Icp-Ms Testing for Supplements, Explained

This is a plain-English guide to ICP-MS, the analytical method behind essentially every credible heavy metal result in the supplement industry. It explains how the instrument works, why it replaced the old colorimetric test, which official methods anchor it, what detection and quantitation limits really mean, and how to read the lab reports it produces. It is written for quality leads and founders who need to understand the science well enough to buy it, question it, and stand behind it.

You do not need to operate a mass spectrometer to run a supplement brand. But if heavy metal results appear anywhere in your quality file or your marketing, you are effectively vouching for numbers that came out of one. Knowing how those numbers are made, and where they can mislead, is the difference between a brand that can answer a buyer's technical diligence and one that forwards the question to a lab and hopes.

What is Icp-Ms and how does it work?

ICP-MS stands for inductively coupled plasma mass spectrometry. Strip away the jargon and it is a two-stage machine: one stage tears the sample apart into charged atoms, the other counts those atoms by weight.

Stage one: the plasma. The dissolved sample is sprayed as a fine mist into a stream of argon gas flowing through a torch. A radio-frequency coil energizes the argon into a plasma, a state of matter so hot (thousands of degrees Celsius, hotter than the surface of the sun) that molecules cannot survive. Everything in the sample is atomized, and the atoms are ionized, meaning they are stripped of an electron and acquire an electric charge. This matters because charged particles can be steered with electric and magnetic fields.

Stage two: the mass analyzer. The ions are pulled out of the plasma into a vacuum chamber and passed through a mass filter, most commonly a quadrupole, which acts like a tunable gate: at any instant it lets through only ions of one specific mass-to-charge ratio. Lead 208 gets through when the gate is tuned for lead 208; everything else is deflected away. A detector counts the ions that pass, and the count is compared against calibration standards of known concentration to produce a number: so many micrograms of lead per kilogram of sample.

Because every element (and even each isotope) has a distinct mass, ICP-MS can measure dozens of elements in a single run, at extraordinarily low levels: detection capability reaches roughly parts per trillion for many elements. For scale, one part per trillion is about one second in 32,000 years. This sensitivity is why "non-detect" claims need careful reading, which we will get to shortly.

Why did Icp-Ms replace the old heavy metals test?

For most of the twentieth century, pharmacopeial heavy metals testing meant USP General Chapter <231>, a colorimetric method dating to the era of wet chemistry. The analyst precipitated metals as sulfides and compared the color of the resulting solution against a lead standard by eye. It had three fundamental problems: it could not tell you which metals were present (everything was expressed as an undifferentiated "heavy metals" number), it could not reliably measure the low levels modern limits require, and the result depended partly on the analyst's eyesight.

USP retired it. Effective January 1, 2018, Chapter <231> was omitted from the USP-NF and its references deleted, replaced by the modern elemental impurities framework: USP <232> and <233> for drug products, with <2232> providing the corresponding limits for dietary supplements. The replacement methods are instrumental, element-specific, and quantitative: they tell you exactly which element is present and at exactly what concentration.

The practical takeaway for a brand: if a supplier's documentation still references "USP <231>" or an unspecified "heavy metals" pass, the testing behind it is obsolete by nearly a decade. That is a supplier-qualification red flag, covered further in our guide to supplier certificate of analysis verification.

Which official methods anchor supplement testing?

Two references do most of the anchoring work in the US:

• USP <233>, Elemental Impurities: Procedures defines the compendial analytical procedures, built around ICP-OES and ICP-MS, including the validation requirements a lab must meet (accuracy, precision, specificity) to claim compliance. USP <2232> points dietary supplement testing at these procedures.
• FDA Elemental Analysis Manual, EAM 4.7 is FDA's own validated ICP-MS method for foods and related products, using microwave-assisted digestion, and it has been validated across multiple laboratories. When FDA tests a supplement, this is the family of methods it uses.

A quick word on ICP-OES, the related technique you will see on quotes: it uses the same plasma but measures light emitted by excited atoms rather than counting ions by mass. It is robust and cheaper but typically less sensitive. For nutrient minerals at percent levels it is fine; for lead at fractions of a ppm against benchmarks like the Prop 65 MADL, ICP-MS is the appropriate tool. When your specification lives in µg/day territory, as described in our guide to heavy metal limits in dietary supplements, insist on ICP-MS.

How does a powder become a number? (Sample preparation)

An ICP-MS instrument drinks liquids, so solid samples must first be dissolved completely. The standard approach, specified in EAM 4.7, is microwave-assisted digestion: a weighed portion of the sample is sealed in a pressure vessel with concentrated nitric acid and heated by microwave until the organic matrix is destroyed and the metals are held in clear acidic solution.

Why should a business reader care about this plumbing? Because incomplete digestion is a silent source of falsely low results. If particles survive digestion, the metals inside them never reach the plasma and are never counted. Difficult matrices (mineral-rich formulas, some botanicals) need validated digestion programs, which is one reason supplement-matrix experience matters when choosing a laboratory, a topic we cover in how to evaluate a heavy metal testing laboratory.

What is the difference between a detection limit and a quantitation limit?

