Microplastics have been identified in commercially distributed edible vegetable oils, including products sold in PET and glass bottles. Studies in Italy and Spain, Türkiye and Bangladesh report widely differing particle abundances; methodological differences limit direct comparison. This review evaluates polymer identity, packaging pathways, regional evidence gaps and maternal–infant exposure research, while distinguishing particle detection from demonstrated health harm.
1. Scope and methodology
Targeted searches of scholarly databases, indexed abstracts and primary research journal pages focused on edible oils, microplastics, PET, HDPE/LDPE, polypropylene, glass containers, FTIR and Raman spectroscopy, pregnancy, human placenta and breast milk. Only directly measured edible-oil studies were used to describe oil particle concentrations. Studies of other food-contact materials were treated as mechanistic context, not vegetable-oil observations.
This is a preliminary scoping review, not a PRISMA-compliant systematic review. The included studies differ in their particle-size detection limits, solvent preparation, contamination controls and polymer identification. Study means therefore cannot support a regional ranking or a pooled global concentration.
2. Quantification in commercial edible oils
| Region / study | Product and package information | Reported particles/L | Technique and interpretation |
|---|---|---|---|
| Italy and Spain — Battaglini et al. 2024 | Olive, sunflower and mixed vegetable oils; PET and glass | 1,140 ± 350 average | Automated μ-FTIR, 5-μm filtration; fragments common. [1] |
| Türkiye — Şirin et al. 2026 | 28 olive and sunflower oils; PET, tin and glass | 1,357 ± 663 average | μ-Raman; oil types and packages not fully matched. [2] |
| Bangladesh — Meem et al. 2026 | 60 samples of branded and non-branded soybean oils; not a controlled glass-versus-plastic trial | 53,922 ± 10,510 average | Optical screening and FTIR; fibers dominate, identification and size criteria differ. [3] |
| Unassigned retail market — Guo et al. 2023 | Four commercial PET-bottled edible oils | 134,000–580,000 observed range | Raman-based detection; >80% below 10 μm. Sampling-market geography not established in reviewed abstract. [4] |
Figure 1 · Reported mean microplastic counts in three retail-oil studies
3. Regional overview
Europe — Italy and Spain
Microplastics were reported across all investigated retail oil types and bottles, including PET and glass packaging. Of identified particles, 81.2% were fragments and 77.5% were smaller than 100 μm. Polyethylene represented 50.3% and polypropylene 28.7%. Investigators did not find a statistically significant difference between the oil samples or types in the reported analysis. An isolated PET-versus-glass causal effect was not established. [1]
Türkiye — a transcontinental market
The 2026 study analyzed 28 commercial olive and sunflower oils with packaging categories including plastic, tin and glass. PE, PP and PET were prominent. Because packaging and oil-type categories were not fully crossed, the data are informative about occurrence but are not a clean packaging-material experiment. The article's highlights and abstract differ in the reported minimum–maximum range; readers should refer to the full paper for category-specific figures. [2]
Asia — Bangladesh and storage research
The Bangladesh study found higher microplastic counts in non-branded than branded soybean oil and a predominance of fibers. Open-market handling and reused containers were discussed as potential influences. That comparison is not evidence of a specific PET-versus-glass effect. Separately, a 2026 controlled storage study found changes in particle abundance and size distributions in edible oils over time, supporting assessment of storage temperature and duration. [3] [5]
North America
The reviewed evidence does not establish a harmonized, representative PET-versus-glass comparison for North American retail cooking oils. The 2023 bottled-oil paper was authored by U.S.-affiliated researchers, but its sampling-market geography cannot be inferred from institutional affiliations. Regional market mapping remains an open research priority. [4]
Africa
No sufficiently comparable, packaging-stratified quantitative study of commercial retail vegetable oil was verified for an African market in this targeted search. That is an evidence gap, not evidence that contamination is absent. Follow-up sampling should distinguish sealed, commercially bottled oils from bulk and refill distribution systems.
4. Packaging systems and polymer identity
PET bottle bodies, polyethylene (including HDPE and LDPE) components, polypropylene closures, glass bottles, and lined metal containers require separate evaluation. Detection of PE, PP or PET in an oil sample cannot alone identify the source; polymers may be introduced during oil processing, packaging, handling, storage or sample preparation.
Figure 2 · Polymer composition identified in the Italy–Spain study
Glass bottle bodies are not themselves a source of synthetic plastic particles, but caps, liners, transport equipment, filling operations and background contamination can still contribute. Likewise, metal packages may have polymer coatings or seals.
In an accelerated PET-bottle study at 40°C for ten days, Guo and colleagues reported no detected PET particle release attributable to the bottles under the tested conditions. This result cannot establish that PET is universally free of migration under all manufacturing or storage conditions. [4]
Figure 3 · Conceptual packaging-to-exposure pathways
5. Pregnancy: what human placental findings mean
The placenta has been studied as one possible compartment in which particles may be identified. Yun and colleagues used Raman microspectroscopy in 50 placentas and detected 40 particles in 31 samples; their results did not establish statistically significant differences in the assessed maternal and neonatal outcomes between the studied polymer groups. [6]
Laboratory work suggests possible mechanisms—oxidative stress, inflammatory signaling, and cellular responses—but experimental exposures do not necessarily reproduce human consumption of retail vegetable oils. Neither a causal route from edible-oil packaging to placental particles nor an attributable pregnancy outcome has been demonstrated. Other possible sources include other foods, beverages and airborne particles.
