Pharmaceutical Elemental Analysis

USP <233> Elemental Impurity Testing for Pharmaceutical Compliance

Sterling Analytical provides USP <233> elemental impurity testing, the validated ICP-OES and ICP-MS analytical procedures that General Chapter <233> (Elemental Impurities — Procedures) requires pharmaceutical manufacturers to use when demonstrating compliance with the elemental impurity limits established under ICH Q3D and USP <232>. Where our Pharmaceutical Heavy Metals Testing page covers the broader regulatory framework — why elemental impurity testing exists, how the four-class system works, and which compliance pathway applies to your product — this page is specifically about <233> itself: the chapter that defines what a laboratory actually has to do, and prove, for an elemental impurity result to count as compliant.

This distinction matters because <232> and <233> serve genuinely different functions within the same regulatory framework. USP <232> sets the limits — the numbers a result has to meet. USP <233> sets the procedures — the validated methodology that produced the number in the first place, and the documented evidence that the methodology is actually capable of producing a trustworthy result. A laboratory can run an ICP-MS panel and report numbers that happen to be below the <232> limits without ever having demonstrated, in the way <233> requires, that the method generating those numbers is accurate, precise, and specific enough to be believed. <233> is what closes that gap.

Pharmaceutical Heavy Metals Testing

The Harmonized 2026 Update: What Changed and Why It Matters Now

This is worth addressing directly because it’s current and because it affects how existing validation documentation should be evaluated. A harmonized revision of USP <233> became official on May 1, 2026, formally approved through Stage 4 of the Pharmacopeial Discussion Group process — the final stage representing full agreement across the United States Pharmacopeia, the European Pharmacopoeia, the Japanese Pharmacopoeia, and the Indian Pharmacopoeia. This is a meaningful regulatory alignment event: elemental impurity testing methodology is now harmonized across the major global pharmacopeias rather than existing as parallel, potentially divergent regional standards.

One specific, citable change in the harmonized chapter: what was previously titled “Quantitative Procedures” has been renamed “Procedures for Quantitative Tests” — a naming change that reflects a broader integration of ICH Q3D principles more explicitly into the chapter structure than the prior version achieved, shifting how validation requirements are organized around product-specific limits rather than generic compendial defaults.

For laboratories and sponsors with existing validation documentation built against the pre-harmonization chapter, the practical question isn’t whether the regulatory story changed — most coverage of the harmonization has focused on that — it’s whether existing validation work holds up against what the chapter now requires. Programs where validation was structured around a genuine, product-specific ICH Q3D risk assessment and the resulting target limits generally align well with the updated framework, since the underlying validation logic (limit procedure vs. quantitative procedure, accuracy/precision/specificity requirements) carries forward. Programs that validated against generic, non-product-specific limits are the ones that warrant a closer review against the harmonized chapter before assuming continued compliance.

Limit Procedures vs. Quantitative Procedures: The Central Distinction in

This is the organizing concept of the entire chapter, and getting it right at the start of method development determines what validation work is actually required.

USP <233> recognizes two fundamentally different kinds of analytical procedure, and each carries its own distinct validation requirements:

A limit procedure answers a yes/no question: is the elemental impurity present above or below a specified threshold? It does not need to tell you precisely how much of the element is present — only whether the sample passes or fails relative to the limit. Limit procedures require validation of detectability, repeatability, and specificity.

A quantitative procedure (renamed “procedures for quantitative tests” in the 2026 harmonized chapter) answers a how-much question: what is the actual measured concentration of the elemental impurity, expressed as a number? Quantitative procedures require the fuller validation package: accuracy, precision (both repeatability and intermediate precision/ruggedness), and specificity.

