Why Nanoplastics in Drinking Water Are So Hard to Measure

If you have read that scientists keep finding nanoplastics in drinking water, the honest answer is more complicated than any single number suggests. According to a 2026 PRISMA-guided review in Molecules, the central problem is not whether these particles exist — it is that the field still cannot measure them in a way that lets one study be compared to another.[Molecules, 2026]

Nanoplastics are usually defined as plastic particles smaller than one micrometer. At that scale, the review argues, the reported concentration in a water sample depends as much on the analytical workflow as on the water itself. That is an uncomfortable conclusion, and it is the real story behind the headlines.[Molecules, 2026]

Drinking water is the end of a longer pollution pathway

The review’s framing is deliberately broad: plastic in your glass is the “downstream analytical endpoint” of a continuum that begins with the production, abrasion, weathering, and fragmentation of plastic in rivers, estuaries, and marine environments. Those fragments move through source waters, treatment infrastructure, and packaging before they ever reach a human.[Molecules, 2026]

That matters because the matrix — the stuff surrounding the particles — changes continuously along the way. River-derived water carries natural colloids, mineral particles, and fluctuating organic matter. Treated water may have fewer particles but can retain residual treatment chemicals. Bottled water adds packaging-specific contributions. A method that works in one segment can fail in another.[Molecules, 2026]

The review is blunt about the implication: validating a method in clean laboratory water is informative but not sufficient. Real-world transferability has to be demonstrated across the actual aquatic-to-potable pathway.[Molecules, 2026]

The core problem is measurement science

This is the review’s central argument, and it is worth stating plainly. Nanoplastics are not a single conventional analyte. They are a polydisperse particle distribution whose reported concentration depends on operational size cutoffs, extraction chemistry, surface adsorption losses, spectral library matching, polymer marker selection, calibration strategy, and blank correction.[Molecules, 2026]

The review cites authoritative agencies reaching the same conclusion. The U.S. Environmental Protection Agency states that no standalone method can characterize the broad diversity of micro- and nanoplastic particles found across environmental matrices. NIST notes that methods commonly used for microplastics generally lack the sensitivity needed for nanoplastics, and that available test materials are insufficient for the metrology regulators require.[Molecules, 2026]

What each method measures — and what it misses

The review assesses the main analytical platforms and treats them as a toolkit where every tool solves one problem while leaving another unsolved.[Molecules, 2026]

Conventional FTIR is robust for micrometer-scale plastics but its spatial resolution is poorly matched to true nanoscale particles, making it insufficient as a primary method for the smallest drinking-water particles. Routine Raman workflows are similarly inadequate for genuine nanoscale analysis.[Molecules, 2026]

Raman microscopy can reach down to about 100 nm under carefully controlled conditions, offering particle-level chemical fingerprints — but it suffers from low throughput, fluorescence interference, and matrix effects. Surface-enhanced Raman spectroscopy (SERS) boosts sensitivity dramatically, with recent aqueous studies reporting detection down to roughly 20 nm. The catch: the signal depends on the plasmonic substrate as much as on the particle, so a strong SERS signal is not automatically a reliable concentration estimate.[Molecules, 2026]

AFM-IR can chemically interrogate individual particles from 20 to 1000 nm with morphology context, but it is low-throughput and demands specialist instrumentation — better suited as a confirmatory reference tool than a survey method. Pyrolysis-gas chromatography-mass spectrometry (Py-GC/MS) delivers robust polymer-specific mass concentrations, but it destroys the particles, so it preserves no size, shape, or count information.[Molecules, 2026]

Why the sample is lost before the detector turns on

One of the review’s most useful points is that sample preparation is not a neutral step — it determines which particles are even available to measure. Filtration, ultrafiltration, oxidative digestion, centrifugation, and even transfer between vessels can change the apparent abundance and size distribution of the target particles.[Molecules, 2026]

Filters may exclude deformable aggregates or retain particles differently depending on surface chemistry. Oxidative digestion can reduce organic matter but may also alter surface signatures or cause aggregation. Even moving a sample between glass and polymer vessels can cause wall losses at the nanoscale. The result, the review notes, is that sample-preparation uncertainty often exceeds instrumental uncertainty — yet it is frequently underreported.[Molecules, 2026]

Bottled water is a special case

The review singles out bottled water as the clearest example of why exposure cannot be reduced to source-water contamination alone. One study found nanoscale organic particles in bottled water and identified bottle degradation as a likely source. A later study used SERS to identify PET nanoplastics in commercial bottled water with a mean size of approximately 88.2 nm — a polymer directly compatible with bottle packaging.[Molecules, 2026]

~88 nm

But the review cautions against over-reading these findings. A high apparent particle burden does not reveal which step produced it. Particles may come from bottle walls, caps, liners, filtration components, transport stress, storage, or even handling after purchase. Bottled-water data, it concludes, is “exposure-relevant but mechanistically composite.”[Molecules, 2026]

The honest counterpoint: detection is not the same as certainty

This is where the review earns its credibility — by refusing to overstate what the evidence shows. It concludes that the field has moved past the question of whether nanoplastics can occur in drinking water. The harder question now is how to convert plausible detection into a comparable measurement.[Molecules, 2026]

The review is explicit that it avoids pooled concentration estimates on purpose, because the underlying studies report fundamentally different endpoints — particle counts, hydrodynamic diameters, mean sizes, polymer-specific mass, and qualitative identification. Combining those statistically, it argues, would create “an artificial precision that the underlying measurements do not support.”[Molecules, 2026]

It also acknowledges its own limits. The formal search was deliberately restricted to publicly accessible scholarly records and backward citation chaining rather than subscription-gated export tools — a choice that improved auditability but may have missed records indexed only in paywalled platforms. Of 33 records screened, 22 were included in the final synthesis, with only 8 directly addressing drinking water.[Molecules, 2026]

What would actually fix this

The review’s practical prescription is a tiered, “fit-for-purpose” workflow rather than a search for one perfect instrument. It describes four layers: contamination-controlled sampling and preprocessing with documented blanks; a screening stage using tools like NTA or DLS as provisional descriptors; particle-resolved chemical confirmation using Raman, SERS, AFM-IR, or related methods; and mass-resolved polymer quantification via Py-GC/MS.[Molecules, 2026]

The strongest evidence, it repeatedly emphasizes, comes from orthogonal workflows — where at least two independent methods converge on the same interpretation. A study pairing particle-resolved spectroscopy with polymer-specific mass spectrometry is more defensible than one resting on light scattering alone.[Molecules, 2026]

Above all, the review calls for validated reference materials, LOD/LOQ reporting, matrix-matched recovery testing, and interlaboratory harmonization. These, it argues, “are not optional refinements; they are prerequisites for a mature field.”[Molecules, 2026]

The practical takeaway

For a reader, the useful lesson is not a scary number — it is a healthy skepticism about numbers. When you see a precise figure for nanoplastics in drinking water, the review’s framework suggests asking what was actually measured: particle count or polymer mass, which size window, with what blanks and recovery checks. The science is advancing, and the occurrence of nanoplastics in drinking water is increasingly credible. But as the Molecules 2026 review makes clear, the field has demonstrated detection without yet achieving metrological closure.[Molecules, 2026]