Federal and regulatory limits vs field findings
This is the comparison view for standards developers, regulators, retailers, brands, and legal teams. It shows the applicable regulatory limit next to the current field-evidence state. It is not an HMTc pass/fail table.
| Metal | Regulatory limit | Actual field finding | Decision read | Evidence |
|---|---|---|---|---|
| Arsenic, Inorganic inorganic arsenic (iAs) | EU inorganic arsenic maximum level for non-alcoholic rice-based drinks: EU 2023/915 maximum level 30 µg/kg iAs for non-alcoholic rice-based drinks | D’Amato et al. 2026 report 25 Italian rice-based beverages with iAs from 7 to 24 µg/kg; no sample exceeded 30 µg/kg. Good et al. 2026 measured total arsenic in rice milk at 0.64 µg/L (no iAs speciation). | Direct comparison available for rice-based beverages; matrix, analyte species, and basis match for the D’Amato data. Total arsenic from Good et al. is context only. | EU inorganic arsenic maximum level for non-alcoholic rice-based drinks; Inorganic Arsenic in Rice-Based Beverages: Occurrence in Products Available on the Italian Market and Dietary Exposure Assessment; Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milks |
| Cadmium cadmium (Cd) | No harmonized ML for PBMAs outside infant formula. | Good et al. 2026 measured Cd in soy milk at 0.17 µg/L and in oat milk at 0.10 µg/L; the Finnish review compiled soy Cd at 2 to 3.9 µg/L across multiple studies. | No applicable PBMA-specific limit loaded for comparison. Occurrence evidence only. | Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milks; Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A Review |
| Nickel nickel (Ni) | No harmonized ML for PBMAs outside infant formula (EU nickel ML for soy-based infant formula: 400 µg/kg). | The Finnish review compiled soy Ni at less than 25.7 to 201 µg/kg across studies; one Finnish dataset reported a mean of 275 µg/kg, which the Finnish Food Authority flagged as potentially concerning for nickel-sensitized individuals. | No applicable general-adult-consumer PBMA nickel limit for comparison. Risk context available for sensitized individuals only. | Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A Review |
| Lead lead (Pb) | No harmonized ML for PBMAs. | Good et al. 2026 measured Pb broadly similar across all eight milk types: 0.18 to 0.28 µg/L. The Finnish review notes 90 to 100 percent of monitoring samples below LOQ, with occasional elevated values in soy and rice drinks. | No applicable limit loaded for comparison. Occurrence evidence shows broadly low Pb. | Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milks; Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A Review |
| Chromium (Cr, total) | No applicable limit loaded. | Good et al. 2026 measured Cr in hemp milk at 0.68 µg/L (highest of all types) and oat at 0.49 µg/L. No speciation; values are total chromium. | No applicable limit loaded. Total Cr is not a proxy for Cr-VI. | Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milks |
Ingredient drivers
The contamination profile of a plant-based beverage is primarily inherited from its base plant material and the soil and water conditions of that crop.
Rice inherits inorganic arsenic from flooded paddy soils; the arsenic accumulates in the grain and carries into the final beverage. Rice drinks measured in the Italian market ranged from 7 to 24 µg/kg inorganic arsenic (1).
Soy inherits cadmium and nickel from soil; soy plants are moderate cadmium accumulators and show some of the highest cadmium concentrations among common plant-based beverage bases. The Finnish review compiled soy cadmium at 2 to 3.9 µg/L across multiple studies (3).
Hemp and oat show elevated chromium (hemp) and nickel (oat) relative to other PBMA types. Hemp also carried the second-highest total arsenic of the US study at 0.28 µg/L (2).
Aluminium in soy beverages shows the widest variability of any PBDA type (176 to 3,930 µg/L), likely reflecting processing-water quality differences between manufacturers (3).
