ICP-MS vs. ICP-OES: Which Heavy Metal Testing Method Does Your Product Actually Need?

The question comes up in nearly every intake conversation we have with a new client: “Does it matter which instrument you use?” For heavy metal testing, the answer is yes — and choosing the wrong method can leave you either over-spending on analysis or, more dangerously, missing contamination that a less sensitive technique simply can’t see.

Both ICP-MS (Inductively Coupled Plasma–Mass Spectrometry) and ICP-OES (Inductively Coupled Plasma–Optical Emission Spectrometry, sometimes called ICP-AES) are workhorse instruments in any serious analytical chemistry lab. They share a common front end — a plasma torch that atomizes and ionizes your sample at temperatures exceeding 6,000 K — but diverge sharply in how they detect and quantify elements afterward. That divergence matters enormously when the compliance thresholds you’re working against are measured in parts per trillion.

What the Detection Limits Actually Mean

ICP-OES works by measuring the light emitted when excited atoms relax back to their ground state. Each element emits at characteristic wavelengths, and the instrument can monitor hundreds of those emission lines simultaneously. It’s fast, robust, and well-suited for elements present in the parts-per-billion (μg/L) range. Detection limits typically fall between 1 and 50 μg/L depending on the element and the sample matrix — plenty of sensitivity for nutritional mineral panels, environmental compliance at EPA action levels, or many industrial quality control applications.

ICP-MS takes a fundamentally different approach. After the plasma ionizes the sample, ions are extracted through an interface and separated by mass-to-charge ratio in a quadrupole or sector-field mass analyzer. Because the detector counts individual ions rather than measuring photon intensity, sensitivity is orders of magnitude higher. Most elements reach detection limits in the 0.001–0.1 μg/L range. That’s parts per trillion, not parts per billion — a difference of roughly 10,000-fold for many analytes.

And ICP-MS can screen more than 70 elements in a single analytical run without switching methods or instrument configurations. For a supplement brand trying to build a comprehensive elemental profile, that efficiency matters.

The gap between 50 μg/L and 0.001 μg/L might seem abstract until you run the numbers against an actual regulatory threshold.

Why California Prop 65 Limits Drive Most Brands Toward ICP-MS

California’s Proposition 65 maximum allowable dose level (MADL) for lead is 0.5 micrograms per day. For a dietary supplement with a typical 5-gram serving size, that works out to a product concentration limit of roughly 0.1 mg/kg — or 0.1 ppm. ICP-OES can theoretically reach that concentration in a clean, simple matrix. But “theoretically” is doing a lot of work in that sentence.

Digest a real-world supplement matrix — botanical powders, protein concentrates, multi-ingredient blends — and background noise from co-eluting elements, matrix-induced spectral overlap, and signal suppression will reliably push your practical quantitation limit well above the regulatory threshold. You end up with a result that says “below detection” when, in fact, the instrument just can’t see that low in that matrix.

ICP-MS reaches 0.1 ppm with substantial margin to spare. And because it simultaneously quantifies arsenic, cadmium, chromium, mercury, and dozens of other elements in the same run, you get a defensible, comprehensive picture rather than a narrow screen.

The same arithmetic applies to USP <232>, which governs elemental impurities in pharmaceutical products and dietary supplements sold in the US. Permitted daily exposures (PDEs) via the oral route are set at 5 μg/day for lead, 15 μg/day for arsenic, 5 μg/day for cadmium, and 15 μg/day for mercury (class 2 oral elements). These limits sit comfortably within ICP-MS sensitivity but can genuinely challenge ICP-OES — particularly in complex botanical or protein-rich matrices where matrix suppression is the norm rather than the exception.

California’s Prop 65 enforcement has also grown more sophisticated. Plaintiff law firms and citizen enforcers have become adept at identifying products with detectable-but-undisclosed heavy metals, and a Certificate of Analysis showing “not detected” at an ICP-OES limit of 50 μg/L is not legally equivalent to “not detected” at an ICP-MS limit of 0.05 μg/L. Sophisticated retail buyers, Amazon compliance teams, and third-party audit programs increasingly understand this distinction and will scrutinize the reported detection limits on your CoA, not just the pass/fail result.

Where ICP-OES Still Earns Its Place

None of this means ICP-OES is obsolete — far from it. Several applications genuinely favor its characteristics.

Nutritional mineral panels are the clearest case. If you’re quantifying calcium at 500 mg per serving, magnesium at 250 mg, or zinc at 15 mg, ICP-MS is overkill. Those analyte concentrations are in the percent-to-thousands-of-ppm range, and the instrument has to dilute samples heavily to avoid flooding the detector. ICP-OES handles high-concentration analytes with better linearity, wider dynamic range for macro-elements, and without the aggressive dilution requirements that ICP-MS demands for concentrated matrices.

