The Black Plastic Spatula Scare, Revisited

In October 2024, a lot of people threw away a spatula. The trigger was a study in the journal Chemosphere reporting that black plastic kitchen tools carried flame retardants that had been banned for years, and the coverage was fast, loud, and nearly unanimous: get it out of your drawer.

Two months later the authors published a correction. One of the study's central numbers was off by a factor of ten, in the direction that makes the risk smaller. Then the story flipped, and the internet decided the whole thing had been nonsense. That's also wrong. Here's the version with the arithmetic in it.

What the study measured, precisely

The research came from Megan Liu at the advocacy group Toxic-Free Future with co-authors at Vrije Universiteit Amsterdam. The team gathered 203 black plastic household products sold in the US, most of them kitchenware, and screened all of them for bromine, which is a cheap proxy for brominated flame retardants.

Then, and this is the part the headlines flattened, they took the 20 products with the highest bromine readings and ran those through mass spectrometry. Flame retardants turned up in 17 of the 20. That's where the widely repeated "85 percent" came from. It is 85 percent of the 20 worst offenders, not 85 percent of everything on the shelf. Plenty of outlets, and plenty of brands selling alternatives, still quote it the other way.

The individual results were genuinely ugly. One black plastic sushi tray contained 11,900 parts per million of decabromodiphenyl ether, decaBDE, a flame retardant the EPA banned in 2021. Total flame retardant concentrations ran as high as 22,800 ppm in the worst samples. A spatula and a child's beaded necklace were also near the top.

The multiplication error that flipped the headline

To turn contamination into risk, the authors estimated how much decaBDE a person might ingest from a hot spatula, then compared it to the EPA's reference dose. They put estimated daily intake at about 34,700 nanograms.

The reference dose for BDE-209 is 7,000 nanograms per kilogram of body weight per day. For a 60 kilogram adult, that's 420,000 nanograms. The paper printed 42,000. Sixty times seven thousand is not forty two thousand, and that single dropped zero is the whole scare: it turned an exposure that sits comfortably below the threshold into one that appeared to be brushing up against it.

Joe Schwarcz at McGill University's Office for Science and Society caught it and wrote it up. The authors published a corrigendum on December 15, 2024, acknowledging the error and stating that it did not change the paper's overall conclusion. Retraction Watch covered the correction, and then covered a second one after further problems surfaced. The industrial chemist Mark Jones argued publicly that even the corrected version understated how much of the paper needed rewriting.

There's a further wrinkle worth knowing. That same December, Clarivate removed Chemosphere from the Web of Science index for failing to meet its editorial quality standards, after the journal retracted a batch of unrelated papers. None of that proves this particular study was wrong. It does mean the peer review that would normally catch a multiplication error was, at that moment, not in great shape.

What a reference dose is, and isn't

An EPA reference dose is not a poison line. It's a daily intake estimated to carry no appreciable risk over a lifetime, and it's built with safety factors stacked on top of the animal or human data, often ten-fold or hundred-fold. Coming in under it isn't a narrow escape. Coming in slightly over it isn't a diagnosis.

So the corrected picture is this: the estimated daily dose from the most contaminated black plastic products found in the study is roughly one twelfth of a threshold that already has a large margin baked in. That's the number. It is not zero, and we're not going to pretend it's zero, because "no detectable exposure" is a standard essentially nothing in a modern kitchen meets.

The supply chain problem nobody corrected

Here's what survived the correction fully intact. Flame retardants have no business being in a spatula, and the reason they're there is recycling.

Electronics casings are loaded with brominated flame retardants by design, because a television that catches fire is a worse outcome than a television with additives in it. When that plastic is recovered and reprocessed, the additives ride along. Black plastic makes it worse in a specific mechanical way: most recycling facilities sort by near-infrared spectroscopy, and carbon black absorbs infrared, so black plastic reads as nothing to the sorter. It gets pulled out of the stream or lumped into low-grade mixed batches, and those batches are exactly where e-waste ends up.

The result is a food-contact product containing a chemical that was banned from being manufactured, arriving by the side door. No individual purchase decision fixes that. It's a materials policy problem, and pointing at it is the most useful thing the study did.

What to actually do with your drawer

Not a panicked purge. If you own black plastic utensils, use them up and replace them as they wear out, and don't buy more. Keep them away from prolonged high heat, since heat and oil are what mobilize additives out of plastic in the first place. If a tool is melted, gouged, or older than you can remember, retire it, which is advice we'd give about any kitchen plastic.

When you do replace, the good options are boring. A plain beechwood or maple spoon runs about ten dollars, handles heat indifferently well, and cannot leach a flame retardant. A stainless turner, like the OXO Good Grips fish turner, is the right tool for anything you're searing anyway, with the caveat that it will scratch a nonstick pan without hesitation. If you want flex without wood, GIR makes platinum silicone spatulas that hold up to real heat, though they're expensive for what they are and silicone is still a polymer, not a magic material.

What we would not do is buy a "toxin-free" utensil set at three times the price on the strength of this study. Several brands built a marketing season on the uncorrected number, and none of them have gone back to update the copy.