Lab testing & reading a COA

01 Why two tests, not one

Purity and identity are separate questions, and no single instrument answers both well. A batch can be 99% pure and still be the wrong compound. It can also be unambiguously the right compound and be half excipient.

HPLC answers "how much of this sample is one thing?". LC-MS answers "is that thing what it claims to be?". Run together they close the gap that either alone would leave open. A certificate that reports only one of them is telling you half the story.

02 HPLC — purity

High-performance liquid chromatography pushes the dissolved sample through a packed column under pressure. Components travel at different speeds depending on how strongly they interact with the packing material, so they emerge separated in time. A detector at the outlet records what leaves and when, producing the chromatogram.

Purity is reported by area normalisation: the area under the target peak, divided by the total area of all peaks, expressed as a percentage. A 98.6% result means the target accounts for 98.6% of everything the detector saw.

That last qualifier matters. Area normalisation is blind to anything the detector does not respond to — residual water, most salts, counter-ions from synthesis. It is an excellent measure of peptide-related impurities and a poor measure of total mass, which is why peptide content is a separate determination.

03 LC-MS — identity

Mass spectrometry ionises the separated material and measures its mass-to-charge ratio. Every peptide sequence has a calculable exact mass, so the observed value either matches the expected one or it does not.

This is what catches substitution outright. A cheaper sequence sold under an expensive name will pass a purity test comfortably and fail a mass check immediately, because the number is simply wrong.

It also catches common synthesis defects — a deletion sequence missing one residue, or an incompletely deprotected product — which show up as satellite masses offset from the target by the mass of the missing or extra group.

04 Beyond purity and mass

Endotoxin (LAL)

Detects bacterial endotoxin, which survives sterilisation and is not visible in any chromatogram. Reported in EU/mg.

Sterility

Confirms no viable microbial growth in the filled vial.

Water content

Karl Fischer titration. Residual moisture drives hydrolysis during storage, so a dry cake is a stable cake.

Peptide content

How much of the vial mass is peptide rather than water and counter-ion. Distinct from purity, and often confused with it.

Appearance

Recorded before release: a white to off-white lyophilised cake, free of discolouration.

05 Reading a Certificate of Analysis

A certificate is a record of one specific batch on one specific date. Anything presented as a certificate "for the product" rather than for a lot number is marketing, not analysis.

• Batch or lot number, and it must match the number printed on the vial in your hand.
• The compound name and its expected molecular formula and mass.
• HPLC purity as a percentage, with the chromatogram reproduced rather than merely summarised.
• Observed mass from LC-MS, alongside the theoretical value for comparison.
• Test date, and the date the batch was released.
• The testing laboratory, its accreditation reference, and a case or report number the laboratory can confirm.
• A signature or authorisation from the analyst or laboratory director.

On the chromatogram itself, look at the baseline and the peak shape as well as the number. A clean flat baseline with one sharp symmetrical peak is a good result. A drifting baseline, a broad peak, or shoulders on the main peak all suggest impurities that the reported figure may have integrated away.

06 Fabricated certificates

Falsified certificates are common enough in this market that treating any PDF as evidence is a mistake. A document is trivially edited; the underlying analysis is not.

Services exist that will issue whatever result is requested for a fee. They are recognisable: they operate through forums or messaging apps rather than a registered business address, they take payment in cryptocurrency, they cannot be contacted independently to confirm a report, and they never assign a case number that survives verification.

• Confirm the laboratory exists independently of whoever sent you the document, and contact it through details you found yourself.
• Check that the case or report number resolves when you query the laboratory directly.
• Compare the batch number on the certificate against the one on the vial. A mismatch ends the enquiry.
• Be sceptical of a chromatogram image with no axis labels, no retention times, or a baseline too clean to be a real trace.
• Treat a purity figure quoted without a chromatogram as an unverified claim.

07 If you test independently

We would rather you did. An independently confirmed batch is worth more to us than an assertion from us that it is fine.

To assess a third-party result against our retained sample for the lot, we need the batch number, the full report including the chromatogram, and the laboratory's case reference so we can verify the report is genuine. Results from anonymous or forum-operated testing services cannot be assessed, because there is no way to establish which sample was analysed.

Where an independently verified result materially contradicts the certificate we issued, we will replace or refund the order and pull the remaining stock of that lot pending re-analysis. We do not reimburse the cost of testing you commissioned.