How to read a certificate of analysis
Sixteen fields, in the order they matter, and what the absence of each one lets you infer. Plus the arithmetic that reconciles gross vial mass with label claim.
The fields, in the order they matter
| # | Field | What it should say | What its absence implies |
|---|---|---|---|
| 1 | Lot or batch number | An identifier that also appears physically on the vial, not only on the box. | You cannot tie this document to the object in your hand. This is the most consequential omission on the list. |
| 2 | Date of analysis | A date after the manufacturing date. | If analysis precedes manufacture, the certificate belongs to a different lot. This happens more often than it should. |
| 3 | Product identity and claimed content | The compound, and the nominal quantity per vial. | Without a nominal figure there is nothing for a content assay to be measured against. |
| 4 | Method section | Column chemistry, dimensions, mobile phase, gradient, flow rate, detection wavelength, injection volume. | The purity number is not reproducible and not comparable to your own independent result. Treat it as a claim. |
| 5 | Purity, as area % | A figure with the wavelength stated. 214 nm and 280 nm give different numbers on the same injection. | A purity figure without a wavelength is incomplete by construction. |
| 6 | Chromatogram | An actual trace, with the baseline and the impurity peaks visible. | You cannot see whether the 1.5 per cent of non-target material is one impurity or eleven, which is the more informative question. |
| 7 | Impurity table | Retention times and area percentages for each resolved impurity above the reporting threshold. | You are being given a total without a distribution. |
| 8 | Identity confirmation | Observed mass against theoretical, with the charge state and whether it is monoisotopic or average. | "Conforms" without a number is not a measurement. |
| 9 | Quantified content | Milligrams per vial, measured against a named reference standard. | The commonest omission. This is the field that answers the question people think purity answers. |
| 10 | Reference standard | What the quantitation was calibrated against, with the standard’s own certificate referenced. | Every quantitative number on the page inherits the uncertainty of an uncharacterised standard. |
| 11 | Water content | Karl Fischer or loss on drying, as a percentage. | You cannot reconcile gross vial mass with net peptide. Expect its absence at this tier. |
| 12 | Counter-ion content | Acetate or trifluoroacetate, as a percentage by mass. | As above. TFA can be several per cent of what you weighed. |
| 13 | Residual solvents | Acetonitrile, DMF, DCM by headspace GC, against ICH limits. | Almost never present on research-grade documentation. Its absence is expected rather than suspicious. |
| 14 | Endotoxin | EU/mg by LAL or kinetic chromogenic method. | Research material is not released against a pyrogen specification. |
| 15 | Sterility | Growth or no growth in defined media over the full incubation period. | "Sterile filtered" is a process claim, not a result. The two are not equivalent and should not be read as such. |
| 16 | Signature and laboratory identity | A named laboratory with a verifiable report identifier. | A document that cannot be traced to its issuer is a document you are trusting on typography. |
Reconciling gross mass to label claim
This is where the purity-versus-content confusion becomes concrete. Suppose a vial is filled by mass to 11.62 mg of a lyophilised solid, and the certificate reports 99.0 per cent purity. How much peptide is in the vial?
| Step | Value | Working |
|---|---|---|
| Gross solid in the vial | 11.62 mg | What was weighed at fill |
| Less residual water at 4.0% | 11.16 mg | 11.62 × 0.960 |
| Less acetate counter-ion at 7.5% | 10.32 mg | 11.16 × 0.925 |
| Less related impurities at 1.0% | 10.22 mg | 10.32 × 0.990 |
| Net peptide | 10.22 mg | Against a 10 mg label claim: +2.2% |
That vial is in specification, and it is in specification because whoever filled it overfilled by 16 per cent to account for water and counter-ion. Now run the same arithmetic on a vial filled to 7.90 mg of solid at the same purity: net peptide 6.95 mg, against a 10 mg claim, a 30 per cent shortfall — with an identical purity figure on the certificate. The purity number cannot distinguish these two vials. Only a content assay can.
Why two competent laboratories disagree
They will, and by more than you expect. A spread of half a per cent to a per cent on the same vial is normal, and it is not evidence of incompetence on either side. The contributors, in rough order of magnitude:
| Cause | Typical effect on purity | Direction |
|---|---|---|
| Detection wavelength (214 vs 280 nm) | 0.4–1.2 pp | Depends on the aromatic content of the impurities |
| Gradient slope | 0.3–0.9 pp | Shallower resolves more impurities, so reports lower purity |
| Integration parameters and baseline | 0.1–0.5 pp | Either |
| Column chemistry and age | 0.1–0.4 pp | Either |
| Mobile-phase additive (TFA vs formic acid) | 0.1–0.4 pp | Either; TFA usually sharpens peaks |
| Injection mass and detector linearity | 0.0–0.3 pp | Overloading inflates apparent purity |
The practical consequence: a gap of half a per cent between your result and the supplier's is noise. A gap of three per cent, reproduced on a second lot, is a finding. Set that threshold before you send the sample.
A short checklist you can actually use
- Photograph the vial and the certificate together, so the lot numbers are in one frame.
- Check the analysis date against the manufacturing date. Wrong order means wrong lot.
- Find the method section. If there is not one, downgrade everything below it.
- Find the detection wavelength. Note it, because you will need it to compare against your own result.
- Look for a content figure in milligrams. If it is absent, you have a purity certificate and no content claim.
- Look at the chromatogram, not the number. Count the impurity peaks.
- Write down, before you send anything for independent testing, what result would make you reject the material.
- Keep the certificate, the lot number and the invoice in one place. The dispute you might need them for is six months away.
Further reading on this site: the questions tagged coa749 Certificates of analysis: what fields a useful one carries, how to tell a real analytical report from a marketing document, batch and lot… Purity as chromatographic area per cent - the fraction of detected material that is your target peak. It says nothing about how much material is… Quantified content: how many milligrams of peptide are actually in the vial, measured against a calibrated reference standard. A separate test… Certified reference materials and in-house standards: traceability, the certificate of the standard itself, system suitability, and why every…coa
purity
content-assay
reference-standard
PeptideMeter - Independent Peptide Analytics
Aggregated, published test results and vendor ratings built from submitted batches. Methodology stated, dataset browsable, no listing fees.
Browse results