ICP-OES is also better suited for industrial and agricultural testing — soil analysis, fertilizer characterization, wastewater screening — where regulatory thresholds are set in the ppm range and sample throughput matters more than sub-ppb sensitivity. Per-sample cost is generally lower, and the instrument is more tolerant of high total dissolved solids, which is a practical advantage when running large batches.

Some food safety laboratories use both instruments in a tiered approach: ICP-OES for rapid, cost-effective macro-element and nutritional screening, then ICP-MS for confirmation analysis or trace-element work when the product type or regulatory context requires sub-ppb detection. That tiered model is worth discussing with your lab before you submit samples — it can keep per-sample costs reasonable without sacrificing the sensitivity you actually need for compliance.

Matching the Method to Your Regulatory Situation

Here’s a working framework for deciding which technique fits your product:

Use ICP-MS when:

• Your product is sold in California and subject to Prop 65 reporting obligations for lead, arsenic, cadmium, or mercury
• You’re seeking USP <232>/<233> compliance for a pharmaceutical, supplement, or OTC product
• Your matrix is complex — botanical powders, protein concentrates, multi-ingredient formulas — where trace-level accuracy and low method detection limits are non-negotiable
• You need arsenic speciation (i.e., distinguishing organic arsenic from far more toxic inorganic arsenic), which requires HPLC-ICP-MS hyphenated techniques
• Your retailer, Amazon listing, or third-party certification program explicitly requires it

Use ICP-OES when:

• You’re running a nutritional labeling panel where target elements are present at ppm concentrations or above
• Applicable compliance limits are set firmly in the ppm range and your matrix is relatively clean
• You need high throughput on a defined, limited element list and per-sample budget is the primary driver
• You’re doing initial screening on raw materials before deciding whether trace-level confirmation testing is warranted

For most dietary supplement and personal care brands selling through US retail or e-commerce in 2026 — especially if California is any part of your distribution — ICP-MS will be the appropriate choice. The regulatory and litigation environment around undisclosed heavy metals has tightened considerably, and a defensible CoA requires detection limits that match the thresholds you’re certifying against.

What to Ask Your Lab Before You Submit Samples

Before sending samples to any analytical laboratory for heavy metal testing, three questions are worth raising explicitly:

What are your reported detection limits for lead, arsenic, cadmium, and mercury in my matrix? Any laboratory doing compliance-grade work should be able to provide method detection limits (MDLs) or practical quantitation limits (PQLs) in writing, specific to your sample type. If the lead MDL comes back in the tens of ppb, ask whether that’s sufficient for your specific compliance threshold — and get the answer documented.

Is the method validated for my matrix? Running a botanical extract through a method validated for drinking water is not the same thing. Matrix-matched calibration, spike recovery data, and documented interference checks are what separate a regulatory-defensible result from a number on a page. Ask to see the validation summary or the scope of methods the lab has qualified for your product category.

Is the laboratory ISO 17025 accredited for this specific method and matrix? Accreditation means an independent assessor — typically A2LA or Perry Johnson Laboratory Accreditation in the US — has reviewed the lab’s technical competence, measurement traceability, and quality management system. The scope of accreditation matters; verify that the specific test method (e.g., EPA 200.8, USP <233>, or an in-house validated method) and the relevant matrix are explicitly listed on the lab’s accreditation certificate.

The instrument is only part of the story. A well-run ISO 17025 accredited food safety laboratory with ICP-OES will produce more reliable data than a poorly managed facility with ICP-MS. But when your regulatory obligations require sub-ppb detection — and for most US supplement and cosmetic brands in California’s market, they do — no amount of operational discipline compensates for a method that physically cannot see what you need it to see.

---

Written by Nour Abochama, Vice President of Operations, Qalitex Laboratories. Learn more about our team

Talk to our team about your testing needs. Contact us

Related from our network

• FDA Audit Readiness and Analytical Data Integrity — Aurora TIC covers the regulatory framework behind lab data integrity, 21 CFR Part 11, and what FDA inspectors look for in your analytical records.
• Raw Material Heavy Metal Testing and Supplier COA Verification — Ayah Labs explains how robust supplier qualification programs use ICP-MS data to verify incoming material specifications against USP and Ph.Eur. monograph limits.
• Health Canada Elemental Impurity Requirements for NHPs — Androxa breaks down how Canadian natural health product regulations handle heavy metal limits differently from FDA — important context if you sell across both markets.