Stec AA, Hull TR. Assessment of the fire toxicity of building insulation materials. Energy and Buildings. 2011 Feb;43(2-3):498–506.
10. Ebert J, Bahadir M. Formation of PBDD/F from flame-retarded plastic materials under thermal stress. Environment international. 2003 Sep;29(6):711–6.
11. Weber R, Kuch B. Relevance of BFRs and thermal conditions on the formation pathways of brominated and brominated-chlorinated dibenzodioxins and dibenzofurans. Environment international. 2003 Sep;29(6):699–710.
12. Ahrens M. Home Fires That Began with Upholstered Furniture. 2011; National Fire Protection Association
13. According to his LinkedIn profile, Dr. Matthew Blais is a science advisor for the North American Flame Retardant Alliance.
14. At 1 hour, 17 minutes of this video from a hearing on June 26, 2012, in Sacramento, CA, it is stated for the record that Dr. Gordon Nelson receives “a stipend for his time and his travel expenses” from the Citizens for Fire Safety Institute (CFFSI), an industry group funded by Albemarle, Chemtura, and ICL Ltd, the three largest makers of brominated flame retardants.
15. In a May 1, 2005, commentary in Flame Retardancy News titled “Limited Combustible Materials” it is noted that Dr. Marcelo Hirschler represents the American Fire Safety Council, an organization which promotes the use of flame retardants.
Gordon Nelson (November 28, 2012 4:51 PM)
Given that I am named in the last two posts, a response is required. It is unfortunate that one needs to question motives rather than dealing with a discussion of comparative risk of fire safety and environmental health, and how to maximize both fire safety and health. But so be it.

Given the challenge, my research is funded by a federal agency. Unlike those signed above, my work deals with the fire behavior of plastics, including polyurethanes. I have been looking at nanocomposites and other non-halogen approaches (phosphorus) to flame retardancy, approaches which achieve 50-70% reduction in Peak Rates of Heat Release. Approaches include bound in flame retardants which will preclude FRs leaching into the environment. Relevant new approaches to flame retardancy do exist.

I have a day job so little of my income comes from consulting. I have over the years been an expert in testimony in legal and regulatory contexts. I would expect to continue to do so. My expert testimony is based upon my experience in the fire behavior of plastics since the early 1970's, regardless of the source of support for time and travel. I have not been asked to do otherwise.

That said the post by Dr. Baum and colleagues raises a couple of questions. Dr. Baum lists herself as Visiting Scholar, Chemistry, UC Berkeley, yet the College of Chemistry website Directory does not list her, despite the fact that such positions are listed. An alumni news item notes, "Baum, a former visiting scholar in the Tinico lab in the College of Chemistry," is dated January 18, 2011. The posts above note that Dr. Baum has received lecture fees from the Polyurethane Foam Association. The Green Science Policy Council is a project of another organization so that its finances are not publicly disclosed. The question of what funding comes from commercial interests is a fair question, given her admonition to others to disclose "financial conflicts".

Evidence is asked by Dr. Diamond for the effectiveness of TB-117. Yet a reference is provided on p33 of the article. That article by Gordon Damant reviews the successes of TB-117 from 1974 to 1994. During that period upholstered furniture fire incidents in California dropped from 2500 to 800. Deaths dropped 1982-1991 from 60 to 10. This is despite California growing at 750,000 people per year. Damant and colleagues conducted hundreds of full scale furniture burns. TB-117 is based upon a large data base of full scale furniture fires. Numerous published papers exist. To ignore that data gives bogus conclusions.

Dr. Blum has repeatedly misrepresented the paper in her reference 7. One is not told about the context of the data. Cone tests were conducted at 25 and 35kW/m2 external flux. The best formulations showed over 50% reduction in Peak Heat Release Rates and passed open flame tests. Not surprisingly in the Cone there is an increase in CO with FR materials. Fire scientists know that CO generation is not materials specific. A big ventilation controlled fire generates huge CO concentrations. We have done 500 large scale fire tests. CO concentrations of 10-50000 ppm CO were observed in the smoke plume from room flashover fires. The issue is preventing fire, making fires smaller, and delaying a big fire. Making fires less toxic is a fool's game.

Reference 5 is said to be cited by Schulz to show TB-117 effectiveness, when the citation is used to address the more general question of "flame retardant materials work when properly used" (p30).

