Suggestive maximum limits for lead and cadmium in food contact materials

Overview

The DTU National Food Institute prepared this memo for the Danish Veterinary and Food Administration as a working document proposing national maximum limits for migration of lead and cadmium from ceramic, glass, and enamelled food-contact materials. The memo is a regulatory source, not a peer-reviewed primary occurrence paper; it combines proposed limits, analytical LOQ reasoning, and secondary summaries of migration studies. The routeable HMI signal is food-contact-material leachate and food-simulant context for ceramic and glass tableware, decorated rims/coatings, ceramic-coated/enamelled cookware, bakeware, and infant feeding bottles.

Key numbers

• Request and scope: DVFA asked DTU on 12/8 - 2024 to suggest new national maximum limits for migration of lead and cadmium from ceramic, glass, and enamelled food-contact materials. The memo is dated 12th of December 2024 and carries DTU DOCX no. 24/1014518.
• Table 1 suggested lead limits: Category I non-fillable/shallow articles and rims of drinking vessels, soup plates, and dinner bowls 0.30 µg/dm2; Category II fillable materials, including cups, glasses, mugs, infant feeding bottles, and other fillable articles 1.5 µg/l.
• Table 1 suggested cadmium limits: Category I non-fillable/shallow articles and rims 0.40 µg/dm2; infant feeding bottles 0.35 µg/l; drinking vessels such as cups, glasses, and mugs 0.70 µg/l; other fillable articles including bowls, cookware, and containers 2.0 µg/l.
• Table 2 current Danish/EU-aligned limits copied in the memo: lead (Pb) 800 µg/dm2 for category I, 4000 µg/L for category II, and 1500 µg/L for category III; cadmium (Cd) 70 µg/dm2 for category I, 300 µg/L for category II, and 100 µg/L for category III.
• Prior EU working-group starting values cited by DTU: Pb 3 µg/kg food and Cd 2 µg/kg food, both with an allocation factor of 10 %.
• BfR 2020 analytical-position values cited by DTU: no detectable release with detection limits of 10 μg lead/kg foodstuff (simulant) or 5 μg cadmium/kg foodstuff (simulant), equivalent in the memo’s category-I conversion to 2 μg lead/dm2 and 1 μg cadmium/dm2 assuming 5 square decimetres (dm2)/kg foodstuff.
• Table 3 DVFA method LOQs: Cd 0.062 µg/L, 0.062 µg/kg, and 0.012 µg/dm2; Pb 0.133 µg/L, 0.133 µg/kg, and 0.027 µg/dm2.
• Table 4 analytical limit-value floor: five times LOQ for cadmium gives 0.310 µg/L, 0.310 µg/kg, and 0.062 µg/dm2; ten times LOQ for lead gives 1.328 µg/L, 1.328 µg/kg, and 0.266 µg/dm2.
• DTU lead HBGV used in the memo: 0.05 µg/kg bodyweight pr. day; DTU cadmium reference value: 2.5 μg/kg body weight per week.
• Intake assumptions for limit derivation: 20 g food and beverage/kg bodyweight for storage, cookware, tableware except drinking vessels, and upper rims; 50 g beverage/kg bodyweight per day for drinking vessels; 150 g beverage/kg body weight per day for baby feeding bottles. DTU equates 1 kg food and/or beverage to 1 L food simulant.
• Exposure comparison for lead at 1.5 µg/kg into 20 g food and beverage/kg bodyweight: 5.1 % of EFSA lower-bound mean lifetime dietary exposure for lead; for cups/glasses/mugs under the higher beverage-intake scenario, 12.9 %.
• BfR 2020 heavily decorated ceramic plates from Germany (n = 42, standard 84/500/EEC migration test), Table 5: Pb minimum <1, median 10.0, mean 54.7, 90th percentile 129, 95th percentile 152, maximum 500 μg/dm2; Cd minimum <0.2, median 1.1, mean 24.0, 90th percentile 105, 95th percentile 142, maximum 170 μg/dm2.
• Bulut et al. 2021 Turkish brown ceramic cookware summary: cookware sampled from 13 sales points; n = 96 tests; Pb migration ranged from < 0.10 µg/L to 4.49 µg/L; 2 of 13 products had migration levels > 1.5 µg/L; in around 40 % of duplicate tests, migration increased in the second test.
• Rebeniak et al. 2014 Polish decorated ceramic summary: 187 flat dishes with 174 (93 %) below 10 µg/dm2 Cd and 153 (82 %) below 100 µg/dm2 Pb; 233 deep dishes with 202 (87 %) below 10 µg/L Cd and 153 (66 %) below 100 µg/L Pb; 331 mugs with 325 (98 %) below 10 µg/L Cd and 323 (98 %) below 100 µg/L Pb.
• Mania et al. 2018 Polish decorated ceramic summary: 62 flat ceramic dishes with 60 (97 %) below 10 µg/dm2 Cd and 47 (76 %) below 100 µg/dm2 Pb; 69 deep ceramic dishes with 65 (94 %) below 10 µg/L Cd and 55 (80 %) below 100 µg/L Pb.
• DVFA 2023 Danish ceramic testing summary: 47 ceramic cups, bowls, plates, and ovenproof dishes were tested under standard 84/500/EEC conditions. Four products had Pb migration at or above the LOQ (1.0 µg/l / 0.2 µg/dm2); one cup and one plate were relatively high at 124 µg/l and 14.2 µg/dm2. Three products had Cd migration above the LOQ (0.1 µg/l / 0.02 µg/dm2), and only one product exceeded the memo’s suggested maximum limits.
• Soda-lime glass summary from Rebeniak et al. 2014: 393 beverage-glass samples; Pb migration was not detected in 95 % of samples and Cd migration was not detected in 98 %. Mania et al. 2018 tested nine deep glass dishes and detected neither Pb (LOQ: 100 µg/L) nor Cd (LOQ: 10 µg/L).
• Braikova et al. 2023 crystal-glass summary: traditional crystal glasses containing 24% lead oxide reached lead migration just above 100 µg/L; Cd migration was approximately 90 µg/L in water and 66 µg/L in vodka.
• Haldimann and Zimmerli 1991 crystal-glass summary: four brands exposed to 4 % acetic acid for 24 hours migrated Pb at 71 µg/L to 145 µg/L; 14 brands tested with white wine for 180 min ranged from 14 to 93 µg/L; two products tested with cola gave 115 and 145 µg/L.
• Hight 1996 crystal-glass summary: 24 hours with 4 % acetic acid produced 467 µg/L Pb in solvent, and white wine produced 358 µg/kg. DTU’s consumer-use scenario using Hight’s 6 minutes white-wine value of 154 µg/L estimates 38.5 µg Pb from a 250 ml glass and 0.55 µg/kg bodyweight for a 70 kg adult.
• Rebeniak et al. 2014 Polish crystal-glass summary: 129 crystal-glass samples; one sample had Cd migration 447 µg/L; no Pb sample exceeded the current 4000 µg/L maximum limit; 45 % were within 500 to 2000 µg/L for Pb; 18.6 % were below the detection limit.
• Decorated rims summary: Rebeniak et al. 2014 reported 303 rim samples with 46 (15 %) above 100 µg/product for Pb and 18 (6 %) above 10 µg/product for Cd. Mania et al. 2018 found Pb above 100 µg/product in 2 of 30 ceramic drinking vessels and 2 of 14 glass drinking vessels; Cd above 10 µg/product in 1 of 30 ceramic and 2 of 14 glass vessels.
• Turner 2018 decorated drinking-glass summary: among 72 externally decorated glasses, 52 contained Pb in at least one colour and 51 contained Cd in at least one colour. Maximum enamel levels were almost 400 mg/g Pb and 71 mg/g Cd. In 14 products decorated at or near the rim, Pb migration ranged from 0.03 mg to 33.8 mg, with 11 of 14 exceeding 3 mg; Cd migration ranged from 0.04 mg to 5.4 mg, with 9 of 14 exceeding 0.1 mg.
• Enamelled metal products: the memo states that no data were found on migration levels from enamelled cast iron cookware or ceramic non-stick cookware.

