BfR 2020 — Aluminium in food and consumer products (FAQ)

The German Federal Institute for Risk Assessment (BfR) issued this consumer-facing FAQ on 20 July 2020 to consolidate its position on aluminium exposure from food, food-contact materials, cosmetics, and medicines, incorporating a 2020 revision of its earlier antiperspirant assessment in light of new SCCS data on dermal uptake of aluminium chlorohydrate. The document reaffirms that food remains the dominant exposure pathway — the majority of adolescents and adults in Germany already consume roughly half of the EFSA tolerable weekly intake of 1 mg Al/kg body weight from food alone — and that aggregate intake from food plus food-contact migration plus cosmetics plus medication can exceed the EFSA TWI, while remaining below the higher JECFA provisional TWI of 2 mg/kg bw/week. The revised antiperspirant position is that dermal absorption under realistic application conditions is considerably lower than the BfR’s 2014 and 2019 assessments had estimated, making cosmetics a small contributor to total intake relative to food.

Key numbers

Health-based guidance values

• EFSA tolerable weekly intake (2008): 1 mg aluminium per kg body weight per week, derived from a study on developmental disorders in young rats.
• JECFA provisional tolerable weekly intake (2012): 2 mg aluminium per kg body weight per week, derived from a more recent (2011) study on developmental disorders in young rats.

Exposure summary (German population, BfR 2019 total-intake estimate referenced)

• Average aluminium concentration in processed and ready-to-eat foods: less than 5 mg/kg fresh mass; few foods exceed 20 mg/kg fresh mass (BfR pilot total diet study cited).
• The majority of the population, especially adolescents and adults, already consumes approximately half of the EFSA TWI (1 mg/kg bw/week) through food alone. Aggregate intake from food plus food-contact materials plus cosmetics plus medication can exceed the EFSA TWI for the majority of the population.
• The JECFA provisional TWI of 2 mg/kg bw/week is not exceeded by the majority of the population from aluminium intake.

Top food contributors to adult dietary aluminium intake (Germany)

• Instant tea drinks: 11% of total food intake
• Mixed raw vegetable salads: 8%
• Tea drinks: 7%
• Cocoa and chocolate products: 6%
• Multigrain bread: 4%
• Remaining food groups (combined): 64%

Lye pastries (lye pretzels)

• Official control laboratories continue to detect aluminium concentrations exceeding 10 mg/kg in baked lye products, attributed to dipping dough into lye before baking on aluminium baking trays. The BfR’s 2002 recommendation to reduce transfer of aluminium to lye pastries to the technically feasible minimum has not been fully implemented.

Food-contact-material release

• Council of Europe resolution on metals and alloys: suggested specific release limit of 5 mg aluminium per kilogram of food. This is a recommendation; it is not a legally binding limit value. There is no specific EU release limit for ceramic articles. Regulation (EC) No 1935/2004 stipulates that food contact materials may not transfer constituents in quantities endangering human health; in the absence of specific limits, a toxicological individual assessment is required.
• BfR Opinion No 007/2017 (29 May 2017) on uncoated aluminium menu trays: high quantities of aluminium ions are released from uncoated aluminium trays to food, especially when used to keep food warm in the cook-and-chill catering process. Aluminium espresso makers release aluminium below the 5 mg/kg Council of Europe release limit during normal use.

Cosmetic-product aluminium content (scientific-literature values cited by BfR)

• Antiperspirants (roll-ons): industry-reported aluminium chlorohydrate concentrations up to approximately 30%, corresponding to a maximum aluminium concentration of about 7.5%. Scientific-literature average and maximum aluminium concentrations: approximately 2.8% and 5.8% respectively.
• Lipsticks: up to about 3% aluminium.
• Whitening (abrasive) toothpaste: up to 4.5% aluminium.
• Sun creams: up to 0.8% aluminium.

Infant feeding

• Infant formula and follow-on formula contain on average significantly higher aluminium concentrations than breast milk; breastfed infants therefore absorb significantly lower amounts of aluminium salts than non-breastfed infants. Specially adapted baby foods (soy-based, lactose-free, hypoallergenic) can contain significantly higher amounts of aluminium than standard formulas. The BfR advises women to exclusively breastfeed infants up to six months where possible, then transition to normal food.