6. Lactation and breast milk: detection is not a risk conclusion
Ragusa and colleagues detected microplastic particles in 26 of 34 breast-milk samples using Raman microspectroscopy. A later study reported detection in 23 of 59 samples. These results document detection in research samples; they do not quantify infant harm or identify edible cooking oils as the source. Differences in study protocols preclude treating the positive-sample fractions as population prevalence or a reliable trend. [7] [8]
Figure 4 · Detection frequency in selected human biological-sample studies
Breastfeeding should not be discouraged on the basis of these findings. The WHO recommends exclusive breastfeeding for the first six months, followed by appropriate complementary foods and continued breastfeeding up to two years or beyond. The documented benefits of breastfeeding remain relevant despite unanswered questions about microplastic exposure. [11]
7. Infants: food contact, feeding equipment and complementary foods
Infant exposure can include direct contact with feeding materials. Li and colleagues reported experimental polypropylene feeding-bottle particle release reaching 16.2 million particles/L under certain test conditions, with temperature and preparation procedures influencing results. That experiment is not an edible-oil study and does not establish a typical daily infant intake. [9]
After complementary foods are introduced, cooking oils may contribute some dietary particles, but oil-specific proportions of total infant exposure have not been established. Food-safety instructions and hygienic infant-feeding practices should not be compromised in attempts to reduce hypothetical particle exposure.
8. Evidence hierarchy and human-health interpretation
| Question | Current interpretation |
|---|---|
| Are microplastics detected in retail vegetable oils? | Yes. Multiple direct studies report occurrence. |
| Are they detected in some human placentas and breast-milk samples? | Yes. Research detection has been reported, with substantial measurement limitations. |
| Does PET always produce higher oil microplastic concentrations than glass? | Not established. Matching and source-specific control studies are needed. |
| Have oil-derived particles been traced into human placenta or breast milk? | No demonstrated oil-specific causal pathway. |
| Do typical edible-oil exposure levels cause pregnancy or infant health harm? | No established clinical causality. Evidence and exposure models remain insufficient. |
| Does changing cooking-oil packaging improve maternal or infant clinical outcomes? | Not demonstrated. |
EFSA states that it has not completed a full food-related microplastics risk assessment and that knowledge of dietary exposure and human-health effects remains limited. Current evidence therefore does not justify assigning a universal “safe” particles/L limit or making claims that one commercial oil protects fetal or infant health. [10]
9. Recommendations for the next comparative study
The strongest next step would be a harmonized multi-market surveillance program, followed by controlled packaging trials. The initial retail phase should stratify each market by oil category, packaging polymer, bottle size, lot, storage history and shelf age. All samples should be measured using predeclared particle-size cutoffs and spectroscopic criteria, with method blanks and recovery experiments.
Figure 5 · Recommended controlled study design
The primary outcome should be blank-corrected, spectroscopically verified particles/L by size class and polymer identity. Secondary measures could include polymer mass using Py-GC/MS, particle morphology, bottle and closure material analysis, temperature sensitivity and time-dependent change. In oil matrices, recovery and detection thresholds are particularly important to interpret cross-study differences.
10. Conclusions
The emerging literature demonstrates that edible oils may contain microplastic particles across different packaging types. PE, PP and PET are among the materials identified, but polymer identity alone cannot establish contamination sources. Reported concentrations vary substantially, and direct regional comparisons are methodologically inappropriate without harmonized procedures.
The early-life exposure literature documents microplastics in selected placenta and breast-milk samples and laboratory release from some feeding bottles. It does not yet demonstrate that microplastics from cooking-oil packaging cause adverse pregnancy, lactation, or infant health outcomes.
For consumers and the food industry, an evidence-led response emphasizes contamination prevention, packaging-system testing, improved analytical quality control, transparency about uncertainty and preservation of established maternal and infant nutrition advice.
References and source links
- Battaglini E, et al. (2024). Analysis of microplastics in commercial vegetable edible oils from Italy and Spain. Food Chemistry 443, 138567. Source ↗
- Şirin M, et al. (2026). Microplastic contamination and associated health risks in edible sunflower and olive oils: The first evidence from Turkey. Food Control, 112562. Source ↗
- Meem RA, et al. (2026). Microplastic contamination in branded and non-branded edible soybean oil from Bangladesh. Journal of Food Composition and Analysis 152, 109005. Source ↗
- Guo X, et al. (2023). Detection and quantification of microplastics in commercially bottled edible oil. Food Packaging and Shelf Life 38, 101122. Source ↗
- Huang C, et al. (2026). Detection and variation trends of microplastics in edible oils during storage. Food Control 188, 112261. Source ↗
- Yun X, et al. (2024). Raman-guided exploration of placental microplastic exposure. Journal of Hazardous Materials 477, 135271. Source ↗
- Ragusa A, et al. (2022). Raman microspectroscopy detection and characterisation of microplastics in human breastmilk. Polymers 14, 2700. Source ↗
- Detection of microplastics in human breast milk and its association with changes in human milk bacterial microbiota. (2024). Journal of Clinical Medicine 13, 4029. Source ↗
- Li D, et al. (2020). Microplastic release from the degradation of polypropylene feeding bottles during infant formula preparation. Nature Food 1, 746–754. Source ↗
- European Food Safety Authority (2026). Microplastics and nanoplastics in food. Last reviewed October 21, 2026. Source ↗
- World Health Organization (2026). Infant and young child feeding. Updated August 4, 2026. Source ↗