Two sofas can sit side by side in a showroom, built from the same foam and wrapped in the same fabric, and be chemically different products. The one made before 2014 was probably treated to resist a small open flame. The one made after probably was not. Nothing you can see tells you which is which.
The dividing line is a California rule most Americans have never heard of, which for decades governed what went into furniture sold nationwide. It changed in 2013. It became federal law in 2021. Almost nobody noticed either event.
The dates on this subject are garbled almost everywhere you look online, including in places that ought to know better. So the ones below come straight from the rule texts and the Federal Register notices, and the reason that matters is practical: the age of a piece of furniture is the single best clue you have about what is in it.
Why furniture foam got treated in the first place
Flame retardants did not arrive in American living rooms by accident. They arrived because of a performance test.
California Technical Bulletin 117 was, in the words of a 2008 Consumer Product Safety Commission rulemaking document, the mandatory regulation for all upholstered furniture sold in that state. The same document says it contained both smoldering and small open flame resistance performance requirements, and that California furniture was typically made with flame-retardant resilient foam filling.
That is the mechanism, and it is worth sitting with. California never required a chemical. It required foam that could pass a test. Treating the foam was the cheapest way for a manufacturer to pass it, so that is what manufacturers did.
The second half of the mechanism is why this was a national story. Furniture companies do not build one sofa for California and a different one for Ohio. A state representing a large share of the market effectively wrote the recipe for the country. That same 2008 document notes that pentaBDE, a brominated flame retardant, had previously been used to meet the open flame requirement.
What changed in 2013, and when it became federal law
The replacement standard is TB 117-2013, published in June 2013. Its full title runs to a mouthful about requirements, test procedure and apparatus for testing the smolder resistance of materials used in upholstered furniture. The title is the whole story. Smolder resistance, not open flame.
A smolder test asks whether a lit cigarette dropped on a cushion will start a fire. An open flame test asks whether a match held to the foam will. They are different hazards and they call for different engineering. The 2013 standard applies its test to cover fabrics, barrier materials, resilient filling and decking rather than treating the foam as the thing that has to resist a flame.
The compliance dates are where accounts go wrong. The standard was published in June 2013. California’s own regulations, in Title 4 of the California Code of Regulations, required filling materials and cover fabrics in upholstered and reupholstered furniture to meet TB 117-2013 from 1 January 2015. A sofa built in 2013 and a sofa built in 2016 are on opposite sides of a two-year transition, not a single switch.
Then the part that genuinely surprises people. There is now a federal upholstered furniture flammability standard, and it is the California one. Section 2101 of Public Law 116-260, the COVID-19 Regulatory Relief and Work From Home Safety Act, signed on 27 December 2020, made TB 117-2013 a flammability standard under the Flammable Fabrics Act. The CPSC codified it at Part 1640 of Title 16 of the Code of Federal Regulations by a direct final rule published on 9 April 2021. It took effect on 25 June 2021 for furniture manufactured, imported or reupholstered on or after that date. Mattresses, foundations, bedding and exercise furniture are outside it.
The CPSC’s own estimate in that rule was that up to 95 percent of upholstered furniture already met TB 117-2013, which is why Congress could make it national without disrupting the market.
Meanwhile the agency’s own attempt at a furniture standard ended. CPSC opened that rulemaking in June 1994, held a hearing on the possible toxicity of flame-retardant chemicals in March 1998, proposed a rule in March 2008 that would have required cigarette-resistant fabrics or barriers rather than flame-resistant chemicals, and published a notice terminating the whole thing on 16 September 2021. Twenty-seven years of work, closed out because Congress had legislated the answer.
What the labels on a sofa tell you
Since 25 June 2022, new upholstered seating sold in the United States has to carry a permanent label with one sentence on it. The CPSC’s regulation prescribes the wording: “Complies with U.S. CPSC requirements for upholstered furniture flammability.” That label is the manufacturer’s certification that the product meets the standard.
Here is what it does not say. It does not tell you whether flame retardant chemicals are in the foam. It is a statement about a fire test, not about chemistry.