The choice between these isn’t arbitrary — it should be driven by what your actual regulatory and quality objective requires. If your control strategy only needs to confirm that an element stays below its PDE-derived limit, a validated limit procedure may be entirely sufficient and is generally less burdensome to validate and run routinely. If you need trending data across batches, investigation-grade precision, or quantitative documentation for a regulatory submission, a quantitative procedure is required. Raising this question — limit test or quantitative test, and does that match the actual regulatory objective — before a validation study begins avoids the much more painful situation of discovering a documentation gap mid-study or, worse, after submission.

What Actually Requires for Limit Procedure Validation

For a limit procedure to be deemed appropriate under USP <233>, validation must establish:

Accuracy — demonstrated using a control sample: a preparation of certified reference materials for the element of interest at the indicated (target) level, alongside a test sample spiked at that same target concentration, referred to in the chapter as “1J” (J representing the target concentration level, typically the relevant PDE-derived limit). The spiked sample’s measured result must be within ±10% of, or greater than, the result obtained from the standard solution prepared at the same level.

Detectability and discrimination — the procedure must be capable of distinguishing a sample spiked at the target concentration (1J) from a sample of the same material spiked at 80% of the target concentration (0.8J). The measured result for the 0.8J spike must be measurably less than the result for the 1J spike. This requirement exists specifically to confirm the method has real discriminating power around the limit, not just the ability to detect that something is present.

Precision — using six independent samples of the material under test, spiked with appropriate reference materials, with an acceptance criterion of relative standard deviation not more than (NMT) 20% for each target element.

Specificity — the procedure must unequivocally assess each target element in the presence of components that may reasonably be expected to be present, including other target elements and matrix components that could interfere — both false-positive interference (showing signal for an element that isn’t actually elevated) and false-negative interference (failing to detect an element that is) must be ruled out.

What <233> Actually Requires for Quantitative Procedure Validation

Quantitative procedures carry a fuller validation burden, reflecting the higher bar of producing a defensible, trustworthy numeric result rather than a pass/fail call:

Accuracy — assessed using three replicate standard reference material (SRM) solutions containing the target analytes at concentrations spanning 50% to 150% of J, alongside three replicate test samples spiked with SRM across that same 50%–150% J range. The acceptance criterion is spike recovery of 70% to 150% at each tested concentration — a real, specific, and relatively generous range that reflects the genuine analytical difficulty of accurately quantifying trace elements across a complex pharmaceutical matrix.

Precision — six samples spiked with target analytes at the J concentration, with an acceptance criterion of standard deviation not more than 20%.

Intermediate precision (ruggedness) — a repeatability test performed under at least one varied condition: a different day, a different instrument, or a different analyst, with an acceptance criterion of percent relative standard deviation not more than 25% for each target element. This requirement exists because a method that performs beautifully in the hands of one analyst on one instrument on one day isn’t actually validated for routine use — ruggedness testing confirms the method holds up under the realistic variation that occurs in ongoing laboratory operation.

Specificity — the same false-positive/false-negative interference check required of limit procedures, confirmed in the presence of other elements at their indicated limits.

The Two Referee Procedures: 's Built-In Compendial Methods

USP <233> doesn’t just set validation requirements for elemental impurity testing generally — it provides two specific, named compendial procedures, referred to as Referee Procedures, that laboratories can use directly without having to validate a method from scratch:

Referee Procedure 1 — applicable to elemental impurities generally amenable to detection by ICP-OES (Inductively Coupled Plasma — Optical Emission Spectrometry).

Referee Procedure 2 — applicable to elemental impurities generally amenable to detection by ICP-MS (Inductively Coupled Plasma — Mass Spectrometry).

Both referee procedures specify closed-vessel microwave digestion for sample preparation — a meaningful technical choice worth understanding in context: older, simpler sample preparation approaches involving open-vessel ashing can produce low recoveries for volatile elements, since volatile elemental species are lost to the atmosphere during high-temperature, open-system ashing rather than being captured and retained for measurement. Closed-vessel microwave digestion avoids this loss pathway, which is part of why it’s specified as the referee approach rather than left as an option among several roughly equivalent alternatives.

