Selected feeding bottles have released plastic particles during hot-water preparation tests and during repeated opening and closing. The results depend on the bottle, handling protocol, particle-size window, and analytical preparation. These studies establish possible release under stated conditions. They do not measure a baby’s internal dose or demonstrate a clinical health effect.12
What this page covers
The direct evidence reviewed here separates a water-based simulation of formula preparation from an experiment on cap and neck abrasion. These are distinct sources and denominators. The hot-water experiment reports particles in the collected water. The abrasion experiment reports particles in a concentrated ethanol extract prepared for the instrument. Their numbers cannot be compared as equivalent concentrations in a feed.12
The library also contains infant textile and fetal biomonitoring papers. Those do not test feeding-bottle release and are not used as substitutes for bottle measurements. This category is a selected review of direct experimental evidence rather than an exhaustive survey of all bottle designs or feeding practices.
What the studies found
The polypropylene-bottle protocol reports a range of particle concentrations after cleaning, sterilization, and shaking with hot deionized water. Raman measurements identified the polymer in sampled filter regions, and image analysis related the observed regions to the filtered volume. This is a release test in water, without milk powder or a measured amount ingested by an infant.1
Scroll horizontally for methods and measurement limits.
| Experiment | Reported result | Identification and size context | Interpretation |
|---|---|---|---|
| PP bottle preparation protocol1 | 1.31-16.20 million particles/L | Raman in representative filter fields; classification bins begin at 0.8 µm | Product-dependent water-simulant concentrations, with spatial extrapolation. |
| Cap/neck abrasion in one plastic feeding bottle2 | 112 confirmed plastic particles/mL of concentrated ethanol extract | LDIR chemical identification; reported plastic sizes 20-500 µm | Concentrated ethanol analytical extract; not a milk concentration. |
How the hot-water experiment was done
The protocol cleaned and rinsed new bottles, immersed them in hot water for sterilization, then placed hot deionized water inside and shook the bottle. Glass hardware minimized additional plastic sources. The collection water was filtered, and low-power Raman spectroscopy compared particle spectra with a polypropylene reference. Atomic force microscopy provided separate surface-topography information.1
Only representative filter regions were examined rather than every point on the membrane. The resulting concentration therefore depends on how well those fields represent the full collection area. The protocol tested recovery with standard particles, which supports the collection workflow, but the stated filter pore size alone is not proof of complete and equal detection at that size for every shape or polymer.1
The reviewed artifact is a repository author manuscript with line numbers and figure descriptions. Its representative-results section gives the range used here. It belongs to the same research program as earlier polypropylene bottle work and should not be counted as an independent replication of that earlier dataset. The available manuscript does not provide a numerical routine full-procedure blank distribution for every bottle test.1
Why total microparticles are not all microplastics
The cap experiment first collected material generated by repeated opening and closing, rinsing caps and necks with filtered ethanol. The extract was dried and diluted to a specified final volume before analysis. Laser direct infrared chemical imaging separated identified plastic polymers from cellulose, additives, and other particles. The paper’s total microparticle column is larger than its confirmed-plastic column and must not be quoted as if the two were equivalent.2
Smaller particles were imaged separately by electron microscopy. An image of a small object does not by itself establish that it is plastic, and those observations cannot extend the chemically confirmed infrared count into the nanoscale range. A size-bin statement in the discussion conflicts with the reported instrument window, so this page retains the clearly described infrared range instead.2
Why the studies cannot supply one bottle count
The experiments differ in handling, solvent, recovery, final volume, and chemical-identification method. One evaluates a hot-water preparation protocol; the other isolates abrasion from closures and necks. Neither result is a category-wide release rate. A high result for a tested specimen does not show that every product made of the same polymer behaves that way, and a low result in a larger-particle window does not establish absence of smaller material.12
Both studies address particle generation. Neither follows exposed infants, measures uptake from the prepared liquid, or establishes a relationship between a bottle count and disease. Statements about material release should remain separate from decisions about feeding hygiene. These experiments do not test the health consequences of changing sterilization or formula-preparation procedures.12
Regulatory status
The supplied EU restriction addresses defined synthetic polymer microparticles as substances or intentionally present in mixtures, with exclusions, derogations and transition provisions. A shedding measurement from an article does not by itself establish whether that restriction applies or has been violated.3 The drinking-water methods are useful analytical comparisons, but they are not validated product-specific release tests for this category.45 No universal compliant particle count is inferred from them. Read the regulatory source guide.
What remains uncertain
The reviewed evidence does not provide a representative exposure distribution across all feeding bottles, a reliable count for every particle size, or a clinical threshold for these mixtures. More comparable tests using intact products and realistic food matrices would improve generalization. For now, the clearest interpretation keeps each result attached to its actual protocol and reports the missing information directly.12
References
- Sampling, Identification and Characterization of Microplastics Release from Polypropylene Baby Feeding Bottle during Daily UseResearch article
- Microparticles and microplastics released from daily use of plastic feeding and water bottles and plastic injectors: potential risks to infants and children in ChinaResearch article
- Commission Regulation (EU) 2023/2055: restriction on synthetic polymer microparticlesLegal instrument
- SWB-MP1-rev1: Infrared spectroscopy of microplastic particles in drinking waterOfficial method
- SWB-MP2-rev1: Raman spectroscopy of microplastic particles in drinking waterOfficial method