USP <232> and <233>: Why the Elemental Impurities Framework Still Trips Up Supplement Manufacturers

Every few months, we receive a certificate of analysis from a supplement brand that lists “Heavy Metals: Passes USP <231>” — and the client is genuinely surprised when we explain that’s no longer a sufficient specification for their product. Not because <231> is wrong, exactly. But because USP officially retired that chapter as a general heavy metals limit test back in 2018, and a meaningful portion of the industry still hasn’t fully absorbed what replaced it.

Or they heard about it, filed the guidance document, and kept doing things the old way.

Eight years on, the elemental impurities framework — codified in USP <232> and <233> and harmonized internationally through ICH Q3D — still generates more confusion per page of guidance than almost any other testing topic that crosses our desks. So let’s break down what actually changed, what it means for your finished dietary supplement, and why the difference between a colorimetric screen and a full ICP-MS panel matters more than most brands realize.

What USP <231> Was — and Why It Was Retired

The old USP <231> General Chapter, “Heavy Metals,” was a colorimetric limit test. You dissolved your sample, added thioacetamide or sodium sulfide reagent, and compared the resulting color against a lead standard. If your sample matched or ran lighter than the standard, you passed. That was the whole test.

For raw material and ingredient-level verification, Ayah Labs specializes in contract testing and supplier qualification.

It was designed in an era when lead was the primary heavy metal concern and modern analytical instrumentation was expensive and scarce. As a broad screening tool, it served its time. But the problems were fundamental:

It wasn’t element-specific. You got a single pass/fail answer expressed as lead equivalents. Arsenic, cadmium, mercury — any element that didn’t react predictably with the colorimetric reagent could be missed entirely or dramatically underestimated depending on the matrix.

Sensitivity was limited. The standard colorimetric method had a practical detection threshold around 10–20 ppm for lead equivalents. Modern ICP-MS routinely achieves detection limits below 0.001 ppm — three to four orders of magnitude more sensitive. For a Class 1 element like cadmium with a Permitted Daily Exposure of just 5 μg/day, that gap is not academic.

And the science had simply moved on. ICH Q3D, the international guideline on elemental impurities in pharmaceutical products, was finalized in 2014 after years of development. It established toxicology-based Permitted Daily Exposure (PDE) values for 24 specific elemental impurities, organized by their inherent toxicity and likelihood of appearing in drug products. USP aligned with that framework by creating <232> (Elemental Impurities — Limits) and <233> (Elemental Impurities — Procedures), with full implementation expected for new drug products by 2016 and a compliance transition for existing products through 2018.

The test that had served as the industry default for decades was no longer fit for purpose.

What USP <232> Actually Requires

USP <232> sets concentration limits for 24 elemental impurities across three classes, defined by their toxicological profiles and how commonly they appear in manufacturing environments.

Class 1 elements — arsenic (As), cadmium (Cd), mercury (Hg), and lead (Pb) — represent the highest concern. Their oral Permitted Daily Exposures under USP <232>/ICH Q3D are:

• Lead: 5 μg/day
• Cadmium: 5 μg/day
• Arsenic (inorganic): 15 μg/day
• Mercury (inorganic): 30 μg/day

These must be evaluated and controlled in all finished oral products. There’s no carve-out for botanical complexity or “natural” sourcing.

Class 2A elements — cobalt, nickel, and vanadium — have moderate PDEs and are relevant when naturally present in ingredients or introduced through manufacturing equipment. Nickel, for instance, has an oral PDE of 200 μg/day but appears in virtually every botanical ingredient at detectable levels.

Class 2B elements — silver, gold, iridium, osmium, palladium, platinum, rhodium, ruthenium, selenium, and thallium — are primarily relevant when intentionally used in synthesis (e.g., platinum-based catalysts). In the context of most dietary supplements, these rarely drive compliance issues.

Class 3 elements — barium, chromium, copper, lithium, molybdenum, antimony, and tin — carry relatively higher PDEs and lower inherent toxicity at typical exposure levels. They’re measured in a complete panel but rarely trigger failures in botanical or vitamin supplement matrices.

Here’s the practical point most brands miss: the concentration limit in your finished product isn’t a single fixed ppm number. It’s calculated based on your product’s maximum daily dose. If your serving delivers 2 grams per day and the lead PDE is 5 μg/day, your permitted lead concentration is 2.5 μg/g. If a different product delivers 10 grams per day, the permitted concentration is 0.5 μg/g. Same element. Same PDE. Different limit — because compliance is dose-dependent.

That’s a fundamentally different framework from the old USP <231> world, where a single 20 ppm threshold applied regardless of how much product someone consumed daily.