An important distinction: verification vs. validation. Before the initial use of a referee procedure at a given laboratory, the analyst must perform a verification — a more limited demonstration that the established referee procedure performs as expected on the specific instrumentation, with the specific reference materials, and on the specific product matrix in use at that laboratory. This is meaningfully less burdensome than full validation, because the referee procedures themselves have already been validated by USP as part of the compendial chapter — what remains is confirming the procedure transfers correctly to your specific analytical environment, not re-proving that the underlying methodology works.

Alternative Procedures: When and How They're Justified

USP <233> explicitly permits the use of an alternative procedure when a specified compendial procedure doesn’t meet the needs of a specific application — General Notices 6.30 governs this option. This matters practically because not every pharmaceutical matrix, every required sensitivity level, or every specific element combination is necessarily best served by the referee procedures as written.

Alternative procedures must be validated and demonstrated equivalent to the compendial procedures for the purposes of the test, with the validation principles drawn from General Chapter <1225> (Validation of Compendial Procedures). Critically, where <233>’s own stated parameters and acceptance criteria differ from what <1225> would otherwise require, the parameters and acceptance criteria presented in <233> take precedence — the elemental-impurity-specific requirements in <233> override the more generic compendial validation framework wherever the two diverge.

The level of validation rigor required for an alternative procedure depends, just as with the referee procedures, on whether it’s being used as a limit test or a quantitative determination — the same accuracy, precision, and specificity framework described above applies, scaled to the procedure type.

How ICH Q3D and USP Connect

It’s worth being explicit about the division of labor between these two frameworks, since they’re often referenced together but serve distinct functions:

ICH Q3D identifies which elements require evaluation for a given drug product, provides the toxicological basis for the applicable Permitted Daily Exposure limits, and supplies the risk-based framework manufacturers use to structure their overall elemental impurity control strategy — covered in depth on our Pharmaceutical Heavy Metals Testing page.

USP <233> specifies the analytical procedures for actually measuring elemental impurities and the validation criteria that confirm a given measurement procedure is genuinely fit for the purpose ICH Q3D’s risk assessment assigned it. Where ICH Q3D answers “what should we be worried about, and at what level,” USP <233> answers “how do we generate a number we can trust, for the specific elements and limits ICH Q3D told us matter.”

This connection is also why product-specific risk assessment matters for validation planning: a validation study built around generic, non-specific target concentrations doesn’t map cleanly onto a chapter that increasingly organizes its requirements around the actual, product-specific PDE-derived limits a given drug product is subject to. Validation planning that starts from your specific ICH Q3D risk assessment — your specific elements of concern, at your specific applicable limits — produces documentation that holds up more cleanly against <233>’s requirements than validation built against a generic reference panel.

Method Selection: ICP-OES vs. ICP-MS Under

<233>’s two referee procedures map onto the same fundamental sensitivity tradeoff discussed across our pharmaceutical testing pages: ICP-OES handles parts-per-million-level work reliably and cost-effectively, while ICP-MS provides the sensitivity required for sub-ppm and ultra-trace requirements — particularly relevant for parenteral and inhalation product PDE limits, which can sit an order of magnitude or more below oral limits for the same element.

Because <233> validation requirements are scaled to the procedure and its intended target concentration, choosing the right base method (Referee Procedure 1 vs. Procedure 2, or an alternative procedure built around either platform) before validation begins is far more efficient than discovering partway through a validation study that the chosen instrument can’t reliably hit the accuracy and precision requirements at the concentration level your specific PDE limits demand.

Common Compliance Gaps We Identify

Who Uses USP Elemental Impurity Testing

Regulatory affairs and CMC teams building elemental impurity control strategies for new drug applications, who need analytical procedures with documented <233>-compliant validation supporting the overall submission.

Quality control laboratories at pharmaceutical manufacturers establishing routine release testing, choosing between limit and quantitative procedures based on actual control strategy needs rather than defaulting to the more burdensome option.

