USP <2232> Elemental Contaminants in Dietary Supplements: What Your Supplier's Heavy Metals Panel Isn't Telling You

Most supplement brands we work with have received a supplier COA listing “heavy metals: compliant” — and stopped there. It’s the QA equivalent of a mechanic telling you your car “passed inspection” without specifying what was checked, by what method, or against what standard. USP <2232> Elemental Contaminants in Dietary Supplements doesn’t work that way. It covers 15 distinct elements, ties limits directly to your product’s actual labeled daily serving size, and requires an analytical method sensitive enough to detect impurities at sub-ppb concentrations.

There’s a real gap between what a conventional 4-metal panel shows and what USP <2232> compliance actually demands. That gap matters not just for FDA audits under 21 CFR 111, but for the genuine contamination risk that some of these elements pose when botanical ingredients arrive from high-burden growing regions.

What USP <2232> Actually Covers — And What a 4-Metal Panel Misses

The conventional heavy metals panel screens for four elements: arsenic (As), cadmium (Cd), lead (Pb), and mercury (Hg). These get the most attention for good reason — they’re acutely toxic and depressingly common in botanical supply chains. But USP <2232> extends the required screening to 15 elemental impurities, including elements that most supplier COAs never touch:

• Antimony (Sb) — migrates from PET packaging and certain pigments; detected in fine herbal powders stored in plastic-lined drums
• Chromium (Cr) — present in agricultural soils; hexavalent chromium (Cr⁶⁺) is classified as an IARC Group 1 carcinogen
• Copper (Cu) — an essential trace mineral, but toxic at sustained supplemental intakes above ~10 mg/day
• Nickel (Ni) — a consistent soil contaminant that appears regularly in turmeric and ginger root from South Asian growing regions
• Selenium (Se) — physiologically beneficial at 55–200 µg/day but acutely toxic above 400 µg/day, a window that matters for high-dose botanical blends
• Thallium (Tl) — rare, but its Permitted Daily Exposure (PDE) is just 8 µg/day, leaving almost no margin in concentrated botanical extracts
• Vanadium (V) — present in certain root extracts; its PDE of 300 µg/day is derived from occupational exposure data

A supplier that tests against a 4-metal panel and stamps “compliant” has told you something true but dangerously incomplete. Chromium, nickel, and selenium aren’t exotic edge-case analytes — they appear routinely in Ayurvedic botanicals. A 2013 PLOS ONE study (Saper et al.) testing 193 Ayurvedic herbal products sold in the US found detectable lead in 65% of samples and chromium in the majority — neither chromium nor any of the 11 elements beyond the standard four would be captured by a conventional heavy metals screen.

Why ICP-MS Is the Required Method — and What the Sensitivity Gap Actually Means

Older heavy metals testing approaches — colorimetric limit tests and flame atomic absorption spectroscopy (FAAS) — were built for narrower analyte sets and higher concentration thresholds. They work adequately for a pass/fail screen against traditional pharmacopoeial heavy metals limits. But they’re not designed for the 15-element USP <2232> profile, and they genuinely cannot achieve the detection limits some elements require.

ICP-MS (Inductively Coupled Plasma Mass Spectrometry) works differently. The digested sample is introduced into a plasma torch operating at approximately 7,000–10,000 Kelvin — hot enough to fully atomize and ionize every element present. A mass spectrometer then separates those ions by mass-to-charge ratio and counts them with extraordinary precision. Routine detection limits land between 0.1 and 1 µg/kg (sub-ppb), roughly 10,000× more sensitive than traditional colorimetric methods.

That sensitivity matters practically, not just analytically. Thallium’s PDE under USP <2232> is 8 µg/day. At a 2 g/day serving of a botanical extract, the allowable thallium concentration calculates to just 4 µg/kg — a concentration you cannot reliably detect or quantify with FAAS. ICP-MS is the only practical method for enforcing USP <2232>‘s low-PDE elements.

The other advantage of running an accredited analytical testing laboratory on ICP-MS: a full 24-element sweep takes similar instrument time to a 4-element AAS panel once the method is validated. There’s no sound analytical reason to screen narrowly when you’re already running ICP-MS. Our team runs a complete elemental profile on all incoming botanical lots as standard practice — the incremental cost of broader coverage is negligible once the method is operating, and the additional information frequently catches contamination that a narrow panel would pass.

The Arsenic Speciation Problem — Why Total Arsenic Is Often the Wrong Number

Arsenic is where the most confusion lives, and it deserves careful attention.

Total arsenic on a COA is nearly meaningless in isolation. Toxicological risk from arsenic depends almost entirely on chemical form. Inorganic arsenic species — arsenite (As³⁺) and arsenate (As⁵⁺) — are IARC Group 1 carcinogens and potent nephrotoxins with no safe lower threshold. Organic arsenicals like arsenobetaine, the form predominant in seafood, are largely metabolically inert and rapidly excreted in urine.

For most terrestrial botanical ingredients, this distinction matters less: plants absorb arsenic from soil primarily as inorganic species, so total arsenic reasonably approximates inorganic arsenic exposure. But the calculation changes meaningfully for marine-derived botanicals. Spirulina, chlorella, bladderwrack, and other algae-based ingredients can carry elevated total arsenic loads where a meaningful fraction is organic arsenobetaine — clinically far less concerning than the total number would suggest.