Methods (brief)

The memo is a DTU regulatory working document rather than a laboratory paper. DTU used health-based guidance values, default food/beverage intake scenarios, and DVFA analytical LOQ constraints to propose migration limits. For market-impact context, the memo summarizes available migration studies using the established Council Directive 84/500/EEC ceramic food-contact-material test approach, usually 4 % acetic acid food simulant with 24-hour contact, and includes a short DVFA 2023 internal testing summary for 47 Danish-market ceramic products. The memo keeps migration units on the regulatory basis: µg/dm2, µg/L, µg/kg, µg/product, mg, or mg/g, depending on article type and cited study.

Related evidence

• Ceramicware
• Glassware
• Decorative decals/coatings on tableware
• Ceramic-coated cookware
• Bakeware
• Bottles (plastic, glass, stainless steel)
• Lead
• Cadmium
• EFSA — Lead in Food (2010), No-Threshold Position
• EFSA — Tolerable Weekly Intake for Cadmium, 2.5 µg/kg b.w./week

Update history

The five most recent substantive edits to this page, classified major (evidence or structure moved), correction (a published value or statement was wrong and has been fixed), or minor (narrative rewritten without changing the underlying evidence). Each description is derived from what the edit did to this page; the linked commit is the authoritative record, routine regeneration passes are excluded, and the full version history lives in git.