Method validation and analytical development teams planning validation studies for new products or transferring methods between sites, who need to scope validation work correctly against <233>’s specific requirements before study execution begins.

Contract manufacturers and CDMOs supporting multiple client products with potentially different applicable PDE limits, requiring validated procedures that map cleanly to each product’s specific risk assessment.

Quality and compliance teams reviewing existing validation documentation against the May 2026 harmonized chapter to confirm continued adequacy or identify where revalidation is warranted.

Sample Submission and Testing Scope

Required sample size and testing scope depend on whether you need referee procedure testing, verification of a referee procedure at our laboratory, full method validation support, or routine <233>-compliant release testing against an already-validated method.

Pricing for routine testing starts from $150 per sample for ICP-OES-based panels, with ICP-MS-based ultra-trace testing and validation study work priced according to scope.

For validation support: Sample quantities are scoped to the specific validation study design — accuracy studies, precision studies, and specificity/interference studies each require defined sample sets; we’ll provide a complete sampling plan once your validation scope (limit vs. quantitative, target elements, applicable J concentrations) is established.

Turnaround Time & Pricing

Standard routine testing turnaround: 3–5 business days Rush service: 24–48 hours available

Validation study turnaround depends on scope — limit procedure validation studies are generally completed faster than full quantitative procedure validation given the smaller required data set; we’ll provide a study-specific timeline once scope is confirmed.

Validation study turnaround depends on scope — limit procedure validation studies are generally completed faster than full quantitative procedure validation given the smaller required data set; we’ll provide a study-specific timeline once scope is confirmed.

What You Receive

For routine testing, clients receive a Certificate of Analysis documenting results against applicable limits with full method references.

For validation support, deliverables include:

Methods & Standards

Explore related services:

Request a Quote

Submit your product details, applicable elemental impurity classification, and testing or validation scope to receive a tailored quote and method recommendation.

Frequently Asked Questions

USP <232> establishes the elemental impurity limits — the numerical thresholds a result must meet. USP <233> establishes the analytical procedures and validation requirements that confirm a laboratory's method is capable of producing a trustworthy result against those limits. <232> sets the target; <233> defines how you're allowed to measure against it.
The chapter became formally harmonized across USP, the European Pharmacopoeia, the Japanese Pharmacopoeia, and the Indian Pharmacopoeia through Stage 4 of the Pharmacopeial Discussion Group process. Among the specific changes, "Quantitative Procedures" was renamed "Procedures for Quantitative Tests," reflecting tighter integration of ICH Q3D principles and a stronger organizational link to product-specific limits.
A limit procedure answers whether an elemental impurity is above or below a specified threshold — a pass/fail result. A quantitative procedure produces an actual measured concentration value. Limit procedures require validation of detectability, repeatability, and specificity. Quantitative procedures require the fuller package of accuracy, precision, intermediate precision, and specificity.
Referee Procedure 1 is the compendial method for elements generally detected by ICP-OES; Referee Procedure 2 is the compendial method for elements generally detected by ICP-MS. Both specify closed-vessel microwave digestion sample preparation. Using a referee procedure requires verification (not full validation) before initial use at a given laboratory.
Open-vessel or ashing-based sample preparation can produce low recoveries for volatile elements, since they're lost to the atmosphere during high-temperature open-system processing. Closed-vessel microwave digestion retains volatile elemental species for accurate measurement, which is why it's specified in both USP <233> referee procedures.
procedure doesn't meet your specific application's needs, under General Notices 6.30. The alternative must be validated and demonstrated equivalent to the compendial procedure, following General Chapter <1225> validation principles — except where <233>'s own specific parameters and acceptance criteria differ, in which case <233>'s requirements take precedence.
Spike recovery of 70% to 150% at each tested concentration across the 50%–150% J range, using three replicate standard reference material solutions and three replicate spiked test samples.