Heavy Metals in Botanicals: Arsenic, Lead, Cadmium & Mercury

Heavy metals are not added to a botanical extract — they are inherited. Plants are efficient accumulators: they draw water and minerals from the soil, and where that soil or water carries toxic elements, the plant takes them up alongside the nutrients it needs. The four that dominate every specification are arsenic (As), lead (Pb), cadmium (Cd) and mercury (Hg) — the “big four” elemental impurities. They share a quality of being toxic at trace levels, persistent in the body, and impossible to destroy by processing. You cannot cook them out, filter them away reliably, or wash them off a finished powder. Control therefore starts in the field and is proven in the laboratory.

Why these four elements

Of the dozens of trace elements a plant can absorb, regulators and pharmacopoeias converge on the same short list because these four combine real toxicity at low exposure with a genuine likelihood of being present in plant material. They are the elements most consistently specified, most tightly limited, and most likely to fail a lot — which is why a heavy-metals panel almost always means As, Pb, Cd and Hg first, with other elements added by exception.

The big four at a glance

• Arsenic (As) — a metalloid widespread in soil and groundwater, often geologically rather than industrially sourced; toxicity depends heavily on chemical form (species), which makes speciation a live question for some materials.
• Lead (Pb) — historically the most ubiquitous environmental contaminant, from leaded fuel and paint legacies, mining, smelting and contaminated dust settling on foliage; cumulative and a particular concern for sensitive populations.
• Cadmium (Cd) — strongly soil- and fertiliser-linked, taken up readily by certain crops; phosphate fertilisers and sewage-sludge amendments are classic routes into agricultural land.
• Mercury (Hg) — less common in most botanicals but high-consequence; sourced from industrial emissions, artisanal gold mining and certain mineral deposits, with organic mercury species being the most toxic.

Where heavy metals come from

Understanding the entry points tells you where control has to live. Contamination is rarely a single source; it is the sum of geology, agriculture, environment, processing and — occasionally — deliberate adulteration. Each route calls for a different defence.

Soil and geology

The dominant route is the soil itself. Some regions carry naturally elevated background levels of arsenic or cadmium from the parent rock, entirely independent of human activity — a botanical can be organically grown, pesticide-free and still exceed an arsenic limit because the bedrock was arsenic-rich. Soil pH and composition also govern how available a metal is for uptake: acidic soils, for instance, mobilise cadmium. This is why provenance and soil testing at the growing site matter as much as anything done downstream.

Irrigation water and the environment

Irrigation water is a second major route, especially for arsenic, where contaminated groundwater is a well-documented problem in several growing regions. Airborne deposition adds another: mining, smelting, coal combustion and vehicle exhaust settle lead and other metals onto leaves and topsoil, so proximity to industrial zones and busy roads raises risk. Fertilisers and soil amendments contribute too — phosphate fertilisers are a recognised source of cadmium, and sewage-sludge amendments can introduce several metals at once.

Processing and adulteration

Some contamination enters after harvest. Drying on bare ground, milling with worn or non-food-grade equipment, and contact with poorly chosen process aids can all add metals to otherwise clean material. Rarely but seriously, heavy metals appear through adulteration — mineral-based colourants or weighting agents added to inflate mass or improve appearance, or substitution with cheaper, more contaminated material. Adulteration is the one source that defeats good agriculture entirely, which is why identity and authenticity testing sit alongside the metals panel.

Why some botanicals accumulate more

Not all plant parts or species behave alike. Some botanicals are natural accumulators, and knowing which raises the right flags before a single test is run. Plant part is the first variable: roots and rhizomes sit in direct, prolonged contact with soil and tend to carry more soil-derived metal than aerial parts, while leaves accumulate what settles on them from the air. Species matters too — certain plants are physiologically efficient at taking up specific metals.

Higher-risk material types

• Roots and rhizomes — long soil contact makes root botanicals (and many traditional tonics derived from them) a recurring source of arsenic, lead and cadmium.
• Leafy material — broad surface area accumulates airborne deposition and, for some leaf crops, cadmium from soil.
• Seaweeds and other algae — marine botanicals concentrate elements from seawater and are a classic high-arsenic material, where speciation becomes essential to interpret the result sensibly.
• Fungi and mushroom-derived ingredients — known to accumulate cadmium and other elements from their growth substrate.
• Concentrated mineral-rich botanicals — any material prized for its mineral content can, by the same biology, carry toxic elements alongside the beneficial ones.

