Are Microplastics Chemically Inert? What the Nature Study Found

What did the Nature microplastics study find?

The paper is “Intrinsic physicochemical activity at microplastic–water interfaces” by Naixin Qian, Wei Min and colleagues at Columbia University, with simulations from Teresa Head-Gordon’s group at UC Berkeley. It was published in Nature on October 14, 2026 and is open access. Columbia’s news office and phys.org covered it the same day.

The core claim is simple to state. Plastic has long been described as chemically inert: it doesn’t react with much, which is why it makes good packaging. The team found that when ordinary plastic particles sit in water, the thin layer of water touching the plastic behaves differently from water in the bulk, and some hydroxide ions in that layer give up an electron. That turns them into hydroxyl radicals, among the most reactive molecules in chemistry. In Columbia’s release, Min put it as water being “a key active player here, rather than just a spectator.”

So the plastic does not need sunlight, heat, scratching or a leaching additive for this to happen. The authors argue it is a property of the surface itself, which is why they call it intrinsic.

How was it measured, and under what conditions?

This matters more than the headline, because every condition limits how far the result travels.

• The plastics: commercial powders of six polymers you meet daily: polyethylene (PE), polypropylene (PP), polystyrene (PS), polyvinyl chloride (PVC), polyethylene terephthalate (PET) and acrylic (PMMA). These were clean, as-purchased particles, not fragments scraped from old containers or pulled from seawater.
• The water: mostly phosphate-buffered saline at pH 7.4, roughly the pH of blood, at room temperature. Some tests used cell culture medium and serum.
• Light: the main probe assays were incubated in the dark for 90 minutes, which rules out sunlight-driven chemistry as the explanation.
• Detection: three independent readouts. A fluorescent probe (APF) that lights up when it meets hydroxyl radicals, electron spin resonance with a radical trap, and the slow bleaching of a dye called methyl orange.
• Amounts: milligrams of plastic per milliliter of liquid, for example 1 mg/mL for imaging and 10 mg/mL for the dye and binding tests. Those are lab quantities chosen to make a signal visible, not estimates of what is in a glass of water or a bloodstream.
• Theory: the Berkeley group ran simulations showing hydroxide ions near a more water-repellent surface lose electrons more easily, which matches the experimental ranking.

The radical output was modest. The authors estimate at least 0.1 to 1 nanomolar per minute at the interface, and label that a lower bound. Over days it adds up: with polystyrene in water, nearly 90% of a dilute dye was bleached across 10 days.

Which plastics were most reactive?

The ranking followed water-repellence, measured as the contact angle of a water droplet on each material. The two polyolefins, polyethylene and polypropylene, repel water most and produced the strongest signal. Acrylic, the least water-repellent of the six, produced the weakest.

Two cautions. First, the ranking describes the chemistry of particles in water. It is not a ranking of which container sheds the most particles; that is a separate question with its own studies, which our microwaving plastic containers guide covers. Second, the most common plastics in daily life landed at the reactive end, which is why the paper drew attention.

What happened to cells and proteins?

The team then asked whether this surface chemistry could matter to living material. Every test was in vitro, in a dish or tube.

• Red blood cells. Mouse red blood cells, kept at body temperature for an hour with polystyrene particles, broke into fragments and released hemoglobin. Across plastics, the damage tracked surface activity with a correlation of 0.75.
• A protein. Lysozyme, a common test protein, clumped together on polystyrene surfaces, the kind of misfolding linked to tissue damage. The correlation with activity was 0.68.
• A radical scavenger. Adding caffeine, which mops up hydroxyl radicals, suppressed that clumping. That supports the idea that the radicals, rather than simple contact, drove it.
• DNA and proteins sticking. DNA and several proteins gathered on particle surfaces, more so on water-repellent polypropylene than on acrylic. That puts biological molecules right where the radicals form.
• A surfactant shield. Coating particles with a detergent-like surfactant (Triton X-100 at 0.1%) cut the radical signal sharply and prevented the protein damage.

The authors frame all this as a “functional hypothesis” for how microplastics might harm tissue, not as proof that they do. That is the right reading.