Antiperspirants — revised 2020 dermal-uptake position

• The 2020 BfR revision, based on new human data made available to the SCCS by the cosmetics industry under realistic application conditions, concludes that consumers absorb significantly less aluminium dermally than was calculated from the data underlying the BfR’s 2014 and 2019 assessments. Antiperspirants with aluminium chlorohydrate may be used daily; a health risk for consumers is unlikely according to current scientific knowledge.

Health endpoints summarised

• Hazard focus: nervous system, mental and motor development of children, kidneys, bones. Acute toxicity from food ingestion is low; most absorbed aluminium is excreted via the kidneys in healthy people. In chronic renal insufficiency, excretion is impaired and aluminium accumulates (skeletal system, muscles, kidneys, liver, brain). Once stored, aluminium is excreted only very slowly. Aluminium crosses the placenta.
• Alzheimer’s disease: studies have attempted to establish a connection between aluminium absorption and Alzheimer’s disease but found no clear evidence; a connection is unlikely on current information, though final assessment is precluded by the inconsistent data situation.
• Breast cancer: no causal relationship between antiperspirant aluminium intake and breast cancer development has been demonstrated. Findings of elevated aluminium in breast cancer tissue cannot establish direction of causation, and other metals (iron, chromium, nickel) were also elevated. Epidemiological studies are inconsistent. According to current knowledge, antiperspirant use is unlikely to cause breast cancer.
• Vaccines: clinical and epidemiological studies show that aluminium exposure from infant and toddler vaccines can be considered safe from a health point of view; the BfR refers questions on vaccine effects and adverse effects to BfArM, PEI, and RKI.

Regulatory framework referenced

• Commission Regulation (EU) No 380/2012 (3 May 2012) amending Annex II to Regulation (EC) No 1333/2008 restricted the conditions of use and quantities applicable to aluminium-containing food additives. EU permission for basic aluminium in food is limited to coating confectionery and decorating cakes and fine pastries, subject to the “quantum satis” principle; certain food colourants may also be used as aluminium coatings.
• Regulation (EC) No 1935/2004 on food contact materials: general safety requirement that materials not transfer constituents in quantities endangering human health.
• Council of Europe resolution on metals and alloys: non-binding suggested release limit of 5 mg Al/kg food for metal/alloy food-contact articles.

Practical exposure-reduction guidance issued

The FAQ records the BfR’s consumer-facing positions on reducing avoidable aluminium exposure:

• The BfR states that acidic or salty foods should not be stored, heated, or kept warm in aluminium foil, uncoated aluminium grill trays, uncoated aluminium food trays, or uncoated aluminium cooking pots and pressure cookers, because aluminium solubility increases under acid and salt exposure. Examples the BfR lists include cut apples, tomatoes, rhubarb, salted herring, marinated fish, cheese, apple purée, and tomato purée. The FAQ identifies reusable stainless steel trays and coated aluminium trays as the BfR’s recommended alternatives.
• The BfR considers aluminium-tray use for grilling justifiable because it prevents fat-dripping-induced PAH formation; the FAQ recommends salting and seasoning grilled meat only after cooking, and notes stainless steel or ceramic grill trays as alternatives.
• The BfR records that aluminium espresso makers contribute only a small share of total exposure and sees no reason to advise against their use, but recommends not washing them in a dishwasher (which removes the protective oxide layer). The FAQ states that aluminium espresso capsules are coated on the inside and do not release aluminium into the beverage, and that thermal flasks are generally constructed so that aluminium parts do not contact the liquid.
• The BfR states that reducing or omitting whitening (abrasive) toothpaste use reduces aluminium absorption from cosmetics.
• The BfR’s general dietary recommendation in the FAQ is variety of food sources to reduce partial exposure to any single contaminant, including aluminium.