Controls and mitigation
| Lever category | Specific action | Estimated magnitude | Source |
|---|---|---|---|
| Sourcing | Rice-based beverages: source rice from low-iAs growing regions; Italian market data show iAs range 7 to 24 µg/kg across 25 products despite all products being mostly organic | D’Amato 2026: iAs range 7 to 24 µg/kg; no sample exceeded EU 30 µg/kg | 1 |
| Testing/QC | Use HPLC-ICP-MS speciation for iAs rather than total arsenic; iAs averaged 84 percent of tAs in Italian rice drinks but ranged from 33 to 97 percent at the sample level | D’Amato 2026: %iAs range 33 to 97%; total arsenic alone does not support regulatory compliance against the EU 30 µg/kg iAs cap | 1 |
| Sourcing | Soy-based beverages: ingredient specification for cadmium content in soy; soy carries the highest Cd of all PBMA types in multiple studies | Good 2026: soy Cd 0.17 µg/L (vs cow 0.03 µg/L); review compilation: soy Cd 2 to 3.9 µg/L | 2, 3 |
| Sourcing | Nickel-sensitive consumers: soy-based beverages carry the highest nickel of all PBDA types; oat-based beverages are an intermediate option (oat Ni 127 to 130 µg/kg vs soy 275 µg/kg mean in one Finnish dataset) | Sohlberg 2026: soy Ni mean 275 µg/kg flagged as potentially concerning for nickel-sensitized individuals | 3 |
| Formulation | Blending rice base with lower-iAs plant bases reduces per-serving iAs burden; note that one rice-and-almond blend showed the highest tAs in the D’Amato study (58 µg/kg) due to organic arsenic from almond, while its iAs was low (14 µg/kg) | D’Amato 2026: rice-and-almond blend sample B: tAs 58 µg/kg, iAs 14 µg/kg (33% iAs) | 1 |
Measured concentrations
Rice-based beverages carry the highest inorganic arsenic burden of any plant-based milk alternative studied to date, with Italian market data showing inorganic arsenic ranging from 7 to 24 µg/kg and all samples below the EU maximum level of 30 µg/kg. Soy-based beverages carry the highest cadmium and nickel of any plant-based type, and hemp milk carries the highest chromium; lead concentrations are broadly comparable across all subcategories and cow milk. No harmonized EU or US maximum levels currently apply to heavy metals in plant-based beverages other than inorganic arsenic in rice-based drinks.
What these beverages are
Plant-based beverages include commercially produced liquid products made from soy, oat, almond, rice, coconut, hemp, cashew, spelt, and other plant sources. They are consumed as dairy-milk alternatives by adults, children, and infants. The heavy-metal contamination profile of this category is not uniform across subcategories: each base material (rice, soy, oat, almond) inherits a different metal signature from soil, irrigation water, and agronomic practice. Rice-based beverages carry the most studied and best-regulated heavy-metal concern (inorganic arsenic); soy-based beverages carry elevated cadmium, nickel, and aluminium; oat-based beverages carry elevated nickel and chromium; hemp-based beverages carry elevated chromium and total arsenic.
Subcategory pages with deeper evidence records:
Contamination profile by subcategory
The table below summarizes measured concentrations from the three primary sources in this unit. All values are as-consumed liquid unless otherwise noted. Speciation notation: total arsenic = total arsenic; inorganic arsenic = inorganic arsenic; cadmium = cadmium; chromium = chromium (total); lead = lead; nickel = nickel; aluminium = aluminium.
As-consumed liquid concentrations by milk type, US retail (Good et al. 2026, n = 22 products, ICP-MS)
Values are mean plus or minus standard deviation across brands per milk type (n = 2 to 4 brands per type). Total arsenic only; no inorganic-arsenic speciation was performed in this study.
| Milk type | Cr (µg/L) | tAs (µg/L) | Cd (µg/L) | Pb (µg/L) |
|---|---|---|---|---|
| Cow (reference) | 0.11 plus or minus 0.06 | 0.02 plus or minus 0.01 | 0.03 plus or minus 0.01 | 0.20 plus or minus 0.10 |
| Almond | 0.43 plus or minus 0.63 | 0.07 plus or minus 0.05 | 0.04 plus or minus 0.04 | 0.18 plus or minus 0.07 |
| Coconut | 0.24 plus or minus 0.10 | 0.13 plus or minus 0.04 | 0.05 plus or minus 0.02 | 0.28 plus or minus 0.23 |
| Soy | 0.23 plus or minus 0.05 | 0.08 plus or minus 0.02 | 0.17 plus or minus 0.03 | 0.28 plus or minus 0.06 |
| Oat | 0.49 plus or minus 0.09 | 0.13 plus or minus 0.03 | 0.10 plus or minus 0.04 | 0.26 plus or minus 0.06 |
| Rice | 0.26 plus or minus 0.05 | 0.64 plus or minus 0.27 | 0.04 plus or minus 0.01 | 0.26 plus or minus 0.06 |
| Hemp | 0.68 plus or minus 0.55 | 0.28 plus or minus 0.22 | 0.05 plus or minus 0.02 | 0.21 plus or minus 0.03 |
| Cashew | 0.41 plus or minus 0.05 | 0.05 plus or minus 0.02 | 0.03 plus or minus 0.02 | 0.18 plus or minus 0.02 |
Source: Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milks, Table 1. Brand replication per category is small (n = 2 to 4); values reflect between-brand variability, not population-level distributions.