This distinction is where most non-technical readers get misled, so here it is in plain terms:

The critical insight: "non-detect" is a statement about the reporting limit, not about the product. "ND" at a reporting limit of 500 ppb and "ND" at a reporting limit of 5 ppb are wildly different statements. The first is compatible with lead levels that could fail a strict specification; the second is genuinely informative. Two labs can test the same lot and one reports ND while the other reports 40 ppb, with both being correct, simply because their reporting limits differ.

This is why a COA that shows "ND" without stating the reporting limit for each element is close to uninterpretable, and why comparing products by their "non-detect" claims without comparing reporting limits is meaningless. It is also why "zero heavy metals" is not a claim any laboratory can substantiate: an instrument can only ever say "below our limit," and at parts-per-trillion sensitivity, truly agriculture-derived products routinely show detectable traces. Detection is not risk; the dose per daily serving against a benchmark is what matters.

When you set specifications, make the lab's quantitation capability part of the conversation: your reporting limits need to sit comfortably below your specification, or the data cannot demonstrate conformance.

What is speciation, and why does total vs inorganic arsenic matter?

Standard ICP-MS counts atoms by mass, so it reports total arsenic: every arsenic atom, regardless of the molecule it arrived in. But arsenic chemistry matters enormously. Inorganic arsenic species are the toxicologically significant forms, while organic species such as arsenobetaine (common in fish and some marine ingredients) are far less concerning. The same logic applies to mercury, where methylmercury is treated separately from total mercury.

The major benchmarks reflect this: USP <2232> sets its arsenic limit for inorganic arsenic (15 µg/day) and sets both a total mercury limit (15 µg/day) and a stricter methylmercury limit (2 µg/day) (USP <2232>). California's oral NSRL for arsenic likewise addresses inorganic arsenic (OEHHA).

Measuring species requires speciation analysis: a chromatography step that separates the arsenic-containing molecules before the ICP-MS counts them, so each species gets its own number. It costs more and fewer labs offer it. The practical decision rule: for most simple formulations (creatine, vitamins, electrolytes), total arsenic results are usually so low that speciation is unnecessary; if total arsenic is comfortably below the inorganic limit, the inorganic fraction must be too. Speciation earns its cost when an ingredient has meaningful total arsenic and you need to know whether the benign or the significant form is responsible, a situation more common with marine-derived and rice-derived ingredients.

What should you look for on an Icp-Ms lab report?

A credible report should let a stranger reconstruct what was done. Check for:

• Method reference: ICP-MS per an identified method (EAM 4.7, USP <233>-compliant, or an equivalent validated in-house method)
• Sample identity: product name, lot number, sample mass, and date received
• Results with units: for each element, a number with explicit units (ppb, µg/g, mg/kg), never a bare "pass"
• Reporting limit per element: stated alongside every result, especially every ND
• Preparation: digestion method noted
• Accreditation: the lab's ISO/IEC 17025 accreditation and, ideally, whether this method is on its accredited scope
• Species clarity: whether arsenic and mercury results are total or speciated

If you can check all seven, you have a report that supports specifications, buyer diligence, and marketing substantiation. Interpreting the numbers against your serving size is the next step, covered in how to read a heavy metal certificate of analysis.

One more distinction worth keeping sharp: an excellent ICP-MS report is still just a snapshot of one lot from one day. It becomes market-facing trust only when an independent program verifies results against a published standard on an ongoing basis, which is the role of heavy metal certification for supplements. Heavy Metal Certified coordinates its testing through qualified independent laboratories using validated elemental methods, and evaluates results against a transparent published standard.

Faq: Icp-Ms Testing for Supplements

What does ICP-MS stand for? Inductively coupled plasma mass spectrometry. An argon plasma ionizes the sample's atoms, and a mass spectrometer separates and counts those ions by mass-to-charge ratio, producing element-by-element concentrations.

How sensitive is ICP-MS? Detection capability reaches roughly parts per trillion for many elements, which is thousands of times below the levels at which any regulatory or pharmacopeial benchmark operates. This is why detectable traces are normal and why dose-based interpretation matters.

Why was USP <231> retired? The old colorimetric method could not identify individual elements, lacked the sensitivity of modern limits, and depended on visual comparison. Effective January 1, 2018, USP omitted <231> in favor of instrumental methods defined in <232> and <233>, with <2232> covering dietary supplements.

Does "non-detect" mean there are no heavy metals in the product? No. ND means the level was below the lab's reporting limit. The claim is only as strong as that limit, which is why a COA should state the reporting limit for every element. "Zero heavy metals" is not a substantiable claim.

What is the difference between ICP-MS and ICP-OES? Both use an argon plasma. ICP-OES measures light emitted by excited atoms; ICP-MS counts ions by mass and is typically far more sensitive. For contaminant work against µg/day benchmarks, ICP-MS is the standard choice.

What is arsenic speciation and do I need it? Speciation separates arsenic compounds before measurement, distinguishing toxicologically significant inorganic arsenic from less concerning organic forms. Most simple formulations do not need it: if total arsenic is well below the inorganic benchmark, speciation adds little. It matters when total arsenic is elevated and the species composition determines the outcome.

What method should my lab be using? Ask for ICP-MS following FDA EAM 4.7, a USP <233>-compliant procedure, or an equivalent validated method, with microwave-assisted digestion and stated reporting limits per element.

If you would rather have this rigor verified than just described, request a preliminary certification assessment: we review your current lab reports, methods, and reporting limits against the published standard. Learn about heavy metal testing and certification or apply for certification.