Does this mean microplastics are harming you?

Not on this evidence alone. Health research sits on a ladder, and this paper is on the lower rungs.

Leonardo Trasande, a pediatrician who directs NYU’s Center for the Investigation of Environmental Hazards and was not involved, called it early-stage but an important step in an Inside Climate News interview published October 31. He also pointed out that it looks at one pathway and does not address the additives and impurities in plastics, which are a separate concern.

There is also a reason the effect might be smaller in the body than in a beaker. Inside us, particles get coated almost immediately in proteins and fats, often called a corona. The paper’s own surfactant experiment showed that a coating can damp the activity. Whether a biological coating does the same is an open question the authors did not settle.

One disclosure readers should know: the paper states that Columbia University has filed a provisional patent application based on the work. That is common for university research and doesn’t undermine the data, but it belongs in the picture.

How does this fit with other microplastics health studies?

It fills a gap. Observational studies have found plastic particles in human tissue and linked them to worse outcomes; the best known, a 2024 New England Journal of Medicine study, found people with microplastics in their carotid artery plaque had about four times the risk of heart attack, stroke or death over the follow-up period. Observational links like that cannot show cause, and critics have asked what plastic would physically do to tissue. A surface that produces oxidizing radicals is one candidate answer, and oxidative stress is a well-studied route to inflammation.

That makes the Nature paper useful as a piece of a puzzle, not a verdict. Our heart attack study explainer covers the NEJM findings and their limits, and the human health evidence review walks through the tissue studies one by one. The FDA’s position has not changed: it says current evidence does not demonstrate that the levels of micro- and nanoplastics detected in foods pose a risk to human health.

The same Columbia lab is behind a 2024 PNAS study that counted an average of about 240,000 detected plastic particles per liter across three US bottled water brands, about 90% of them nanoplastics. News stories sometimes merge the two. They are different papers asking different questions: one counted particles, the other studied what particles do in water.

Use the App

Want to know which plastics you live with every day?

Scan your water bottle, food containers or kettle. The app reports the declared materials, flags the parts that touch food and drink, and saves your swaps so you can track what you've changed.

What should you do differently?

Very little, and that is the honest answer. The finding strengthens the case for reducing avoidable exposure; it doesn’t add a new rule. What it does offer is a way to sort good advice from bad.

1. Keep heat away from plastic food contact. In container studies, heating is the condition that raised measured particle release the most. Reheat in glass or ceramic, and don’t pour boiling water into plastic. See microplastics in electric kettles for one common case.
2. Drink filtered tap water rather than bottled where you can. The bottled-water counts above are among the highest measured. Our tap vs bottled comparison has the numbers and their caveats.
3. Replace worn plastic that touches food. A scratched board, a cloudy tub or a frayed utensil is visibly breaking down, so it is the easiest item to swap first.
4. Don’t buy a fix the study didn’t test. The caffeine result is not advice to drink more coffee: the lab used 100 millimolar caffeine. A 12-ounce brewed coffee with 113 to 247 mg of caffeine, the range the FDA lists, works out to roughly 1.6 to 3.6 millimolar, so the test was about 30 to 60 times stronger. Antioxidant supplements and “microplastic detox” products were not studied at all.
5. Don’t chase polymer rankings at the shelf. PE and PP ranked most reactive as powders, yet PP is also the plastic most often labeled for microwave use, so the ranking cannot tell you which container to buy. Glass, stainless steel and ceramic remove the question for the items you use hottest and most often.

What research would settle the open questions?

Three kinds of follow-up would move this from mechanism toward risk. Tests on weathered plastics and particles isolated from real food, water and air, since the paper suggests oxidized surfaces are less active. Experiments in biological fluids that coat particles in proteins, at concentrations closer to measured human exposure. And animal or human-cell studies that track oxidative stress markers against particle dose. Until then, the safest summary is that plastic is less inert than its reputation, and nobody yet knows by how much that matters inside a person.

For background on why particle size changes the picture, see nanoplastics vs microplastics.