HPLC-ICP-MS speciation testing is the only way to split those fractions. A standard total-arsenic ICP-MS result does not give you speciation — you have to request it explicitly, and most contract labs don’t include it in a default panel. For algae-based ingredients, and for any product with California or Amazon national distribution, this matters immediately. California Prop 65 applies its 0.5 µg/day Maximum Allowable Dose Level (MADL) specifically to inorganic arsenic. A spirulina ingredient reporting 1.5 ppm total arsenic might have 0.2 ppm inorganic arsenic — well within Prop 65 — or 1.4 ppm inorganic — significantly over. Without speciation data, you’re making a compliance decision without the relevant number.

PDE-Based Limits: Why the Math Is Product-Specific and Non-Transferable

This is the part that catches even experienced QA teams off guard, and it’s worth working through the logic carefully.

USP <2232> doesn’t specify universal concentration limits for each element that apply regardless of product or dose. Instead, it establishes PDEs — the maximum daily intake considered safe for each element from all dietary supplement sources combined. The allowable concentration in any specific raw material then depends on how much of that ingredient a consumer actually takes per day.

The calculation is straightforward: allowable concentration (µg/g) = PDE (µg/day) ÷ daily dose (g/day)

But supplement brands routinely make two errors with this. First, they accept a supplier COA calculated against a generic or maximum dose rather than their product’s actual labeled serving. If your ashwagandha KSM-66 extract is dosed at 600 mg/day and the supplier ran PDE calculations at 2,000 mg/day, that COA doesn’t transfer to your formulation — your effective concentration limits are more than 3× stricter. The COA may say “pass” while your product is out of specification.

Second, brands forget to allocate the PDE across all contributing ingredients in a multi-botanical blend. If a greens formula combines ashwagandha, turmeric, and moringa, lead exposure from all three sources must sum within the 10 µg/day PDE — you can’t allocate 10 µg/day to each ingredient independently. In practice, this means that for high-dose botanical products (5+ grams of total botanical content per serving), the per-ingredient lead concentration limit may need to be well below 2 ppm to leave allocation room for all components.

Running the PDE math against your specific formulation isn’t optional — it’s the only way the numbers on a COA are actually meaningful for your product. A generic “10 ppm lead limit” check tells you almost nothing unless it’s been calculated against your actual label and your actual serving size.

What a USP <2232>-Compliant ICP-MS Package Includes in Practice

When brands ship raw botanical samples to our Countryside, IL receiving facility, a USP <2232>-compliant analytical testing package typically covers:

1. Full 24-element ICP-MS panel — all 15 elements specified in USP <2232> plus additional analytes for a complete elemental fingerprint of the incoming lot
2. USP <233>-compliant sample preparation — microwave-assisted acid digestion with spike recovery verification, method blanks, and internal standards (Sc, In, Bi) for matrix interference correction
3. NIST-traceable calibration — concentration standards referenced to NIST SRM 3100-series elemental solutions, with traceability documented on the CoA
4. Serving-size-adjusted PDE compliance table — calculated against the brand’s actual labeled daily dose; pass/fail is stated per element for the specific product, not a generic ingredient profile
5. Arsenic speciation flag — automatically triggered for algae-derived and marine botanical ingredients; HPLC-ICP-MS speciation testing available as an add-on with 2–3 additional business days
6. ISO 17025-accredited CoA — issued under Qalitex Laboratories’ California accreditation, typically within 5–7 business days of Chicago receipt

The botanicals that consistently warrant USP <2232> full-panel testing based on what we see across incoming lots: ashwagandha root and standardized extracts, turmeric rhizome and curcumin concentrates (India-origin), shilajit resin and purified extracts (highly variable profiles across suppliers), spirulina and chlorella powders, black seed (Nigella sativa) from Middle Eastern supply chains, and any ingredient with undisclosed geographic origin on the supplier paperwork.

---

If your incoming QC program still relies on a 4-metal COA from the supplier, it’s worth pulling out 21 CFR 111.75 and reading the purity specification requirements carefully. “Purity” under DSHEA means testing against your own established specifications — and if your specifications don’t reflect the full USP <2232> element list, you’re not meeting the standard the regulation requires. A COA that says “passes heavy metals” without specifying the element list, the analytical method, or the serving-size-adjusted calculation doesn’t give you a defensible position when an investigator asks how you verified purity for incoming raw materials.

Request 24-element ICP-MS for high-risk botanical lots. Provide your actual labeled serving size so PDE calculations reflect your product. Flag marine-derived ingredients for arsenic speciation. And run the multi-ingredient PDE allocation math before accepting any individual ingredient result as sufficient for a blended formulation.

That’s the difference between a COA that clears your receiving dock and one that holds up in an audit.

---

Written by Nour Abochama, VP Operations, Qalitex | Quality Consultant, Ayah Labs. Learn more about our team

Ship your sample to our Chicago facility — get a Qalitex CoA in 5–7 days. Contact us

Related from our network

• ISO 17025-Accredited ICP-MS Heavy Metals Testing for Dietary Supplements — Qalitex Laboratories operates the accredited ICP-MS instrumentation and issues the CoAs behind every Ayah Labs elemental analysis
• USP <232>/<233> Elemental Impurities Testing for Pharmaceutical and Supplement Manufacturers — Deep-dive technical resources on method validation, internal standards, and California Prop 65 compliance for supplement brands