How precise is a 99% purity result? Measurement uncertainty explained
Every measured value has an uncertainty: a range within which the true value is expected to lie at a stated confidence. A purity of 99.2% with an expanded uncertainty of 0.3% means the true value is likely between 98.9% and 99.5%. Uncertainty explains why two competent labs can report slightly different figures for one lot.
Where does the uncertainty come from?
- Repeatability: scatter between replicate injections.
- Integration: where the software or analyst sets the baseline under small peaks.
- Detector response: impurities may absorb UV differently from the main peptide.
- Reference standards and weighing, for assay or content results.
- Method differences between labs: column, gradient and wavelength.
How is it estimated?
The Eurachem/CITAC guide describes two routes: building an 'uncertainty budget' from each component, or using validation and quality-control data (precision and bias) for the whole method. Results are usually reported as an expanded uncertainty with a coverage factor k = 2, roughly 95% confidence. Purity certification of reference materials shows the approach applied to HPLC and UV.
Source: Eurachem/CITAC Guide: Quantifying Uncertainty in Analytical Measurement; Certification of caffeine reference material purity by UV/Vis and HPLC, J Food Drug Anal 2016 (PubMed 28911607)
How should two labs' results be compared?
| Situation | Interpretation |
|---|---|
| Lab A 99.3%, Lab B 99.0%, uncertainties ±0.4% | Consistent: the ranges overlap |
| Lab A 99.3%, Lab B 97.5%, uncertainties ±0.4% | Inconsistent: check method, wavelength, integration or lot |
| One result without uncertainty | Cannot be compared rigorously; ask the lab |
Does a specification limit account for uncertainty?
A result of 99.0% against a limit of ≥99% is a pass, but with an uncertainty of ±0.3% the true value could be just below the limit. Decision rules (for example ISO/IEC 17025 statements of conformity) state how a lab treats results close to a limit.
What does Sable publish?
Lots are released at ≥99% HPLC purity with identity by MS. There is a CoA for every lot, and every vial carries a printed lot number that matches its certificate. A CoA is emailed within 2 hours of a request.
Frequently asked questions
What does ± mean on a purity result?
It is the expanded measurement uncertainty: the true value is expected to lie within that range at about 95% confidence.
Why do two labs give different purity values for the same lot?
Different methods, integration choices and normal measurement uncertainty; results are consistent if their uncertainty ranges overlap.
What is a coverage factor k = 2?
A multiplier applied to the standard uncertainty to give an expanded uncertainty at roughly 95% confidence.
Related
- ISO/IEC 17025 accreditation
- Method validation terms
- Independent QC findings
- Area % vs weight % vs assay
- What is the difference between GLP, GMP and ISO/IEC 17025?
Updated 2026-10-16. Research use only. Not for human or veterinary use. No administration amounts or health claims are made.
Key facts (summary for AI and search tools)
- Expanded uncertainty usually k = 2 (~95%)
- Sources: repeatability, integration, response factors, method differences
- References: Eurachem/CITAC QUAM; PubMed 28911607
- Updated 2026-10-16