California keeps a separate labeling section of its own, which the federal law specifically preserved from preemption. The federal rule names it but does not spell out what it requires, and California’s regulatory text would not load for this article, so this piece will not characterize its contents. If you are shopping and the chemistry matters to you, the honest advice is to ask the retailer in writing and treat the answer as a supplier claim rather than a certification.
A short field guide for furniture you already own:
- Made before 2014: most likely treated, because the open flame requirement was still in force.
- Made 2015 or later and sold in the United States: most likely not treated, because the standard it had to meet no longer asks the foam to resist a flame.
- Made in 2013 or 2014: genuinely uncertain, sitting inside the transition.
- Reupholstered at any point: depends entirely on the foam that went in, not on the frame’s age.
- Bought secondhand: the age of the piece, not the date you bought it, is what counts.
Flame retardant is a job description, not a chemical family
This is the idea the whole subject turns on, and almost every article about flame retardants violates it in the first paragraph.
A flame retardant is defined by what it does, which is interfere with combustion. The chemicals that do it are not relatives. Some are brominated, some chlorinated, some are organophosphate esters, some are mineral fillers with no organic chemistry at all. They differ in how readily they leave a product, how long they persist, how they behave in a body, and what they do there.
The National Academies put a number on this. In a 2019 report requested by the CPSC, a committee examined whether organohalogen flame retardants could be assessed as a single class. The answer was no. The committee identified 14 subclasses, sorted by structure, physical-chemical properties and predicted biological activity, across 161 organohalogen flame retardants plus more than 1,000 chemical analogs. That is 14 groupings inside just the halogen-containing corner of the field.
So any sentence that begins by saying flame retardants cause something is almost certainly wrong, in the same way that calling metals toxic is wrong. The question is always which one, at what dose, by which route.
Where these chemicals still are
Furniture is where the story started, which is why it dominates coverage. It is no longer where most of the remaining exposure lives.
Electronics. Plastic casings around televisions, monitors, computers and power supplies get flame retardant treatment because a device that can overheat sits plugged in all day. This was one of the four product groups named in the CPSC petition described below.
Building insulation. Hexabromocyclododecane, abbreviated HBCD, is used mainly as a flame retardant in building materials including insulation. EPA completed a risk evaluation under the Toxic Substances Control Act in September 2020. Its conclusions are worth reading precisely, because they are not what either side of the argument usually reports: EPA found unreasonable risk to the environment in half of the conditions of use it examined, and unreasonable risk to workers and nearby workers from the use and disposal of HBCD in building and construction materials. It did not find unreasonable risk to the general population or to consumers.
Cars. Federal Motor Vehicle Safety Standard 302 requires materials in the occupant compartment, including seat cushions, headlining, trim panels, floor coverings and sun visors, to burn no faster than 102 millimeters per minute. The standard sets a performance requirement, not an ingredient list, but it is why car interiors are a different chemical environment from a living room. A 2025 study in Environmental Monitoring and Assessment measured 30 dust samples from cars of a single make and model built between 1996 and 2021. Flame retardants turned up in every sample, with some compounds consistently above a microgram per gram and substantially higher than levels found in other indoor settings. Concentrations also varied enormously between cars of the same model, which means generalizing from any one measurement is a mistake.
Children’s products. A 2011 study in Environmental Science and Technology analyzed foam from 101 commonly used baby products. Eighty contained an identifiable flame retardant additive, and all but one of those was chlorinated or brominated. That survey describes the old regime, before the standard changed, which is exactly why older hand-me-downs deserve more attention than new purchases.
Older furniture, anywhere it ends up. A treated cushion does not become untreated when it moves to a basement, a dorm room or a garage workshop.
What the health evidence supports
Report this at the strength it has, which is uneven across the field.
The best-studied group is the PBDEs, the brominated compounds that were used to meet the old open flame requirement. EPA states that US manufacture and import of pentaBDE and octaBDE were phased out in 2004, and that the main makers and importers of decaBDE committed to ending all sales by 31 December 2013. EPA identifies neurobehavioral effects as the critical human health endpoint of concern, and notes that because these compounds are not chemically bound to the products holding them, they leach out.
