Total vs. Inorganic Arsenic in Botanical Ingredients: Why ICP-MS Speciation Changes the Compliance Picture

Three weeks ago, a Chicago-area supplement brand nearly rejected a 200 kg spirulina shipment. Their third-party CoA showed 6.2 ppm total arsenic — well above the 1.5 ppm internal limit their quality team had written into supplier specifications. They called us before pulling the trigger on the rejection.

The speciation panel we ran told a different story. Of that 6.2 ppm total, 5.8 ppm was arsenobetaine — an organic arsenic compound so biologically inert that the European Food Safety Authority has assigned it a hazard ranking that is, for all practical purposes, zero. Inorganic arsenic, the form that causes cancer, peripheral neuropathy, and cardiovascular disease, came in at 0.4 ppm. That’s comfortably within USP <232> limits by any rational calculation.

They accepted the shipment. Then they rewrote their incoming inspection specification.

This scenario plays out more often than it should. Total arsenic figures on supplier CoAs are technically accurate but regularly misinterpreted by buyers who don’t know what they’re looking at. If your raw material specs treat all arsenic as equally dangerous, you’re making procurement and compliance decisions on the wrong number — and probably blocking safe ingredients while potentially missing the ones that actually warrant concern.

Why “Arsenic” Is Not a Single Substance

Chemists have understood this for decades, but the supplement industry hasn’t fully caught up. Arsenic exists in multiple chemical forms with radically different toxicological profiles. The two categories that matter for raw material testing are inorganic and organic arsenic.

Inorganic arsenic — arsenite (As³⁺) and arsenate (As⁵⁺) — is the form that IARC classifies as a Group 1 human carcinogen. Chronic exposure at 50 µg/day or higher drives dose-dependent increases in skin, lung, and bladder cancer risk. It accumulates in groundwater, contaminates rice paddies flooded with arsenic-bearing irrigation water, and is the subject of virtually every toxicological guideline you’ll encounter.

Organic arsenic compounds are chemically and biologically distinct. Arsenobetaine and arsenocholine — the dominant arsenic species in marine organisms — are absorbed rapidly and excreted intact without metabolic conversion to inorganic forms. EFSA’s comprehensive 2009 assessment concluded arsenobetaine poses “no nutritional or toxicological concern” at realistic dietary intakes. Dimethylarsinic acid (DMA) and monomethylarsonic acid (MMA) sit in a middle tier: they’re metabolic intermediates produced during inorganic arsenic biotransformation in mammals, mildly toxic at high doses, but orders of magnitude less hazardous than As(III) or As(V).

Marine-derived botanicals bioaccumulate arsenic from seawater with impressive efficiency. Spirulina, chlorella, kelp, bladderwrack, dulse, and hijiki routinely show total arsenic concentrations of 3 to 50 ppm in commercial ingredient lots. In most well-characterized marine sources, 80 to 95% of that total is arsenobetaine or arsenocholine. The total arsenic number is high. The inorganic arsenic exposure it actually represents is often negligible.

Contrast that with rice. Rice paddies in arsenic-affected regions — parts of South Asia, the Sacramento Valley in California — accumulate arsenate through root uptake from soil and water. Unlike marine organisms, rice doesn’t preferentially store organic forms. Rice protein concentrates and rice bran extracts can carry inorganic arsenic at 0.1 to 0.5 ppm or higher, and that inorganic fraction is genuinely worth scrutinizing against serving-size exposure calculations.

What USP <232> Requires — And Where the Standard Falls Short

USP <232> (Elemental Impurities — Limits) establishes a permitted daily exposure (PDE) for arsenic of 15 µg/day for oral products. That PDE traces back to ICH Q3D, the international pharmaceutical standard that USP <232> harmonizes with. The math is straightforward: multiply the concentration in your ingredient by the daily intake of that ingredient, and compare against 15 µg/day.

But here’s the gap that routinely catches brands off guard. USP <232> specifies total elemental arsenic, not inorganic arsenic. The chapter was developed for pharmaceutical drug products, where the arsenic encountered in excipients, reagents, and water is overwhelmingly inorganic. The standard’s authors weren’t thinking about marine botanicals with ppm-level arsenobetaine. When dietary supplement companies adopted USP <232> as their reference standard — either voluntarily or under FDA’s 21 CFR Part 111 cGMP framework — they imported an assumption that doesn’t hold for this entire class of ingredients.

ICH Q3D does permit a route out. Section 2.1 of ICH Q3D states explicitly that the PDEs are based on total elemental exposure, but “when the chemical form of the element is known,” sponsors may use the toxicological data for that specific form in a risk assessment. In practice, this means a brand with validated IC-ICP-MS speciation data demonstrating that inorganic arsenic comprises less than 10% of total arsenic in their marine botanical can build a documented, defensible risk assessment showing actual inorganic arsenic exposure is within limits — even when total arsenic isn’t.

California adds a second compliance layer that makes speciation even more important. Prop 65 lists inorganic arsenic compounds as carcinogens with a No Significant Risk Level (NSRL) of 10 µg/day — stricter than the USP PDE. For brands distributing nationally into California, that’s the binding constraint on inorganic arsenic. And if your purchasing team is only looking at total arsenic, you either can’t demonstrate compliance with Prop 65 or you’re unknowingly rejecting ingredients that would sail through a proper speciation-based analysis.

