How to read a certificate of analysis — what purity actually means on a peptide vial

A certificate of analysis is a record of what was measured on one batch of material, by whom, and by which method. It is not a warranty and it is not a specification. A specification states what a batch is supposed to be; a certificate states what one particular batch was found to be on the day a sample of it was tested. Read that way the document is genuinely useful. Read as a badge, it says almost nothing.

Nearly every certificate that arrives with a peptide vial leads with a single number — a purity figure, usually somewhere above 98%. That number is the most quoted and the least examined thing on the page. What follows is an account of what it is a percentage of, what the other analyses on the sheet are for, and which single check on arrival is worth more than the rest of the document put together.

On a peptide certificate, purity is almost always a chromatographic figure. A sample is separated by reversed-phase HPLC, a detector records what comes off the column against time, and the resulting peaks are integrated. The purity figure is the area of the main peak expressed against the total area of every peak the detector registered. McCarthy and colleagues describe the same arithmetic from the reference-standard side, where the impurity total is the sum of the detected impurity peak areas over the total detected area (3). Pharmacopoeial limits are written in those terms too: the United States Pharmacopeia monograph for calcitonin salmon sets individual impurities at not more than 3.0% and total impurities at not more than 5.0% of the total area of all chromatographic peaks, as reported by Zeng and colleagues (5).

So a certificate reading 99.5% says that 99.5% of the absorbance the instrument integrated belonged to one peak. It does not say that 99.5% of the mass in the vial is peptide. Those are two different statements, and only the first one has been measured.

Two things open the gap. The first is that the detector does not respond equally to everything it sees. Peptides without aromatic residues are commonly monitored near 214 nm, where the peptide bond itself absorbs; Kuipers and Gruppen measured the peptide bond's molar extinction coefficient at that wavelength as 923 M⁻¹ cm⁻¹ and found tryptophan absorbing roughly thirty times more strongly, with phenylalanine, tyrosine and histidine about six times more (4). An impurity rich in tryptophan is over-represented in the area count; one without it is under-represented.

The second is that a chromatogram counts only what reaches the detector as a resolved peak. Anything that does not elute, that co-elutes underneath the main peak, or that carries no chromophore at the monitored wavelength is absent from the arithmetic rather than counted against it. Zeng and colleagues put the consequence plainly for regulated peptide products, reporting that many of the HPLC-UV quality-control methods submitted for peptide drugs did not appear adequate to distinguish and quantify process impurities and degradation products, and that a high-resolution mass spectrometry method detected more peptide-related impurities in a calcitonin salmon nasal solution than the pharmacopoeial HPLC-UV method did (5).

A vial label states a mass. What that mass consists of is a separate question, and it is the question the rest of the certificate exists to answer.

The reference-standard literature answers it by mass balance: everything present that is not the intended peptide is identified and measured, and what is left over is the peptide. McCarthy and colleagues describe that accounting for a peptide reference standard — the sum of the HPLC impurities, acetic acid, trifluoroacetic acid, residual solvents and residue on ignition, with water content determined separately by Karl Fischer titration under the United States Pharmacopeia general chapter on water determination (3). The same work keeps apart two terms that a sales sheet tends to merge: purity is the fraction of the bulk material that is native peptide, while peptide content is the mass of peptide actually present in a vial (3).

The size of the gap is not hypothetical. Melanson and colleagues, assigning a value to a candidate certified reference material of angiotensin II, measured the trifluoroacetic acid counter-ion by a validated fluorine-19 qNMR method at nearly 25% of the mass, and assigned the material a final purity of 691 ± 9 mg/g (2). That is a rigorously characterised reference material rather than a poorly made one. Mass that is water, residual solvent or counter-ion is mass a chromatogram of peptide peaks was never counting in the first place, which is why a certificate that reports purity and nothing else has answered only part of the question.

Everything above is a way of reading a document that has already been handed over. The check that matters most is simpler than any of it and takes about ten seconds: compare the batch number printed on the vial with the batch number on the certificate.

It is the only test on this page that needs no instrument. A certificate that names no batch describes no particular material. A certificate whose batch differs from the vial describes some other material, however good its figures are. Neither failure requires anyone to understand a chromatogram to spot, and no amount of analytical detail elsewhere on the sheet compensates for either.

This is why Peptio publishes the issuer, the method and the batch on every certificate rather than a purity figure on its own: those three fields are what make the check possible. The issuer records who measured it, the method records what was measured, and the batch records which material was in front of them. A number without those three is a claim. A number with them is a record, and a record can be traced back to one vial.