Microplastics at home: what to do, what’s hype and what the research shows

Cutting boards. North Dakota State University researchers chopped on polypropylene and polyethylene boards. Polypropylene boards released more than polyethylene boards, and on polyethylene boards, chopping carrots released more than chopping without them. Most particles were under 100 μm. Based on their own assumptions about use, the authors estimated that a person could be exposed to about 7.4 to 50.7 g of microplastics a year from a polyethylene board and about 49.5 g from a polypropylene board. A preliminary test on mouse cells showed no adverse effect.⁴

Plastic containers. Researchers tested containers and reusable pouches with water and with 3% acetic acid. Microwaving released the most, up to 4.22 million microplastic and 2.11 billion nanoplastic particles per cm² in three minutes. Storing food for more than six months also released millions to billions. The authors modeled the highest daily intake for infants drinking microwaved water and for toddlers eating microwaved dairy. Particles extracted from the plastics killed about 77% of cultured kidney cells, but at a very high dose (1,000 μg/mL), so it doesn’t show what everyday exposure does.⁵ The FDA says it doesn’t have enough evidence to show packaging plastics migrate into food and drinks, and notes that many studies used methods of “variable, questionable, and/or limited accuracy and specificity.”¹

Tea bags. One plastic tea bag steeped at 95 °C released about 11.6 billion microplastic and 3.1 billion nanoplastic particles in the lab. The bags were nylon and PET. Water fleas showed dose-dependent effects.⁶ The study makes no claim about paper or cotton bags, so neither do we.

Kettles. Researchers boiled water in eight new polypropylene kettles. On the first boil, nanoplastics measured 0.011 ± 0.005 µg/cm² and microplastics 0.032 ± 0.016 µg/cm². By the tenth boil release had dropped, and after 150 boils it was about 0.0004 µg/cm² for nanoplastics and 0.002 for microplastics. Hard tap water lowered nanoplastic release versus ultrapure water. The authors suggest discarding the first few boils; rinsing alone was less effective. The nanoplastics were chemically confirmed; the particle counts were not.⁷

Nonstick pans. Researchers scratched PTFE pots and used Raman imaging. Their abstract says thousands to millions of Teflon micro- and nanoplastics might be released during a mimic cooking process.⁸ We read the abstract only. Our ceramic nonstick guide covers the main alternative.

Bottled water. Qian and colleagues counted about 2.4 ± 1.3 × 10⁵ particles per liter (about 90% nanoplastics) across three unnamed bottled-water brands, using a new imaging method.¹⁰ The paper was a methods study. Materić argued the blank samples were contaminated at levels comparable to the bottled water, so the quantification is “fundamentally unreliable.” The authors replied that their blanks controlled for the contamination that mattered, mostly airborne, and that the alternative blank water was itself contaminated.^11,12 Both sides agree blanks matter. ANSES, a French agency, measured far lower counts using a different method: plain water averaged 4.5 particles per liter in glass bottles and 1.6 in plastic bottles and cartons.⁹ The WHO said in 2019 that, based on limited information, microplastics in drinking water don’t appear to pose a health risk at current levels.²

Glass and caps. ANSES found glass-bottled cola, lemonade, iced tea and beer averaged about 100 particles per liter. It linked them to paint on metal caps. Cleaning the caps (blowing air, then rinsing) cut particles from 287 to 87 per liter. ANSES says that without toxicological reference data, it can’t say whether those levels pose a health risk.⁹

Filters. Researchers tested pour-through devices on spiked water with PET and PVC fragments and nylon fibers (30–1,000 µm). Two membrane-filter devices removed 78–86% of PVC and 94–100% of PET fragments. A device using only granular activated carbon and ion exchange produced more particles in its filtered water than it took in. Only three devices were tested, and the test didn’t address nanoplastics.¹³

Boiling. Researchers found that polystyrene, polyethylene and polypropylene particles (0.1 to 150 μm) co-precipitate with calcium carbonate when hard water (over 120 mg/L CaCO₃) boils. Boiling removed at least 80% of them. The particles end up inside the scale, so the benefit depends on leaving it behind.¹⁴

Brain tissue (Nature Medicine, 2025). Nihart and colleagues analyzed autopsy tissue. They reported polyethylene-dominated plastic particles in brain, liver and kidney. Brain levels were higher in 2024 samples than in 2016 ones, and highest in people with dementia. The main method was pyrolysis gas chromatography–mass spectrometry (Py-GC/MS).¹⁵

Critics responded:

• A nine-author critique (a “Matters Arising”) in the same journal points to limited contamination controls and a lack of validation, and says lipids can give false-positive polyethylene signals.¹⁶
• Germany’s BfR called the study noteworthy but said the reported amounts “appear implausibly high” and that false or misinterpreted signals can’t be ruled out.³
• A 2025 paper from Rauert and colleagues, on blood samples, argues Py-GC-MS is “not currently a suitable technique for identifying polyethylene or PVC due to persistent interferences.”¹⁹
• The Guardian reported on concerns about seven challenged studies. It quotes the study’s senior author, Matthew Campen, who says most criticism of the work to date “has been conjectural and not buffeted by actual data.”²⁰ Another scientist, Béen, told the Guardian there is “less and less doubt about the fact that MNPs are there in tissues. The challenge is still knowing exactly how many or how much.”²⁰

Arteries (NEJM, 2024). Marfella and colleagues followed 257 patients after carotid surgery (average 33.7 months). Polyethylene was found in 58.4% of plaques. Compared with patients whose plaques had no detectable particles, those with particles had a hazard ratio of 4.53 (95% CI 2.00 to 10.27) for heart attack, stroke or death.¹⁷ This was an observational study, so it can’t show that the particles caused the events. In correspondence, Holda and Batko raised contamination concerns: no operating-room blanks, one analytical method (GC-MS) for every plaque, and detected particles smaller than those in earlier studies.^17,18 The authors replied in the same issue; we haven’t read the reply.

Where this leaves us. Particles in human tissue seem likely. How many, and what they do, is unsettled.