Diets
industrial refining of seed oils (high-heat deodorization) produces harmful contaminants (3-MCPD esters, glycidyl esters, trans fats)
In plain terms: Does refining seed oils create harmful chemicals like trans fats and 3-MCPD?
Part of: 🔍 seed oils
True in food chemistry — high-temperature deodorization measurably generates 3-MCPD esters, glycidyl esters (a probable carcinogen precursor) and small amounts of trans fats; the contaminant chemistry is well documented (EFSA-relevant), though typical dietary exposures are low and human-harm estimates rest on animal toxicology + margin-of-exposure modeling, not human outcome trials.
📅 Last reviewed: 2026-07-14 ⓘ
Evidence ladder
How far up the ladder this claim has climbed. A high consensus on a low rung means "consistent so far," not "proven in people."
Top evidence so far: All trials, pooled (Meta-analysis)
How the studies fall
The evidence (18)
| Source | Grade | Stance | Quality | Finding |
|---|---|---|---|---|
| Gharby S. 2022 · ScientificWorldJournal | observational | supports | moderate | Reviews chemical and physical refining of vegetable oils (soybean, rapeseed, palm, corn, sunflower); states refining removes essential nutrients and 'often generates other undesirable compounds such as 3-MCPD-esters and trans-fatty acids,' which 'dir |
| Makarenko MA 2024 · Foods | in-vitro | supports | moderate | Analytical study found repeated deodorisation cycles of sunflower oil increased saturated fatty acids, MUFAs, trans fatty acids (TFAs), volatile aldehydes, and MCPDE/GE contents with each successive cycle, concluding 'most of these compounds have the |
| Lambelet 2003 · J Agric Food Chem | in-vitro | supports | moderate | Deodorization of rapeseed oil under severe conditions (250C, 5-6 h) produced substantial trans fatty acids (over 5% of total) and cyclic fatty acid monomers, while earlier refining steps barely altered the fatty-acid profile. |
| Qin 2022 · Crit Rev Food Sci Nutr | observational | mixed | low | Highest fried-food (refined-oil) intake associates with modestly higher obesity/T2DM/hypertension risk; confounded proxy for refined-oil harm, not direct |
| Fournier 2006 · J Chromatogr A | in-vitro | supports | moderate | Deodorization at >=220C forms trans geometric isomers of EPA/DHA in refined oils; confirms refining heat generates trans-PUFA isomers |
| Zhang 2021 · Foods | in-vitro | mixed | moderate | Frying conditions modulate 3-MCPD-ester and glycidyl-ester levels in palm oil; both classes noted as high-toxicity refining/processing contaminants |
| Silva 2019 · Food Addit Contam A | in-vitro | supports | moderate | Confirms 3-MCPD/2-MCPD/glycidyl esters form during oil refining (esp. palm); washing can reduce them, implying they are processing artifacts not intrinsic |
| Zhang Y 2023 · Food Chem X | observational | supports | low | [mitigation/below-limits nuance] Review of tiger nut oil extraction technologies, bioactive compounds, oxidative stability, and 'potential hazards,' but abstract gives no specifics on refining-induced contaminants like trans fats, glycidyl esters, or |
| Yao 2019 · J Agric Food Chem | mechanism | supports | moderate | Simulated deodorization plus computational modeling identifies diacylglycerols as precursors and maps the reaction pathway generating 3-MCPD esters during oil refining and heating. |
| Li 2020 · Am J Clin Nutr | meta-analysis | contradicts | high | Higher dietary/biomarker linoleic acid associated with LOWER total, CVD and cancer mortality across cohorts; undercuts 'omega-6 oil = harm' at dietary doses |
| Oey 2019 · Compr Rev Food Sci Food Saf | meta-analysis | supports | moderate | Systematic literature review confirms refining of vegetable oils forms 2-/3-MCPD esters and glycidyl esters, with deodorization the key generating step, and catalogs mitigation strategies. |
| Yıldız K 2024 · Food Technol Biotechnol | mechanism | supports | high | Reviews formation of 3-MCPD esters and glycidyl esters (GE) in edible oils during heat treatment/deodorization; IARC classifies 3-MCPD as group 2B and glycidol as group 2A carcinogens, with toxicological studies showing these genotoxic/carcinogenic c |
| An 2022 · J Oleo Sci | in-vitro | supports | high | Sunflower-oil deodorization (210-270C) progressively raises 3-MCPD esters (0.47->11.2), glycidyl esters (0.24->18.4) and trans fats (0.06->0.70%) |
| Ji Y 2022 · Foods | in-vitro | mixed | moderate | Heating DAG-enriched olive-based oil (baking/frying at 160-180C) found 3-MCPDE actually decreased while GE slightly increased with prolonged heating; final levels stayed low (3-MCPDE <1.25 mg/kg, GE <1.00 mg/kg), with no significant relation to DAG c |
| Yao YY 2026 · Foods | observational | supports | low | [mitigation/below-limits nuance] Broad review of enzyme technology in food processing; mentions enzymatic degradation of harmful compounds like phytic acid, tannins, allergens, and acrylamide, but does not address seed-oil refining, deodorization, tr |
| Arris FA 2020 · Foods | observational | supports | moderate | Review of 3-MCPD as a heat- and process-induced contaminant in vegetable oils (especially palm oil), covering formation routes and mitigation; frames it as a food contaminant with 'potential health risk on humans' above tolerable daily intake. |
| MacMahon 2013 · Food Addit Contam A | observational | supports | high | US market survey: refined oils carry 3-MCPD up to 7.2 ppm and bound glycidol up to 10.5 ppm vs near-absent in unrefined oils; highest in refined palm |
| Oey 2022 · Food Res Int | in-vitro | supports | moderate | Pilot-plant refining of palm oil generated 3-MCPD, 2-MCPD and glycidyl esters that chemical-refining adjustments reduced by roughly 50-73%, confirming these are refining-induced contaminants. |
Disagree, or know a study we missed?
We grade by evidence, not opinions. The way to weigh in is to point us to a study we haven't cited (check the evidence table above first), or to flag a problem with one we have. Every submission is reviewed; if it holds up, the grade updates and shows in Science Changes Its Mind.
Opens a short form. You'll sign in with Google so submissions are tied to a real account — we don't display your identity, and we only accept a link we can verify (PubMed, DOI, ClinicalTrials.gov).
Educational only, not medical advice. Grades and scores reflect published evidence weighted by study design and quality; see the methodology.