Burrell and Exley 2010 — Aluminium in infant formulas

Burrell and Exley measured aluminium in infant formula products from the UK market, covering both ready-made liquid formulas and powders, cow-milk-based and a soya-based product, and first-infant, follow-on, growing-up, preterm, and good-night formulations. Ready-made formulas (n=8) ranged from 175.5 to 700.4 µg/L; the highest was the preterm formula. Powder-format entries in Table 2 (n=8), evaluated as prepared-milk equivalents per manufacturer instructions, ranged from 333.3 to 629.0 µg/L; the highest was the soya-based powder. The authors compared these values to typical breast-milk aluminium concentrations (~15-30 µg/L) and characterized formula aluminium content as 10 to 40 times higher.

Key numbers

• Methods state 15 commercial infant formula products; Tables 1-3 enumerate 8 ready-made liquid rows and 8 powder-format rows (one soya-based, seven cow-milk-based). Five replicate samples per product, each measured three times; replicate accepted if % RSD < 10%.
• Ready-made formula Al, mean concentrations (n=8): range 175.5 to 700.4 µg/L. Replicate maximum 863.0 µg/L in the preterm ready-made product. The preterm formula was the highest mean at 700.4 µg/L (Table 1).
• Powder Al, raw powder basis (n=8): range 2.4 to 4.3 µg/g. Replicate maximum 10.8 µg/g in one follow-on powder (Table 2).
• Powder Al, prepared-milk equivalent (per manufacturer reconstitution instructions): range 333.3 to 629.0 µg/L. The soya-based powder was the highest at 629.0 µg/L (Table 2).
• Estimated aluminium ingestion at 6 months, using manufacturer recommended feeding volumes: range 206 to 592 µg Al/24 h across ready-made and prepared powder products. The maximum 592 µg Al/24 h appears for the soya-based powder and one follow-on powder (Table 3).
• Estimated aluminium ingestion at 12 months (growing-up milks): 88 and 107 µg Al/24 h across the two growing-up milk rows.
• Estimated aluminium ingestion for a preterm infant from very-low-birth-weight (<1 kg) up to term (~2.5 kg): 112 to 263 µg Al/24 h.
• Authors’ comparison anchor: typical breast-milk aluminium reported as ~15-30 µg/L; formula values are 10 to 40 times higher.

Methods (brief)

Samples were taken directly from packaging (ready-made liquids shaken before sampling). Microwave digestion used a 50/50 mixture of 14 M HNO3 and 30% w/v H2O2. Digestates were diluted in ultra-pure water (conductivity <0.067 µS/cm). Aluminium was measured by transversely heated graphite atomiser on a PerkinElmer Analyst 600 atomic absorption spectrometer. Five replicate preparations per product × three measurements per preparation; means accepted only if % RSD <10%. LOD/LOQ are not reported in the published methods.

Ingredient and process signals

The paper identifies cow-milk-based formulas, a soya-based formula, powder and ready-made liquid formats, and a preterm formulation as distinct evidence categories. The high soya-based-powder value (629.0 µg/L prepared-milk equivalent) is discussed in terms of aluminium accumulation in soybean plants and the aluminium tolerance of certain soybean cultivars grown on acid soils. Process contamination from individual ingredients, equipment, storage, and packaging is named as a likely contributor across all formats. [[ingredients/soy]] is the relevant agricultural-platform node for the soya-based finding; [[ingredients/milk-and-dairy]] and [[ingredients/infant-formula-ingredients]] are the formula-platform nodes for the cow-milk-based products.

Implications

This source contributes direct aluminium occurrence data for infant-formula products sampled on the UK market in 2010, with separate evidence for ready-made liquid formulas, powdered formulas prepared according to manufacturer directions, a soya-based powder, and a preterm ready-made formula.

The strongest source-page use is format and ingredient-platform differentiation. Powdered formulas generally produced higher prepared-milk aluminium estimates than ready-made liquid formulas, and the authors discuss the soya-based powder in relation to aluminium accumulation in soybean plants grown on acid soils. The preterm ready-made formula is also a distinct evidence category because the authors discuss infant vulnerability and compare formula concentrations to typical breast-milk aluminium.

The paper provides product-level occurrence and feeding-direction intake estimates, but the wiki stores those as aggregated format descriptors rather than sampled-product names. It does not provide a regulatory maximum level or a safe exposure threshold for aluminium in infant formula.

Wiki pages this source may touch

• infant-formula-powder-non-soy
• infant-formula-powder-soy-based
• infant-formula-rtf-liquid-non-soy
• infant-formula-ingredients
• milk-and-dairy
• soy
• soy-based-infant-formula
• non-soy-infant-formula
• aluminum

Verification notes

• Cross-vendor audit (Codex, 2026-05-17) corrected Table 2 powder row count from n=7 to n=8 and cow-milk-based powder count from n=6 to n=7; reassigned the 863.0 µg/L ready-made replicate maximum to the preterm product per Table 1; removed avoidable product-name-attributed values from Key numbers; and corrected raw_path to the matching SHA-256 PDF.
• The current product taxonomy has no preterm-formula slug; the preterm-formula evidence is captured under infant-formula-rtf-liquid-non-soy for routing purposes but is a distinct subcategory in the literature. Flag for Karen per Part 10: consider infant-formula-preterm-rtf as a future product page if more preterm-formula evidence accumulates.
• This page was enhanced 2026-05-17 from a Mode B (direct PDF) GPT-5.5 fast-mode draft via the verification pipeline. Prior version (2026-04-29) had a legacy ## Summary heading and thin Key numbers (4 rounded bullets); enhanced revision adds replicate-sample structure, EDI estimates, breast-milk comparison anchor, and per-table source citation. GPT draft initially listed product-name-specific values in Key numbers (Part 12 violation); aggregated to format/category level using product-form descriptors (preterm, soya-based, ready-made, powder) without naming sampled products. Modernized ## Wiki pages updated on ingest heading to ## Wiki pages this source may touch.
• Strict Part 12 recheck (Codex, 2026-05-17) removed sampled-product names from the page body. The method instrument vendor/model remains in Methods because the current Part 12 Exception 2 permits scientific-method vendor names for reproducibility.
• Manual-fetch recheck (Codex, 2026-05-18) updated the provenance to the MFK_ PDF path, which is byte-identical to the older raw/studies/1471-2431-10-63.pdf file (SHA-256 3f79353a3f8c325ab6ae57a1b12ccaa3504176286228428bffbba261e3630279), and rewrote the legacy Certification/Courses/App implications into source-evidence context only.

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.