NFPA 2009 data are clear: 27% of upholstered furniture fires and 49% of deaths are due to smoking materials. That leaves 73% of upholstered furniture fires and 51% of deaths due to a different ignition source. Testing only smoldering cigarette performance as proposed in California is not a fire safety standard.

The Schulz article gave Dr. Baum a lot of column inches. But there are disagreements with Dr. Baum's conclusions in the fire science community. One needs to get both fire data and environmental/health data on the table for rational discussion. Clearly the Chicago Tribune chose not to do that. The regulatory processes to come will require it. The issues are complex. I for one hope that a full scientific discussion will occur, rather than what occurred at the Sacramento hearing mentioned on June 26th.

The ACS Symposium Series Volume 1118 "Fire and Polymers VI: New Advances in Flame Retardant Chemistry and Science" will issue shortly. There are 31 peer reviewed chapters. New approaches exist, flame retardants and flame retardant materials do work.

Gordon Nelson (November 29, 2012 9:56 AM)
Paragraphs 4 and 9 contain typos: Dr. Baum = Dr. Blum. It was hard to write with two air conditioning technicians (Florida)working with a ladder next to your desk. Also I had just returned from a meeting involving a colleague, Dr. Baum. Nonetheless, one hopes that in the processes to come, the discussions will be science based, rather than what has occurred at hearings so far.
Professor Miriam Diamond (December 6, 2012 2:22 PM)
Re: Evidence of the effectiveness of TB 117
Dear Dr. Nelson,
Thank you for pointing out data that clearly show the drop in upholstered furniture fire incidents since 1974 and that coincide with the implementation of TB 117. However, this downward trend in the fire data to not provide a cause-effect explanation for the trend. In other words, how do we know that TB 117 was responsible for the decline in upholstered furniture fire incidents?
The fire statistics I have seen (and I fully disclose that I am not an expert in this area) show decreases in fire incidences in developed countries worldwide over this time period. Other explanations for the cause of this decreased incidence could be the decline in rates of cigarette smoking, improved sprinkler systems, widespread use of smoke detectors and improved building codes.
What is the evidence for a clear cause-effect relationship between the decrease in upholstered furniture fire incidents and TB 117?
Yes, let's keep the conversation to science and evidence.
Prof. Miriam Diamond
Gordon Nelson (December 13, 2012 4:15 PM)
Over the period since 1974 there has been a decline in smoking. If one looks at smokers in California the big impact is the amount smoked, but less so on the number of people who smoke (fires involving death are not equally spaced across the day). Residential sprinklers are of minor use. Some 60% of fire deaths occur in buildings without operating smoke detectors. Codes have changed. There are also smolder resistant cigarettes. All of these measures go in the right direction. And California has TB 117 and California has the 5th best fire record in the US, despite large numbers of people living in poverty.

Fire kills the young, the old, the drunk, and the poor, not a random sample of the population. Thus having a layered approach to safety is important.

In 2011 California had 406 upholstered furniture fires with 24 injuries and 5 deaths. From 1/1/2003 to 12/31/2011 there were 4494 incidents, 273 injuries and 49 deaths. In the US there were 450 deaths in 2009 alone from fires that began in upholstered furniture (Ahrens Table 12 8/2011). Clearly California is a special place.

From the recent Heather Stapleton paper in ES&T, 81% of furniture outside of California contains flame retardants, and over 90% of furniture complies with UFAC on cigarette ignition. Thus it is not surprising that performance nationally is beginning to look more like that of California. The UK (also Ireland) has a tougher standard on upholstered furniture than California, yet FR compounds in dust are low. For the period 2002-07 the UK standard has resulted in 54 fewer deaths per year and 1065 fewer fires each year(BIS report 12/09).

Bottom line. One needs layered safety requirements. To say that TB-117 has had no impact does not make sense. That there are other (layered) measures that help is true and part of the needed mix for fire safety success.

The hundreds of full scale tests done by the Bureau in the implementation of TB-117 show that TB-117 has impact. A larger ignition source is needed for ignition and time to flashover takes longer versus non-TB-117 compliant furniture.

Again, nationally only 27% of upholstered fires and 49% of deaths are due to smoking materials. California is the defacto national standard. A cigarette only regulatory test in not adequate.

Leave A Comment

*Required to comment