Certified Reference Materials Explained: What They Are and How Labs Use Them
A certified reference material (CRM) is a material with one or more property values that are certified by a technically valid procedure, accompanied by a certificate that states the value, its uncertainty, and the level of metrological traceability. In practice, a CRM is the anchor that lets a laboratory prove its measurements are accurate, not just precise. It differs from a general analytical standard in one critical way: the certified value comes with a documented uncertainty budget and a traceability chain, which is what regulators and auditors look for.
CRM vs. Reference Material vs. Analytical Standard
These terms are often used interchangeably, but they carry different weight in a quality system.
| Term | Certified value? | Uncertainty stated? | Traceability documented? | Typical use |
|---|---|---|---|---|
| Certified Reference Material (CRM) | Yes | Yes | Yes | Calibration, method validation, compliance |
| Reference Material (RM) | Sometimes | Not always | Limited | Internal checks, rough verification |
| Analytical standard (working standard) | No (assigned in-house) | Usually not | Internal only | Routine calibration, day-to-day QC |
A CRM is produced under a quality system that meets ISO 17034, and the certifying body is typically accredited for that specific material. A reference material may be homogeneous and stable but lacks the formal certification. An analytical standard is often a solution you prepare or buy with a nominal concentration, and its accuracy depends on your own verification.
The practical rule: use a CRM when you need to demonstrate accuracy to an external party. Use a working standard when you are monitoring day-to-day drift and have already established your calibration with a CRM.
What Makes a CRM "Certified"
Four properties separate a genuine CRM from a well-characterized material.
Certified value. The concentration or property is determined by multiple independent methods or by a primary method, not a single measurement.
Uncertainty. The certificate states an expanded uncertainty (often at 95% confidence) that includes characterization, homogeneity, and stability components. This is not the same as your instrument's repeatability.
Traceability. The value links to a national or international standard, such as NIST or a national metrology institute, through an unbroken chain of comparisons.
Accreditation. The producer is accredited to ISO 17034, and the certificate identifies the scope. If a supplier cannot show this, treat the material as an RM, not a CRM.
How Labs Actually Use CRMs
Calibration
For techniques like ICP-OES, ICP-MS, or AAS, a CRM calibrant gives you a traceable calibration curve. This matters most when results feed into regulatory reporting or customer specifications.
Method Validation
During validation you need to show accuracy, precision, linearity, and selectivity. A CRM with a matrix similar to your samples lets you demonstrate recovery and bias. Spike recovery alone does not prove traceability.
Quality Control
A CRM run as an independent QC check catches bias that a calibration standard cannot, because it is a separate material from a different production lot. Many labs run a CRM at the start, middle, and end of a batch.
Instrument Verification
Periodic checks with a CRM confirm that an instrument still performs within specification after maintenance, a column change, or a long idle period.
Choosing the Right CRM
Selection is where most labs either save time or create audit findings. Work through these criteria in order.
- Matrix match. A water CRM will not validate a soil digest. Match the matrix as closely as practical to your sample type.
- Analyte and concentration range. Choose a certified value near the middle of your working range, not at the extreme.
- Uncertainty appropriate to your tolerance. If your method requires 5% accuracy, a CRM with 10% expanded uncertainty cannot support that claim.
- Certification documentation. Confirm the certificate states the value, uncertainty, traceability, expiry, and storage conditions.
- Packaging and volume. Single-use ampoules reduce contamination risk; larger bottles suit multi-run programs but require careful handling.
- Availability and lot continuity. If you report long-term trends, plan for lot changes and bridge them with overlap runs.
Handling, Storage, and Documentation
A CRM only supports compliance if you can show it was used correctly.
- Store at the temperature stated on the certificate, and log receipt and opening dates.
- Equilibrate to room temperature before opening to avoid condensation.
- Use clean, dedicated glassware or plasticware; contamination is the most common cause of a failed CRM result.
- Record the lot number, expiry, and the CRM result in your batch record.
- Do not use a CRM past its expiry unless you have documented evidence it remains valid.
- Keep certificates with your quality records; auditors will ask for them.
When a CRM Is Not Necessary
Not every measurement needs a CRM. If you are doing qualitative screening, internal process monitoring, or early method development, a working standard or RM is often sufficient and far cheaper. Move to a CRM when results are reported externally, when a method must be validated, or when a customer or regulation requires traceability.
Common Pitfalls
- Treating a supplier's "certified" label as equivalent to ISO 17034 accreditation.
- Using a CRM outside its certified concentration range and extrapolating.
- Ignoring matrix effects and using an aqueous CRM for a complex sample.
- Failing to bridge a lot change, which creates an unexplained shift in control charts.
- Assuming uncertainty on the certificate equals method uncertainty; they are different quantities.
Bottom Line
A certified reference material is defined by its certified value, stated uncertainty, documented traceability, and accredited production. Use CRMs for calibration, validation, QC, and compliance; use working standards for routine drift checks. Choose by matrix, concentration, and uncertainty, and treat handling and documentation as part of the measurement itself. Get those four things right and your results will stand up to both technical and regulatory scrutiny.