Microplastics in Water Are Not Inert and Create Harmful Chemicals on Their Own

Their impact on people remains an ongoing question.

Its impact on people remains an ongoing question.

Bottled water is displayed for sale at a Randalls grocery store on January 10, 2024 in Austin, Texas.

One of the reasons plastics are so often used is that they are chemically inert, meaning that bags, food containers and cups will not react with what we put in them. However, they can still break down into microplastics that appear in drinking water, human blood, arteries and brains. The consequences are unclear.

"Whether microplastics can directly cause human disease is still under investigation," says Columbia University chemist Wei Min. In a new study in Nature, Maine's team took a step toward answering that question, showing that microplastics are probably not as chemically inert as we'd expect.

Maine's lab has worked on droplet chemistry for years, studying how water broken into droplets behaves at its boundary with air. These droplets have been reported to produce reactive oxygen species, or ROS, a family of harmful, oxygen-containing molecules. "ROS basically hit proteins, lipids, DNA and cause damage," Min said. "You can break a strand of DNA into pieces. You can mutate DNA, you can misfold proteins."

Many animal studies have linked microplastic toxicity to oxidative stress, which is caused by ROS. So the team wondered if the boundary between water and plastic might do something similar. "We've been inspired by that context," Grymmin said.

In the first experiment, the team stirred polystyrene powder in water containing methyl orange, a yellow dye that scavenges ROS. Within 10 days, the solution became almost colorless and lost approximately 90% of the color. In ethanol or other solvents, the color remained intact. Min and his colleagues concluded that any polystyrene chemistry required water for its realization. So they did more tests.

The first was electron spin resonance, a technique that directly detects radicals—atoms or molecules with an unpaired electron that makes them particularly reactive. The mixture of polystyrene powder and water contains hydroxyl radicals, a highly reactive ROS.

Next, the team used a fluorescent sensor that only lights up when it reacts with hydroxyl radicals. "You don't see it, but when it reacts with local ROS, it turns a bright color," Min said. Under the microscope, the glow focused on the surfaces of the plastic particles and continued to build over time.

This effect was not unique to polystyrene either. The scientists tested six common plastics—polyethylene, polypropylene, polystyrene, PVC, PET, and acrylic—and all produced radicals. Waterproof plastics, polyethylene and polypropylene, were the most active, while PET and acrylic, which contain water-absorbing chemical groups, were the least active.

"The simplest answer is that the most hydrophobic responses change the water structure the most," Min said. In normal or bulk water, each molecule is held together by a network of hydrogen bonds with its neighbors. But a few layers of water molecules pressed onto a water-repellent surface lose some of those partners. "Their structure is no longer the normal structure of water. They are disordered and their energy has also increased significantly," Min said.

Water always contains a small number of hydroxide ions, and converting one to a hydroxyl radical requires the removal of an electron. In bulk water, this costs a lot of energy. But the simulations performed by this group showed that on the polyethylene surface, hydroxide ions with shorter hydrogen bonds require much less energy to form radicals. In addition, plastic surfaces create a negative electrical charge in the water, which further reduces energy costs.

But the Maine team still had doubts that needed to be addressed before any definitive claims could be made.

Maine's previous work on droplet chemistry has drawn much criticism, with some researchers arguing that the reported reactive chemicals come from contamination or measurement artifacts.

To avoid similar objections, Min and colleagues used several independent detection methods. They also added a detergent-like surfactant that coats the plastic and disrupts the water layer on its surface. They found that the more surfactant they added, the weaker the plastic's electrical charge and the less radicals it produced. "We're only changing one parameter while keeping the rest [the same]," Min said. "We use the exact same plastic, the same water, the same everything."

Impurities are usually stable molecules that have no reason to spontaneously generate radicals, Maine explained. However, he is careful not to overdo it. "To be absolutely clear, we cannot rule out 100 percent the involvement of impurities," he said. But our controls and theoretical computational studies strongly suggest that this is intrinsic to the microplastic-water interface.

Finally, the team tested whether this interface chemistry is important for biological molecules. And it probably does.

Proteins and DNA can aggregate on plastic surfaces, and lysozyme, a common experimental protein, also began to aggregate in beads there. The addition of caffeine, which scavenges radicals, largely prevented the build-up attributed to radicals.

Mouse red blood cells that touched the polystyrene particles broke into pieces, and the damage caused by each type of plastic tracked the number of radicals produced. This effect continued in blood serum.

However, this suggests that microplastics are making people sick by producing ROS, Min emphasized. While animal studies often point to ROS as possible origins of oxidative stress and chronic inflammation, ROS can be produced in a variety of ways, including through basal metabolic activity.

Min admitted that it would be difficult to compare the magnitude of this effect with the effect of ROS produced by our own cells.

"If you're really at the interface, you might be significantly affected, whereas if you're just a few microns away, you won't feel it." Maine said. Exposure is another unknown. Min also stressed that no one can yet say how much microplastics the average person gets from food, water and air each year.

The plastics studied were mostly fresh, laboratory-purchased powders, as opposed to the weathered particles we encounter in the environment. In the team's experiments, older particles with more oxidized surfaces produced weaker signals. "This aging process is very heterogeneous," Min said. Some microplastic particles were made 50 years ago and some five days ago. Finally, there is the issue of size.

Microplastics are usually defined as particles smaller than five millimeters, while the minimum cutoff ranges from 100 to 1,000 nanometers. Min's team did not test how changes in microplastic particle size affect water surface chemistry, but the researchers suspect it may play a role. The same mass of plastic broken down into nanoparticles has a much larger surface area, which may magnify the effect. "This may be another reason why nanoplastics are becoming more dangerous," Min said.

The researchers argue that if hydrophobic surfaces guide the common chemistry of water, plastics can be designed to be less hydrophobic, either by changing their molecular structure or additives. First, however, the team wants to focus on the question their study left open. We definitely want to see how much this effect directly contributes to disease or biological problems in animals and humans.

Nature, 2026. DOI: 10.1038/s41586-026-11141-y

Jacek Krywko is a freelance science and technology writer covering space exploration, artificial intelligence research, computer science, and all kinds of engineering wizardry.