Measurement Standards and Reference Materials: Selection, Control and Recalibration

A reference can be correctly identified yet unsuitable for a measurement. Its nominal value, an attached certificate or a future date does not, by itself, resolve questions about property, matrix, uncertainty and storage. The useful decision is more specific: which reference can support this measurement under the conditions in which it will actually be used?

The following pathway connects selection, documents and control over time. The matrix and case are original GuideGxP tools, not a universal procedure. Keep metrological control of references separate from specific requirements for compendial standards.

1. Distinguish the role before the commercial name

A measurement standard realizes a quantity with a stated value and uncertainty and serves as a reference: it may be a weight, a thermometric system or a material. A reference material, RM, requires sufficient homogeneity and stability for defined properties and use. Not all RMs have assigned values usable for quantitative calibration.

A certified reference material, CRM, adds authoritative documentation with property values, uncertainties and traceabilities obtained using valid procedures. These categories can overlap: a CRM used as a calibrator can serve as a measurement standard. “Analytical standard” often describes the use rather than proving the metrological category.

A commercial certificate of analysis does not automatically make a reagent a CRM. Check what is certified and for which purpose. Identity data or nominal purity do not automatically authorize assigning a concentration. For compendial applications, the responsible body's documents and instructions remain applicable; do not arbitrarily substitute the required standard.

2. Translate the measurement into selection requirements

Define the property or quantity, unit, level or range, matrix, preparation, environmental conditions and uncertainty needed for the decision. The right concentration in an incompatible solvent may be useless; a weight with a suitable value but excessive uncertainty may not support the intended check. Do not select solely the most precise reference in the catalogue: assess the whole chain of use.

Consider minimum sample quantity, homogeneity at the actual sampling scale, stability after opening and compatibility with extraction, dilution or reconstitution. A CRM certified for one property does not transfer that status to every listed property. Where necessary, commutability also requires evidence: it means behavior comparable to samples in the relevant procedures, not merely visual similarity.

ISO 17034 concerns competence of reference material producers. Relevant accreditation helps assess production but does not replace judging whether the selected material is fit for use. ISO 33403:2024 addresses use, ISO 33401:2024 documentation, and ISO 33405:2024 characterization, homogeneity and stability. Catalogues and abstracts, rather than full texts, were checked here.

3. Connect document, item and application

Check correspondence between material identity, lot or individual number where applicable, certificate and container. Retain the original version used and check producer updates. Read assigned value, units, expression basis, uncertainty and coverage factor where stated; do not treat mass fraction as mass concentration without a justified conversion.

The NIST SRM 1619b certificate dated 6 July 2026 distinguishes certified values from additional uncertified information and links validity to storage and use. It is a documentary example in a fuel-oil matrix, not a proposed pharmaceutical reference. It demonstrates why the heading “certificate” does not justify selecting any number found in the document.

At receipt record integrity, relevant transport conditions and storage location. Assign status and responsibility; connect opening, aliquots and preparations to the original reference and the measurements using it. Movement records help reconstruct the history, whereas metrological traceability concerns a result: they are not synonyms.

4. Original matrix for control over time

The matrix distinguishes recalibratable objects from consumable or alterable materials. Recalibration may update the value of an intact weight; it does not automatically restore composition, homogeneity or certified validity of a deteriorated solution. New characterization requires appropriate competence, method and documentation: it is not equivalent to changing a date on a label.

5. Manage working references, checks and dates

For a working reference document how its value was assigned, which contributions determine uncertainty and the authorized scope. Dilutions and transfers add conditions and potentially their own contributions; they do not inherit only the certificate's number. Material without an assigned value can support suitable precision checks but cannot, by itself, demonstrate trueness.

Avoid circular verification: when calibration and control depend on the same altered material, agreement may hide common error. Assess origin, lot, preparation and comparison method without assuming that a different label guarantees independence. Set criteria and actions before reading the result.

Distinguish declared validity, the limit after opening, an internal review date and a recalibration interval. Follow specific producer conditions and applicable obligations. Absence of an explicit date does not prove unconditional validity; a positive check does not automatically extend certification. If the producer updates the validity period, check that the update and conditions apply to the material you actually hold.

6. Simulated case: storage not demonstrated

A solution CRM, nominally 1 000 mg/L, has six months of validity remaining. In this hypothetical case its documentation requires storage between 2 and 8 °C, but 72 hours of records are missing. The bottle looks intact and its label remains legible. Those observations do not demonstrate that the certified value remained applicable throughout the undocumented interval.

Suspend use of the material, retain documentation and reconstruct conditions and exposure using available evidence and material-specific producer information. A recording gap does not automatically demonstrate an excursion or degradation; it does prevent assuming compliant conditions without justification.

Identify dependent preparations and measurements performed after the uncertain interval began. A subsequent comparison with a valid reference may help the investigation but cannot reconstruct the entire past by itself. The responsible function assesses restricted use, additional evidence or replacement, also justifying the impact on results. Do not backdate checks or automatically declare every result invalid or valid.

Explore metrological traceability and measurement uncertainty; return to the Calibration and Metrology hub for other tools.

7. Sources and applicability

Checked: 1 October 2026. Case and matrix are original; numerical conditions are simulated, not general prescriptions.

1. JCGM, VIM3: 5.1 measurement standard, 5.13 RM and 5.14 CRM: official entries consulted.
2. ISO: 17034:2016, 33401:2024, 33403:2024 and 33405:2024: catalogue records and abstracts only. ISO 17034:2016 is published, with revision under development; no unread clauses are attributed.
3. NIST SRM 1619b, certificate dated 6 July 2026: three pages consulted; an example of documentary conditions and limits, outside the case's pharmaceutical application.
4. NIST, Metrological Traceability: FAQ and policy: reference-user responsibility and fitness for purpose.