How a Chemical Testing Laboratory Runs Arsenic Speciation

Total arsenic measurement is conceptually simple. Digest the sample in nitric acid, introduce the solution to the ICP-MS plasma, measure the signal at arsenic’s primary mass-to-charge ratio (m/z 75), and quantify against calibration standards. The plasma reduces every arsenic-containing molecule to individual atoms regardless of its chemical form. Fast, reproducible, and completely blind to speciation.

Speciation requires a fundamentally different approach. The standard analytical testing laboratory method is hyphenated IC-ICP-MS — ion chromatography coupled in-line with ICP-MS. Before the dissolved sample enters the plasma, it passes through an ion-exchange chromatographic column. Each arsenic species — As(III), As(V), DMA, MMA, arsenobetaine — carries a different charge density and elutes from the column at a distinct retention time. The ICP-MS detector quantifies each eluting peak separately, producing individual concentration values for every arsenic form in the sample.

Method validation for arsenic speciation typically references certified reference materials such as NIST SRM 1566b (oyster tissue) or TORT-3 (lobster hepatopancreas), which carry certified values for multiple individual arsenic species. A lab running this method without species-specific reference materials is running an incomplete validation — worth asking about before you accept a speciation CoA.

One critical sample preparation note. Strong acid digestion — the standard prep for total ICP-MS metals panels — destroys organic arsenic compounds through decomposition and chemical conversion. Running speciation on a conventionally digested extract defeats the purpose entirely. Proper speciation requires a mild enzymatic or aqueous extraction that preserves native chemical forms. If a lab proposes to run speciation from the same digest as your full metals panel, push back and ask for their extraction protocol. The answer will tell you whether they actually understand what they’re doing.

Which Ingredients Actually Warrant Speciation Testing

Not every incoming material needs the added complexity and cost. A risk-based triage approach keeps testing budgets rational while addressing the materials where speciation actually changes decisions.

Run speciation as a default on all marine-derived botanicals: spirulina, chlorella, kelp, bladderwrack, dulse, hijiki, marine collagen, carrageenan, and fish-derived concentrates where total arsenic exceeds 1 ppm. Speciation almost certainly reveals that inorganic arsenic is a small fraction of the total, protecting your brand from unnecessary rejections and equipping you with the documentation to defend that decision if challenged.

Run speciation to confirm risk — not to exonerate — for rice protein concentrates, rice bran extract, and rice flour. These ingredients carry inorganic arsenic at meaningful fractions of the total. A total arsenic figure of 0.3 ppm in rice protein sounds low, but if 70% is inorganic, you’re at 0.21 ppm inorganic — worth calculating against serving size before moving on.

Lower speciation priority for most terrestrial botanicals — ashwagandha, turmeric, valerian, saw palmetto, ginkgo. Total arsenic in these materials typically runs below 0.3 ppm, and even if the inorganic fraction were 100%, the resulting daily exposures at normal dosing would be modest. That said, if a lot comes in with a total arsenic figure above 0.5 ppm in a non-marine botanical, speciation is worth the additional spend before you accept or reject.

When our team receives samples at the Countryside facility for incoming inspection, marine and rice-derived ingredients now trigger automatic speciation add-ons before the samples ship to the analytical labs for processing. Getting the right test ordered at intake eliminates the expensive scenario where a total metals panel comes back flagged, inventory goes on hold, and you’re rushing a speciation order to rescue a procurement decision that needed the right data two weeks earlier.

What to Ask Your Lab, and How to Read the Report

Request IC-ICP-MS arsenic speciation with quantitation of at minimum: inorganic As(III), inorganic As(V), DMA, MMA, and arsenobetaine. Any accredited analytical testing laboratory running this method should be able to report all five fractions with method detection limits below 0.05 ppm.

When the speciation report arrives, the compliance-relevant number is the sum of As(III) + As(V) + MMA + DMA. Arsenobetaine comes out of that calculation entirely. Multiply the summed inorganic-and-metabolite concentration by your daily intake of the ingredient in grams, and compare the resulting µg/day figure against 15 µg/day (USP <232> PDE) and 10 µg/day (Prop 65 NSRL for California).

Document the risk assessment in your batch record. Note the speciation result, the calculation, and the conclusion. If you’re operating under 21 CFR Part 111, this documentation belongs in your component qualification records alongside the CoA. The ICH Q3D framework explicitly supports species-specific risk assessments when analytical data backs them up — your records need to demonstrate that the data exists and that someone with appropriate training evaluated it.

Finally, consider pushing your speciation requirement upstream. For marine botanicals, it’s reasonable to require that supplier-provided CoAs include speciation data, not just total arsenic. Established spirulina and chlorella producers — particularly those supplying pharmaceutical-grade material — can typically provide this. If a supplier can’t or won’t, that’s a supplier qualification data point that belongs in your vendor file regardless of what else their documentation looks like.

---

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

• Full ICP-MS Elemental Impurity Panels Under ISO 17025 Accreditation — Including arsenic speciation by IC-ICP-MS for marine-derived botanicals and rice-based ingredients.