Verification notes

Routing-frontmatter decisions made during ingest:

• products: bread-and-baked-goods added to route the FAQ’s substantive lye-pretzels finding (BfR detection of >10 mg/kg in baked lye products, with the 2002 reduction recommendation not yet fully implemented) onto the broad baked-goods product page rather than only the bread ingredient page. The taxonomy snapshot has no pretzel-specific or lye-pastry slug; bread-and-baked-goods is the appropriate umbrella.
• ingredients: vegetables added for the BfR’s “mixed raw vegetable salads (8%)” finding among top adult dietary Al contributors. The taxonomy snapshot offers vegetables, leafy-vegetables, non-root-vegetables, and several specific vegetables; the BfR’s “Rohkostsalat” framing covers a heterogeneous raw-vegetable category (not exclusively leafy), so the broadest accurate slug vegetables is chosen.
• infant-formula-powder only (not soy/non-soy split) — the BfR FAQ contrasts breast milk vs infant/follow-on formula in general, and notes that “specially adapted baby foods (soy-based, lactose-free, hypoallergenic) can contain significantly higher amounts of aluminium” without giving a soy-vs-non-soy split for standard infant formula. The routing audit’s broad-powder-context rows for both soy/non-soy locked rows are appropriate; this source does not itself resolve the soy/non-soy row split.
• matrices: [] — this is a risk-communication FAQ with no analytical matrix of its own; the analytical matrices are in the underlying BfR opinions, not in this document. Leaving the matrices array empty triggers an advisory-severity malformed entry in the routing audit, consistent with Karen’s standing decision to treat optional-routing-input absence as advisory rather than blocking.

Audit-application notes:

• Audit subagent (2026-06-03) flagged missing [[products/bread-and-baked-goods]] slug; verified against PDF p.5 lye-pretzels section — added.
• Audit subagent (2026-06-03) flagged missing vegetable ingredient slug for the “mixed raw vegetable salads (8%)” finding; verified against PDF p.3 — added [[ingredients/vegetables]].
• Audit subagent (2026-06-03) flagged that the Practical exposure-reduction guidance bullets read as wiki-voice imperatives rather than BfR-attributed positions; verified against PDF pp.2-3, p.5, p.7 — bullets reframed to clearly attribute each recommendation to the BfR.

Methods (brief)

Risk-communication FAQ. No original measurement, no sample frame, no analytical method, no exposure calculation produced for this document. The underlying data and positions are drawn from prior BfR scientific opinions (2002 lye pastries; 2014 and 2019 risk assessments on aluminium; 2017 Opinion 007/2017 on uncoated aluminium menu trays; 2019 total Al intake estimate; 2020 revised antiperspirant assessment incorporating SCCS data), from the EFSA 2008 AFC panel TWI derivation, and from the JECFA 2012 provisional TWI derivation. Cosmetic-product aluminium concentration ranges are sourced from industry data and scientific literature referenced by BfR but not enumerated in the FAQ itself.

Implications

Certification: This document records the formal BfR position that food is the dominant aluminium exposure pathway in Germany, that aggregate intake from all sources can exceed the EFSA TWI for the majority of the population, and that lye pastries continue to be detected above 10 mg/kg despite an 18-year-old reduction recommendation — useful regulatory-context anchors for any HMT&C aluminium threshold deliberation that references the EFSA TWI as the literature baseline. The 2020 BfR revision on antiperspirant dermal uptake materially shifts the relative weighting of cosmetic vs dietary exposure compared with the BfR’s own 2014/2019 view.

Courses: The food-category contribution rankings (instant tea 11%, raw vegetable salads 8%, tea 7%, cocoa/chocolate 6%, multigrain bread 4%) and the BfR’s specific consumer guidance on acidic/salty foods plus uncoated aluminium contact surfaces provide concrete, agency-sourced talking points for QA-team and consumer-facing course material.

App: The cosmetic-product aluminium concentration ranges (antiperspirants ~2.8–5.8% Al; lipstick up to 3%; whitening toothpaste up to 4.5%; sun cream up to 0.8%) and the lye-pretzel >10 mg/kg finding can populate Al contamination-profile data points on the corresponding product pages.

Page history

The five most recent substantive edits to this page. The full version history lives in git; when DOI minting comes online (see schema docs), each entry below will also link to a version-pinned DataCite DOI.