How extraction changes the picture

A critical and frequently misunderstood point: extraction can concentrate heavy metals, not remove them. When an extract is made at a 10:1 ratio, ten kilograms of raw herb become one kilogram of extract — and any metal that follows the active into the extract is now present at up to ten times the concentration of the starting herb. A raw material that comfortably passes at the herb stage can fail at the extract stage purely through concentration. This is why a heavy-metals result on the raw herb never substitutes for testing the finished extract.

Solvent partitioning

Whether a metal concentrates depends on whether it follows the solvent. Water and hydro-alcoholic extraction can mobilise water-soluble metal species into the extract; non-polar solvents and supercritical CO₂ tend to leave most ionic metals behind in the spent biomass. So an oil or CO₂ extract often carries a lower elemental burden than a water extract of the same herb — but this is a tendency, not a guarantee, and it must be confirmed by testing the actual extract rather than assumed from the process.

Testing: ICP-MS and ICP-OES

Modern heavy-metals testing is elemental and quantitative — a far cry from the old colourimetric “total heavy metals” limit tests, which were non-specific and have been formally retired in favour of instrumental methods that name and quantify each element. Two techniques dominate, and the distinction matters when reading a method on a COA.

ICP-OES vs ICP-MS

Both start by introducing the digested sample into an inductively coupled plasma (ICP) — an argon plasma hot enough to atomise and excite the elements. ICP-OES (optical emission spectroscopy, sometimes written ICP-AES) reads the light each element emits; it is robust, cost-effective and well suited to elements present at higher concentrations. ICP-MS (mass spectrometry) instead counts ions by mass-to-charge ratio, delivering far lower detection limits — into the parts-per-billion range and below — with the ability to resolve individual elements and even isotopes.

For the big four at the low limits now in force, ICP-MS is the de facto standard: its sensitivity sits comfortably below the specification, so a result is a genuine measurement rather than a “less than the detection limit” shrug. Both methods require proper sample preparation — typically closed-vessel microwave acid digestion — to bring the solid extract fully into solution; a careless digestion is a common hidden source of unreliable results. Mercury in particular needs care, as it is volatile and prone to loss during preparation, sometimes warranting a dedicated technique.

Limit frameworks that apply

A heavy-metals number means nothing without the framework it is judged against. Several overlapping systems set the limits, and a robust specification states explicitly which one it follows. The dominant frameworks for botanical ingredients are the USP general chapters and the ICH elemental-impurities guideline.

USP <232> / <233> and ICH Q3D

USP <232> defines the permitted limits for elemental impurities, while USP <233> defines the validated analytical procedures (centred on ICP-OES and ICP-MS) used to demonstrate compliance. The two are designed as a pair: <232> says how much is allowed, <233> says how to prove it. These align closely with ICH Q3D, the harmonised guideline on elemental impurities that organises elements into classes by toxicity and likelihood, and sets permitted daily exposures (PDEs) translated into concentration limits based on dose. Class 1 — arsenic, cadmium, lead and mercury — is precisely the big four, reflecting their status as the highest-priority elements to control.

Regional frameworks and context

• EU frameworks — European Pharmacopoeia and EU food/contaminant regulations set maximum levels for lead, cadmium, mercury and arsenic in foods and food supplements, with limits that vary by category.
• California Proposition 65 — not a pharmacopoeial limit but a US right-to-know exposure framework with stringent thresholds for lead and other listed substances; products sold into California are frequently held to these levels, which can be tighter than general food limits and effectively shape national specifications.
• Pharmacopoeial monographs — individual herb or extract monographs may carry their own specific heavy-metals limits that take precedence for that material.

Limits are commonly expressed in ppm (parts per million, equivalent to mg/kg) or, for exposure-based frameworks, as a permitted daily intake in micrograms. Because dose-based limits (ICH Q3D, Prop 65) depend on how much of the product a person consumes per day, a single concentration limit only makes sense once the intended daily dose is fixed — a higher daily dose forces a tighter concentration limit. A specification that cites a framework without its basis is incomplete.

