The average store-bought bottle of water contains somewhere in the neighborhood of 10 to 100 times more minute plastic particles than previously believed, judging from a study published this week.

We've long known about the presence of microplastic particles in bottled water and proliferating through nature, while nanoplastics (pieces less than a micron, or 0.001 mm) have been difficult to accurately measure. This is where boffins at the Columbia and Rutgers universities in the US come in with a newly developed microscopy technique to image some of the smallest plastic particles yet spotted in our water. 

Using a technique dubbed stimulated Raman scattering (SRS), the academics discovered an average of 240,000 plastic particles per litre of bottled water across three of the brands tested, with measurements ranging from 110,000 to 370,000 fragments per litre. This is, we're told, 10 to 100 times greater than earlier estimates. Around 90 percent of the particles detected using SRS qualified as nanoplastics, the rest being microplastics.

"It is not totally unexpected to find so much of this stuff," Columbia graduate student Naixin Qian, lead author of a PNAS-published paper describing this study, explained. "The idea is that the smaller things get, the more of them there are."

Previous studies of micro and nanoplastics in the water and food supply have linked them to reproductive abnormalities, increased mortality, oxidative stress and other biological harms to humans and other animals. 

Taking a page from the AI training playbook

The team developed SRS to overcome three hurdles found with other techniques used to measure plastic particles in water: Poor instrument resolution, poor detection sensitivity, and an inability to distinguish different types of plastic. According to the team, they've cleared all three hurdles. 

SRS works by shooting lasers at water samples to make specific molecules resonate and using software to identify plastic particles from those results. The team said they were able to train their SRS system to detect seven types of plastics with an identification rate of more than 96 percent and a false positive rate of just one percent, the team said. Speed of recognition was also not a problem.

"Compared to the traditional Raman technique which may need days to map a filter surface, SRS only needs several hours for a sample," Beizhan Yan, associate research professor at Columbia and one of the study's authors, told The Register.

Plastics, plastics and more plastics

While SRS has allowed the team to say with confidence that they've determined there are orders of magnitude more plastic in bottled water than previously believed, there's still a lot the study's method isn't able to detect. 

According to the paper, the software was only able to account for about 10 percent of the total particles imaged using SRS. The rest, said Yan, could be organic, or inorganic particles including plastics that weren't among the seven types the algorithm was trained to identify. 

For Yan, the biggest find by the study was "the variety of the plastic types and different size distributions" found in water samples.

"Previous studies found the major chemical composition to be PET, which is expected since the bottle is made of PET," Yan told us. "Therefore, for a long time, we had the impression that the majority of plastic particles inside bottled water would be Polyethylene terephthalate (PET) particles of a few hundred nanometers in size."

"That was incorrect," Yan said. 

Ironically enough, the most commonly detected plastic was polyamide, a material commonly used in water filtration systems and "the most popular membrane material used in reverse osmosis … a common water purification method shared by all three brands [of water tested]," the team wrote in the paper. 

The boffins hope that their ability to identify specific types of plastic particles present in water can help find sources of microplastic contamination. PET and PE, both commonly used in the plastic bottles used to hold water, were, as noted, found in the samples. Other polymers were also found in the bottles, but aren't part of the packaging process, hence a likely contamination source "before or during water production," the team said. 

SRS could be tuned to similarly detect non-plastic particles in water, which Yan told us the team is working on. They also plan to use their SRS technique to analyze tap water for nanoplastic contamination, as well as air samples and biological tissues. ®

https://www.theregister.com//2024/01/10/water_nanoplastic_pollution/