A certificate of analysis is not a grade; it is a set of measurements made under stated conditions, and the conditions are the part that carries the information. Two vials can both say 99% and be entirely different materials — one measured on a shallow gradient at 214 nm with a resolved chromatogram, the other measured in a way that hides everything eluting near the void. Reading a COA like a chemist means reading the method before reading the number.
Start with the header, not the percentage
Four header fields determine whether the rest of the document means anything: the product name with sequence, the lot number, the date of analysis, and the instrument or method identifiers. A COA without a lot number that matches the vial label is a marketing document. A COA without a date cannot be tied to a stability window. A COA that names no column, no gradient and no detection wavelength is asserting a result you cannot evaluate.
Sequence on the header matters more than people expect, because it is what the mass spectrum is checked against. If the sequence is absent, the identity confirmation has no stated target.
The chromatogram: what the purity number is actually measuring
Peptide purity on a COA is almost always reversed-phase HPLC area percentage: the area of the main peak divided by the total integrated area, expressed as a percentage. Three things follow from that definition.
First, it is a relative measure, not a mass measure. It says nothing about water, salt or counter-ion — only about the proportion of UV-absorbing material eluting as the main peak. Second, it depends entirely on detection wavelength. Peptide bonds absorb around 214–220 nm, so that is the standard analytical wavelength and the one that sees essentially everything peptidic. Detection at 280 nm only sees tryptophan, tyrosine and to a small extent phenylalanine — a peptide lacking those residues is nearly invisible at 280 nm, and reporting purity at 280 nm on such a sequence is meaningless. Third, it depends on the gradient. A steep gradient compresses closely related impurities — deletion sequences, oxidised species, diastereomers — under the main peak. A shallow, longer gradient resolves them and typically reports a lower, more honest number.
So the useful question is not "is it 98 or 99" but "what would this material report on a 1%/minute gradient at 214 nm on a C18 column". The method section is where you find out. More detail sits in HPLC purity explained and the definition at purity.
The mass spectrum: do the charge-state arithmetic yourself
Electrospray ionisation produces multiply charged ions, and for peptides above roughly 1,000 Da you will usually see a series rather than a single peak. The observed mass-to-charge ratio for a given charge state z is approximately (M + 1.00794z) / z, where M is the neutral average molecular weight.
Worked example with BPC-157, average molecular weight 1,419.55 Da, sequence Gly-Glu-Pro-Pro-Pro-Gly-Lys-Pro-Ala-Asp-Asp-Ala-Gly-Leu-Val. The singly protonated ion should appear near m/z 1,420.6; the doubly protonated ion near (1,419.55 + 2.02)/2 ≈ 710.8; the triply protonated near 474.2. If a COA shows a peak at 1,420.6 and the vial is labelled BPC-157, identity is consistent. If it shows 889 — the molecular weight of the TB-500 heptapeptide — the label and the contents disagree, and no purity percentage rescues that.
Two refinements separate a careful reader from a casual one. Average mass and monoisotopic mass are not the same number; for a 1.4 kDa peptide they differ by roughly 1 Da, and for a 4.7 kDa peptide by around 3 Da, so a small offset may be a units mismatch rather than an error. And a deconvoluted spectrum is more informative than a raw one, because it reports the neutral mass directly and shows whether adducts — sodium at +22, potassium at +38 — are inflating apparent mass. The full treatment is in mass spectrometry and peptide identity.
Net peptide content: the number that changes your maths
This is the most consequential field and the most commonly omitted. A vial labelled 5 mg is usually 5 mg of gross lyophilised solid. That solid contains peptide plus counter-ion plus residual water plus any excipient. For a peptide purified by reversed-phase chromatography with trifluoroacetic acid in the mobile phase, the isolated product is a TFA salt, and TFA content is not trivial: basic residues each carry a counter-ion, so a lysine- and arginine-rich sequence can carry a substantial mass fraction as trifluoroacetate. Add hygroscopic water uptake and the actual peptide content of a nominal 5 mg vial may be meaningfully below 5 mg.
Amino acid analysis or quantitative nitrogen determination is how net peptide content is measured properly. If a COA states it, your molar calculations are trustworthy; if it does not, treat concentrations as approximate and say so in your methods.
Red flags, ranked
| Signal | Why it matters |
|---|---|
| No lot number or a lot that does not match the vial | The document cannot be tied to the material in hand |
| Chromatogram with no axes, no scale, or no method conditions | The purity figure is uninterpretable |
| Identical retention times across chemically different peptides | Suggests a reused template rather than a real run |
| Observed mass that does not match the stated sequence | Identity failure; purity becomes irrelevant |
| Purity reported at 280 nm for a sequence with no Trp or Tyr | The detector could barely see the analyte |
| Flat image scan with no instrument metadata | Unverifiable; ask for the native report |
| No net peptide content, no water content, no counter-ion statement | Molar concentrations will be systematically overstated |
Verifying independently
The strongest position is not trusting a better-looking COA but having a second opinion on the same lot. Independent laboratories will run identity and purity on a submitted sample, and a vendor that publishes third-party reports alongside in-house ones is making a falsifiable claim rather than an assertion. How that process works, and what to ask for, is set out in third-party testing explained, with the step-by-step reading guide in how to read a peptide COA and the term itself at certificate of analysis. Recurring questions are answered in the purity, COA and testing FAQ.