Arsenic speciation and why form matters

For most elements, the total amount is what is limited and reported. Arsenic is the important exception. Arsenic exists in inorganic forms (arsenite and arsenate) and organic forms (such as arsenobetaine), and these differ enormously in toxicity: inorganic arsenic is the highly toxic, regulated species, while several organic forms are comparatively benign. A material such as seaweed can show an alarmingly high total arsenic while being dominated by low-toxicity organic species — so a total-arsenic result alone can be misleading for high-arsenic botanicals.

Speciation analysis — typically HPLC coupled to ICP-MS — separates the species and quantifies the inorganic fraction that actually drives the limit and the risk. For most low-arsenic botanicals, total arsenic against a conservative limit is sufficient and proportionate. For known accumulators (seaweeds, certain marine and mineral-rich materials), inorganic-arsenic speciation is the meaningful test, and a specification for those materials should call for it explicitly rather than leaning on a total figure that flatters or alarms without informing. The same logic applies, more rarely, to mercury, where organic (methylmercury) species are the toxicologically critical fraction.

Source, effect and control at a glance

The four elements differ in where they come from and how they are best controlled. This table maps each to its dominant sources and the practical levers that keep it within specification — a working reference for both sourcing decisions and lot investigation.

How to specify heavy metals

A defensible specification does more than name a number. It names the elements, the limit and the framework behind it, the method, and the basis on which the limit is set. Writing it tightly is the single most effective control a buyer holds, because it dictates exactly what the supplier must demonstrate before a lot can ship.

1. 1Name each element explicitly — As, Pb, Cd and Hg as separate line items, not a single combined “heavy metals” figure, plus any additional elements your market requires.
2. 2State the limit and its framework — e.g. conformance to USP <232>/<233>, ICH Q3D Class 1, the relevant EU level, or a Prop 65-aligned lead limit, with the value in ppm (mg/kg) or as a daily-exposure basis.
3. 3Fix the dose basis where the limit is exposure-driven — a dose-based limit is only meaningful tied to the intended daily intake.
4. 4Specify the method — ICP-MS (or ICP-OES where sensitivity allows), with reference to USP <233> validated procedures and a digestion approach appropriate to the matrix.
5. 5Call for speciation where it matters — inorganic-arsenic speciation by HPLC-ICP-MS for seaweeds and other arsenic accumulators.
6. 6Require testing on the finished extract — at the shipped extraction ratio, not only on the raw herb.

What the COA must show

A complete certificate of analysis turns the specification into evidence. It should report a discrete result for each of arsenic, lead, cadmium and mercury — each with its own limit and a clear pass against it — never a single lumped value. It should state the method (ICP-MS or ICP-OES) and, ideally, the limit framework it conforms to. Where a result is below the method's detection limit, the COA should give that limit (e.g. “< 0.01 ppm”) so you can confirm the method was sensitive enough to sit below the specification, rather than a bare “none detected” that could hide an insensitive technique. For accumulator materials, the inorganic-arsenic figure should appear alongside or in place of total arsenic.

Questions to ask your supplier

Before you accept a heavy-metals claim into your own specification and onto your own label, work through these questions in order. The answers separate a controlled supply chain from a hopeful one.

1. 1Is heavy-metals testing performed on the finished extract at its shipped ratio, or only on the raw herb?
2. 2Which elements are tested and to which framework — USP <232>/<233>, ICH Q3D, EU levels, Prop 65 — and at what numeric limits in ppm?
3. 3What method is used (ICP-MS or ICP-OES), and what are the detection limits relative to the specification?
4. 4For arsenic-accumulating materials, do you run inorganic-arsenic speciation, and what fraction is inorganic?
5. 5Where is the material grown, and do you have soil and irrigation-water data for the source region?
6. 6How do you guard against adulteration, and is identity/authenticity confirmed alongside the metals panel?
7. 7Does the COA report a discrete pass for each of As, Pb, Cd and Hg, with the detection limit shown for any “not detected” result?
8. 8Is every lot tested, or do you test by skip-lot/periodic schedule — and on what basis?