Why ICP-MS Is the Method — and What <233> Specifies

USP <233> defines the validated analytical procedures for measuring elemental impurities. It recognizes two techniques: Inductively Coupled Plasma-Mass Spectrometry (ICP-MS) and Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-OES).

ICP-MS is the workhorse for most supplement testing, and for two reasons that matter enormously in practice: detection limits and element specificity. A well-maintained ICP-MS system achieves detection limits below 0.001 μg/g (1 ppb) in prepared solution — which, after accounting for the dilution that occurs during acid digestion, typically translates to sub-ppb detection in the original sample matrix. That sensitivity is essential when you’re trying to confirm compliance with a 5 μg/day PDE in a low-dose product like a concentrated botanical extract or a vitamin tablet.

ICP-OES is faster and less expensive to run, and for elements with higher PDEs or products with large daily servings, it may be sufficient. But for the Class 1 elements in concentrated or low-dose formulations, ICP-MS is usually the only approach that can demonstrate compliance with adequate confidence.

USP <233> also specifies the validation requirements the method must satisfy: specificity, limit of quantitation, accuracy expressed as percent recovery (within 70–150% for most elements), and inter-day precision. A COA from an ISO 17025-accredited laboratory with a validated ICP-MS method referenced to USP <233> provides documentation that will hold up under FDA review, retailer audits, and Amazon’s third-party verification requirements. A COA that reads “Heavy Metals by USP <231>: Passes 20 ppm as Pb” does not.

The Compliance Gap Specific to Supplements

Here’s where the nuance matters, and where we see the most confusion from brands: USP <232> and <233> are mandatory for pharmaceutical drug products but technically “informational” for dietary supplements under current FDA guidance. The agency hasn’t issued a specific rule mandating the ICH Q3D framework for supplements the way it has for OTC and prescription drug products.

So why does this matter for your supplement? Three reasons that are anything but theoretical.

21 CFR Part 111 — the GMP regulation for dietary supplements — requires that finished products meet specifications for identity, purity, strength, and composition. “Purity” is intentionally broad. FDA investigators have cited elemental contamination as a purity failure under this authority, and the agency doesn’t need a specific elemental impurities rule to act on GMP violations. The regulatory exposure is real.

Retail and e-commerce requirements have moved faster than rulemaking. Amazon, major natural retailers, and club stores now routinely require test reports from accredited third-party laboratories, and those reports are expected to include specific, quantitative elemental data. A one-line colorimetric pass doesn’t satisfy those requirements, and supplement brands have lost listings over the gap.

Botanical ingredients are high-risk matrices. Turmeric, ashwagandha, spirulina, rice protein, and similar ingredients concentrate heavy metals from agricultural soils, irrigation water, and processing equipment in ways that are highly variable and often invisible to colorimetric screening. At Qalitex, we routinely see arsenic in rice-based protein concentrates at concentrations ranging from 0.3 to 1.8 μg/g. For a protein powder with a 30-gram daily serving, the upper end of that range translates to 54 μg of arsenic per day — more than three times the inorganic arsenic PDE. A USP <231> screen at 20 ppm would pass that batch comfortably. A dose-corrected USP <232> assessment shows a clear exceedance.

That’s not a theoretical risk. That’s a batch that should not have shipped.

What a Compliant COA Should Actually Show

When you’re reviewing an elemental impurities COA — whether from a raw material supplier or a finished-product third-party lab — here’s what a compliant report looks like and what to watch for.

Individual elements should be listed by name with specific numeric results. Not “Heavy Metals: Pass.” You need to see “Lead: 0.04 μg/g,” “Arsenic: 0.11 μg/g,” and so on for each element in the panel — at minimum the four Class 1 elements, and ideally the full 24-element suite.

The analytical method should be identified — ICP-MS or ICP-OES — along with the applicable standard (USP <233>, AOAC 2013.06, or equivalent). The report should also state the method’s LOQ for each element, so you can assess whether the technique was sensitive enough to detect a failure at your dose level.

The laboratory must carry ISO 17025 accreditation with scope covering elemental analysis in the relevant sample matrices (botanical extracts, finished dietary supplements, etc.). This is the only accreditation standard that involves a formal third-party assessment of method competency and measurement validity — not just a self-declaration.

And if you’re making a USP <232> compliance determination, you still have work to do after the COA arrives. The compliance calculation — taking the measured concentration, multiplying by your maximum daily serving, and comparing to the applicable PDE — is your responsibility as the brand owner. The COA gives you the measurement; the risk assessment is yours.

If your current supplier is still handing you a USP <231> colorimetric screen and calling it a complete heavy metals test, that’s a conversation worth initiating before the next batch ships. The framework changed. The regulatory and commercial expectations have caught up. The question now is whether your testing program has.