When the Spec Sheet Is Not Enough

A procurement director recently asked us to review a heavy metal CoA from a mineral ingredient supplier. The document listed results for lead, arsenic, cadmium, and mercury — all passing. The method listed was “ICP.” No further detail. No detection limits. No sample preparation description.

That level of documentation is insufficient for a defensible quality program. “ICP” is not a method — it is a family of techniques. The difference between ICP-OES and ICP-MS in terms of detection capability spans two to three orders of magnitude for some elements. A result that “passes” on one instrument may not be reportable at all on another.

This post explains the practical differences between the two dominant heavy metal testing platforms — ICP-MS and AAS — and gives quality managers a framework for deciding which method is appropriate for a given raw material and regulatory context.

The Regulatory Baseline: USP <232> and <233>

For dietary supplement manufacturers selling in the US market, the primary regulatory framework for elemental impurities is USP <232> (Elemental Impurities — Limits) and USP <233> (Elemental Impurities — Procedures). These chapters establish permitted daily exposure (PDE) limits for 24 elements across oral, parenteral, and inhalation routes, and specify the validation requirements for analytical procedures used to demonstrate compliance.

USP <233> is method-agnostic — it does not mandate ICP-MS or AAS. It requires that whatever method you use be validated for specificity, accuracy, precision, and detection/quantitation limits appropriate to the PDEs in <232>. In practice, most laboratories use ICP-MS for <232>/<233> compliance work because the sensitivity requirements for several elements (notably arsenic, cadmium, and mercury at oral PDEs) are achievable with ICP-MS but challenging with flame AAS.

For cosmetic manufacturers, the relevant framework varies by market. The EU Cosmetics Regulation and ISO 17025-accredited testing programs typically reference ICP-MS or ICP-OES for trace element analysis, with detection limits specified per element.

ICP-MS: High Sensitivity, Multi-Element, Higher Cost

Inductively coupled plasma mass spectrometry (ICP-MS) ionizes the sample in an argon plasma and separates ions by mass-to-charge ratio. The technique is capable of sub-ppb (parts per billion) detection limits for most elements of regulatory concern.

Strengths:

Detection limits typically in the 0.001–0.1 µg/L range in solution, depending on the element and matrix
True multi-element capability — a single run can quantify 20+ elements simultaneously
Isotope-specific detection reduces interference from isobaric overlaps (with appropriate correction)
Required or strongly preferred for USP <232>/<233> compliance at oral PDE limits

Limitations:

Higher instrument cost and more complex maintenance than AAS
Matrix effects from high-TDS (total dissolved solids) samples require dilution or matrix-matched calibration
Polyatomic interferences (e.g., ArCl on ⁷⁵As) must be addressed through collision/reaction cell technology or mathematical correction
Results vary by matrix — a method validated for botanical extracts may not perform identically on mineral salts without re-evaluation

In our laboratory work, ICP-MS is the default method for any raw material where USP <232> compliance documentation is required, or where the material is destined for a product with a low daily serving size (which compresses the allowable elemental burden per gram of ingredient).

AAS: Targeted, Cost-Effective, Appropriate for Specific Applications

Atomic absorption spectrometry (AAS) measures the absorption of light at element-specific wavelengths. It comes in two primary configurations relevant to supplement and cosmetic testing:

Flame AAS (FAAS): Suitable for elements present at higher concentrations (typically >0.1 ppm in solution). Commonly used for calcium, magnesium, zinc, iron, and copper in nutritional applications. Detection limits are generally in the low ppm range — insufficient for USP <232> compliance for lead, arsenic, and cadmium at oral PDEs.

Graphite furnace AAS (GFAAS): Significantly more sensitive than flame AAS, with detection limits in the low ppb range for many elements. Can be appropriate for single-element determinations (e.g., lead in a specific matrix) when ICP-MS is not available. However, it is a sequential technique — each element requires a separate run — making it less efficient for multi-element screening.

Hydride generation AAS (HGAAS): Specifically useful for arsenic, selenium, and mercury. Converts these elements to volatile hydrides before atomization, improving sensitivity. AOAC 986.15 uses HGAAS for arsenic in food matrices. Results can be matrix-dependent; consult your method validation data before applying to novel ingredients.

When AAS makes sense:

Targeted single-element testing where ICP-MS is cost-prohibitive and the element of concern is present at concentrations detectable by GFAAS
Nutritional labeling compliance for minerals (calcium, zinc, iron) where ppm-level accuracy is sufficient
Screening applications where a positive result will trigger ICP-MS confirmation

Choosing the Right Method: A Decision Framework

Scenario	Recommended Method
USP <232>/<233> compliance documentation	ICP-MS
Multi-element screening of botanical raw materials	ICP-MS
Nutritional mineral content (Ca, Mg, Zn, Fe)	FAAS or ICP-OES
Single-element confirmation (e.g., lead in a mineral)	GFAAS or ICP-MS
Arsenic speciation (inorganic vs. organic)	HPLC-ICP-MS
California Prop 65 compliance documentation	ICP-MS

Note that arsenic speciation — distinguishing inorganic arsenic (the regulated form) from organic arsenic compounds — requires a hyphenated technique (HPLC-ICP-MS or HPLC-ICP-OES). A total arsenic result alone is not sufficient for rice-based ingredients or marine-derived materials where organic arsenic may constitute a large fraction of the total. Consult your regulatory affairs team on the appropriate specification for your specific ingredient and market.

What to Ask Your Contract Lab

When reviewing a heavy metal CoA or selecting a contract testing partner, ask:

Which specific technique was used (ICP-MS, ICP-OES, FAAS, GFAAS)?
What are the method detection limits (MDLs) and reporting limits for each element?
What sample preparation procedure was used (microwave digestion, dry ashing, acid dissolution)?
Is the method validated per USP <233> or an equivalent standard?
Is the laboratory ISO 17025 accredited for this specific test?

Practical Checklist: Heavy Metal Testing for Raw Materials

Confirm which elements are required by your product specification and applicable regulations (USP <232>, Prop 65, EU limits)
Verify the test method meets the sensitivity requirements for each element at the relevant PDE or limit
Request method detection limits and reporting limits on every CoA — not just results
For rice, marine, or mineral-derived ingredients, request arsenic speciation (inorganic vs. organic)
Confirm the lab’s accreditation scope covers the specific matrix and elements you are testing
Document the method version and instrument type in your batch record for traceability
Investigate any result reported as ”< reporting limit” — confirm the reporting limit is below your specification limit

Heavy Metal Testing for Raw Materials — ICP-MS vs. AAS and When Each Applies