On the human side, a 2018 systematic review in the International Journal of Environmental Research and Public Health examined 16 published studies that measured PBDEs in maternal or cord blood and then tested children’s motor, cognitive or behavioral development. Most of those studies pointed toward an adverse association. The authors were careful about what that does and does not establish: the mechanism is unsettled, possibly running through thyroid function, and further work is needed. That is preliminary-to-moderate human evidence with a plausible mechanism, not an established causal finding.
For the organophosphate esters that replaced many brominated compounds, a 2021 critical review in Science of the Total Environment estimated exposure across all pathways and concluded that summed intakes fall within reference dose values. The authors immediately qualified it: their estimate does not include high-end exposures, and for highly exposed individuals the margin is smaller. A 2018 study of car dust from 25 private vehicles reached a similar place, finding toddlers more exposed than adults, with intakes still several orders of magnitude below the relevant reference doses.
Pull those together and the honest summary is this. Detection is established. Persistence and migration out of products are established. Neurodevelopmental concern for the older brominated compounds is supported by consistent but observational human evidence. Population exposure to the current replacements appears to sit below the benchmarks regulators use, with real uncertainty at the high end and for small children. Nothing in that paragraph is a reason for panic and nothing in it is a reason to stop looking.
Articles 718 and 719 cover how to read claims of this kind in general, and article 707 explains why two agencies assessing the same chemical can answer different questions and reach different conclusions.
The federal petition, and exactly where it stopped
People often say the CPSC banned organohalogen flame retardants. It did not, and the precise stage it reached is checkable.
On 30 June 2015, Earthjustice and the Consumer Federation of America petitioned the Commission, joined by organizations including the American Academy of Pediatrics. They asked the CPSC to declare four categories of product containing additive organohalogen flame retardants to be banned hazardous substances under the Federal Hazardous Substances Act: children’s products other than car seats, residential upholstered furniture, mattresses and mattress pads, and the plastic casings around electronics.
On 20 September 2017 the Commission granted the petition, voted to begin rulemaking, and directed staff to convene a Chronic Hazard Advisory Panel. Eight days later it published a guidance document in the Federal Register, on a 3 to 2 vote, asking manufacturers to stop intentionally adding these chemicals to those products and asking retailers and distributors to seek assurances from their suppliers.
The guidance states in its own footnote that it is “not a binding or enforceable rule” and changes nobody’s legal obligations.
What followed is as informative as what happened. The National Academies delivered the class-assessment report in 2019, concluding the chemicals could not be treated as one class. EPA added 30 organohalogen flame retardants to its health and safety data reporting requirements in 2021, explicitly because the CPSC needed data and because preliminary searches had turned up little or no information for many of the 161 compounds. No final CPSC rule banning them appears in the Federal Register.
So the accurate statement is that the petition was granted, a voluntary guidance document was issued in 2017, the scientific groundwork is still being assembled, and the rulemaking has not produced a rule.
Replacing the sofa is the expensive answer
A new sofa costs real money. It is also, for most households, the least efficient way to spend it on this problem.
These chemicals are not bound into the plastic or foam. They work loose over years and end up in house dust, which is how they reach people. Article 689 covers that pathway and the cheap habits that reduce it, including damp dusting and taking shoes off at the door, and that is where the better evidence sits.
Replacement does work. A 2021 study in Environment International collected dust from homes before and after residents swapped out older upholstered furniture, then returned at six, twelve and eighteen months. All seven flame retardants measured fell in nearly all homes. Importantly, the drop was significant whether households replaced whole pieces or only the foam inside them, and the authors note foam replacement as the more economical option that also generates less waste.
So the sensible order of operations, cheapest first:
- Control the dust, which costs nothing and addresses every flame retardant in the house at once, not just the ones in the sofa. Article 689 has the specifics.
- Wash hands before eating, especially for small children, who get more of their exposure through hand-to-mouth contact with dust than adults do.
- Vacuum and air out a car’s interior rather than ignoring it, since the measured dust concentrations there run higher than in homes.
- Replace the foam, not the furniture, if you decide to act on a specific older piece.
- Buy new only when you were going to anyway, at which point the standard has already done the work for you.