Inorganic arsenic in rice-based beverages, Italian market (D’Amato et al. 2026, n = 25, HPLC-ICP-MS)
Arsenic speciation was performed; inorganic arsenic (inorganic arsenic = As(III) + As(V)) is the regulatory analyte. Sampling period: April 2022 to March 2023. All products were non-alcoholic rice-based beverages; 88 percent were certified organic.
| Statistic | tAs (µg/kg) | iAs (µg/kg) | iAs as percent of tAs |
|---|---|---|---|
| Mean | 23 | 15 | 84% |
| Median | 22 | 15 | 86% |
| Minimum | 9 | 7 | 33% |
| Maximum | 58 | 24 | 97% |
| EU maximum level | n/a | 30 | n/a |
| Samples exceeding ML | n/a | 0 of 25 | n/a |
Source: Inorganic Arsenic in Rice-Based Beverages: Occurrence in Products Available on the Italian Market and Dietary Exposure Assessment, Table 3. LOQ for inorganic arsenic was 0.7 µg/kg; no left-censoring occurred. The highest total arsenic (58 µg/kg) was in a rice-and-almond blend sample whose inorganic arsenic was only 14 µg/kg (33 percent of species sum), suggesting organic arsenic from the almond fraction. No significant correlation was found between declared rice content and inorganic arsenic concentration (Pearson r = 0.238).
Cross-study concentration ranges compiled from multiple published studies (Sohlberg et al. 2026, review)
The ranges below are compiled by the review from multiple primary studies and span mean values across those studies; they are not single-study measurements. Units vary by study basis; mixed units within a cell are noted.
| Metal | Soy | Oat | Almond | Rice |
|---|---|---|---|---|
| aluminum (Al) (µg/L or µg/kg) | 176 to 3,930 µg/L | 65 to 304 µg/kg | 695 µg/kg | 50 to 375 µg/kg |
| total arsenic (tAs) (µg/kg) | 1.1 to 16 | 0.9 | 0.8 | 12.8 to 16.3 |
| inorganic arsenic (iAs) (µg/L or µg/kg) | not separately tabulated | not separately tabulated | not separately tabulated | 2.3 µg/L to 15.6 µg/kg |
| cadmium (Cd) (µg/L or µg/kg) | 2 to 3.9 µg/L | 0.4 to 0.5 µg/kg | 0.09 µg/kg | 0.6 to 1.2 µg/kg |
| nickel (Ni) (µg/kg) | less than 25.7 to 201 (mean 255 in one dataset) | 127 to 130 | 36.6 | 30 to 57.5 |
| lead (Pb) (µg/L or µg/kg) | 0.3 to 10 µg/L | 0.2 µg/kg | 0.9 µg/kg | 0 to 7 µg/L |
| chromium (Cr) (µg/kg) | 12.9 | 8.3 | 38 | 16.1 |
Source: Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A Review, Table 2. Tin values reported for nut- and coconut-based drinks in the review derive from packaging-migration work and are not listed here as intrinsic beverage contamination.
Applicable regulations
The EU regulatory framework for heavy metals in plant-based beverages is sparse outside of two areas: inorganic arsenic in rice-based drinks and maximum levels within infant formula regulations.
| Instrument | Metal | Product scope | Value | Status |
|---|---|---|---|---|
| EU 2023/915 | iAs | Non-alcoholic rice-based drinks | 30 µg/kg | Binding |
No harmonized EU or US maximum levels for chromium, cadmium, lead, nickel, or aluminium in plant-based beverages other than infant formula are reported in the literature covered by this unit. Good et al. 2026 note that “no harmonized heavy-metal maximum limits currently apply to plant-based milk alternatives” in the US market. Sohlberg et al. 2026 confirm the same regulatory gap for the European market outside of the rice-drink inorganic arsenic limit.
Exposure and risk context
For the general population, all heavy-metal concentrations measured in the Italian rice-beverage study fell below the EU maximum level of 30 µg/kg inorganic arsenic, and the population-averaged dietary exposure from rice drinks was low (MOE 14 for toddlers at population level). However, among toddlers who regularly consume rice drinks, the consumer-mean inorganic arsenic intake was 0.27 µg/kg body weight per day with a margin of exposure of 0.2, which D’Amato et al. 2026 describe as raising a potential health concern for this subgroup (1). The EFSA reference point for inorganic arsenic is 0.06 µg/kg body weight per day (BMDL05 for skin cancer from epidemiological data).
For nickel, the Finnish Food Authority risk assessment cited by Sohlberg et al. 2026 found that mean nickel in soy drinks (275 µg/kg in one dataset) may be of concern for nickel-sensitized individuals at daily consumption levels, though the review does not characterize population prevalence of nickel sensitization (3).
Good et al. 2026 do not compute formal hazard quotients or margin-of-exposure values; they frame the PBMA heavy-metal findings as qualitative concern for chronic low-level exposure in frequent consumers, particularly children and those who rely heavily on plant-based products (2).
Historical recalls and enforcement
No public recalls or enforcement actions specifically targeting heavy metals in plant-based beverages are documented in this unit’s source corpus. The EU maximum level for inorganic arsenic in non-alcoholic rice-based drinks (30 µg/kg, EU 2023/915) is the operative regulatory instrument for this product class in the EU market. This section will be populated when specific enforcement events are ingested.
Related records
- Plant milks, soy-based
- Rice beverages (rice milk)
- Plant milks, non-soy non-rice
- Inorganic arsenic
- Total arsenic
- Cadmium
- Nickel
- Lead
- Aluminium
- Chromium
- EU 2023/915 inorganic arsenic maximum level for rice-based drinks
- Category 5 beverages
Methodology
This page reports what the cited sources say about heavy-metal concentrations in plant-based beverages. HMT&C certification thresholds for products in this category are developed under the certification program at heavymetalcertified.com, not on this page.
Speciation is treated as non-substitutable. Inorganic arsenic (inorganic arsenic) and total arsenic (total arsenic) are separate analytes with different toxicological significance; total arsenic is not a proxy for inorganic arsenic. Total chromium (chromium) is not interpreted as hexavalent chromium (hexavalent chromium) unless the source explicitly speciates hexavalent chromium.
Basis is preserved and labeled. Concentrations are reported as the source states them (as consumed, wet weight). Silent unit conversion is never performed.
Non-detect handling preserves source convention. Where a source reports a value below its LOD or LOQ, this page preserves the source’s handling.
Source pooling across different LOQ thresholds, collection periods, and geographies is not performed. The cross-study ranges compiled by Sohlberg et al. 2026 reflect the range of means across primary studies, not a statistical pooling.
Evidence tiers: A-tier sources are peer-reviewed primary studies and government reports; B-tier are NGO reports and trade publications. This unit contains three A-tier sources.
References
Works cited in this page’s text, in first-appearance order. See Sources for this page’s source inventory. Each title links to a source record with bibliographic details, reported findings, and document provenance.
- Comparative exposure and risk assessment of heavy metals, nutrients, and organochlorine pesticides in cow and plant-based milksPeer-reviewed
- Inorganic Arsenic in Rice-Based Beverages: Occurrence in Products Available on the Italian Market and Dietary Exposure AssessmentPeer-reviewed
- Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A ReviewPeer-reviewed
Sources
Source records associated with this topic. Inclusion does not establish that every finding applies to this product or ingredient. References above identify works cited in the text.
| # | Citation | Year | Type | Used on this page for |
|---|---|---|---|---|
| 1 | Sohlberg et al. 2026. Food Safety and Nutritional Aspects of Plant-Based Dairy and Meat Alternatives: A Review, Food Science and Nutrition | 2026 | Peer-reviewed | Finnish narrative review synthesizing Al, tAs, iAs, Cd, Ni, Pb, Cr, Sn ranges across PBDA types from multiple primary studies; regulatory context for EU MLs; risk context for nickel in soy |
| 2 | Wall et al. 2025. Macromineral and trace element concentrations in conventional and organic milk and plant-based beverages in the UK: implications for population intakes, Food Chemistry 490 (2025) 145083 | 2025 | Peer-reviewed | Contribution to this topic under review; UK mineral and trace element data for plant-based beverages |
| 3 | Ruzik et al. 2022. Speciation of Arsenic(III) and Arsenic(V) in Plant-Based Drinks, Foods 2022, 11, 1441 | 2022 | Peer-reviewed | Contribution to this topic under review; arsenic speciation in plant-based drinks |
| 4 | Walther et al. 2022. Comparison of nutritional composition between plant-based drinks and cow’s milk, Frontiers in Nutrition | 2022 | Peer-reviewed | Contribution to this topic under review; Swiss retail tAs in 27 plant-based drinks across 8 species |
Update history
No substantive edit history is available in this build. The full commit record is available in git.