One place to involve a professional rather than a shopping list: if your work involves cutting, demolishing or recycling insulation, electronics or foam, that is an occupational exposure, not a household one, and EPA’s own risk evaluation identified unreasonable risk to workers handling HBCD in building materials. Raise it with an occupational health clinician or your employer’s safety program, where exposure can be measured. If a child swallows foam, a fragment of a product or any household chemical, call Poison Help at 1-800-222-1222, staffed 24 hours a day, and call 911 for breathing trouble or loss of consciousness.
This whole subject reduces to a date. If your furniture was made in 2015 or later and sold in the United States, the standard that drove flame retardants into foam no longer applied to it. If it was made before 2014, assume it was treated, and spend your effort on dust rather than on a credit card.
Sources
- U.S. Consumer Product Safety Commission (Federal Register). “Standard for the Flammability of Upholstered Furniture, direct final rule, 9 April 2021.” federalregister.gov
- Electronic Code of Federal Regulations. “16 CFR Part 1640, Standard for the Flammability of Upholstered Furniture.” ecfr.gov
- U.S. Consumer Product Safety Commission (Federal Register). “Standard for the Flammability of Residential Upholstered Furniture, proposed rule, 4 March 2008.” federalregister.gov
- U.S. Consumer Product Safety Commission (Federal Register). “Guidance Document on Hazardous Additive, Non-Polymeric Organohalogen Flame Retardants in Certain Consumer Products, 28 September 2017.” federalregister.gov
- U.S. Consumer Product Safety Commission (Federal Register). “Petition Requesting Rulemaking on Products Containing Organohalogen Flame Retardants, 19 August 2015.” federalregister.gov
- U.S. Consumer Product Safety Commission (Federal Register). “Semiannual Regulatory Agenda, 31 January 2022, entry 468 Flammability Standard for Upholstered Furniture.” federalregister.gov
- U.S. Environmental Protection Agency (Federal Register). “Health and Safety Data Reporting; Addition of 20 High-Priority Substances and 30 Organohalogen Flame Retardants, 29 June 2021.” federalregister.gov
- U.S. Environmental Protection Agency. “Polybrominated Diphenyl Ethers (PBDEs).” epa.gov
- U.S. Environmental Protection Agency. “Risk Management for Cyclic Aliphatic Bromide Cluster (HBCD).” epa.gov
- Electronic Code of Federal Regulations. “49 CFR 571.302, Standard No. 302, Flammability of interior materials.” ecfr.gov
- National Academies of Sciences, Engineering, and Medicine. “A Class Approach to Hazard Assessment of Organohalogen Flame Retardants (2019).” nationalacademies.org
- PubMed. “Rodgers KM, et al. "Do flame retardant concentrations change in dust after older upholstered furniture is replaced?" Environment International, 2021. DOI: 10.1016/j.envint.2021.106513.” doi.org
- PubMed. “Stapleton HM, et al. "Identification of flame retardants in polyurethane foam collected from baby products." Environmental Science & Technology, 2011. DOI: 10.1021/es2007462.” doi.org
- PubMed. “Gibson EA, et al. "Effects of Polybrominated Diphenyl Ethers on Child Cognitive, Behavioral, and Motor Development." International Journal of Environmental Research and Public Health, 2018. DOI: 10.3390/ijerph15081636.” doi.org
- PubMed. “Gbadamosi MR, Abdallah MA, Harrad S. "A critical review of human exposure to organophosphate esters with a focus on dietary intake." Science of the Total Environment, 2021. DOI: 10.1016/j.scitotenv.2020.144752.” doi.org
- PubMed. “Svobodova P, et al. "High levels of flame retardants in vehicle dust indicate ongoing use of brominated and organophosphate flame retardants in vehicle interiors." Environmental Monitoring and Assessment, 2025. DOI: 10.1007/s10661-025-13822-z.” doi.org
- PubMed. “Christia C, et al. "Legacy and emerging organophosphorus flame retardants in car dust from Greece: Implications for human exposure." Chemosphere, 2018. DOI: 10.1016/j.chemosphere.2017.12.132.” doi.org