What purity actually means

The number everybody quotes

Somebody in a comment thread says a supplier tests at ninety-nine point something and the conversation stops there, as though a decimal had settled it. Our quality circle spends more time on this single number than on anything else, because almost every misplaced confidence in this market rests on a misreading of it.

Here is the short version, and the rest of the guide is the working. A purity figure tells you what proportion of the material the instrument detected in one tested sample was the substance of interest. It does not tell you how much peptide is in your vial. It does not tell you whether the vial is sterile. It does not tell you what else came along that the instrument could not see. And it is about a sample, which — unless you commissioned it yourself from the vial in your hand — is somebody else’s material.

Sunil is not a chemist and says so at the start of every session; several members are, and they correct him, which is how this guide came to be as careful as it is.

What an HPLC purity figure measures

High-performance liquid chromatography pushes a dissolved sample through a column that separates components by how strongly they stick to it. Things emerge at different times, and a detector — usually ultraviolet absorbance at a set wavelength — records a signal as each one passes. You get a trace with peaks, and purity is normally reported as the area of the main peak as a percentage of the total area of all peaks.

Three consequences follow immediately, and they are the whole point of this section. First, it is a relative measure: it compares the main peak with the other peaks the detector saw, not with the true composition of the vial. Second, anything that does not absorb at the chosen wavelength is invisible — salts, many solvents, water, and much inorganic material simply do not appear in the denominator. Third, anything that emerges at the same time as the main peak is counted as the main peak. Closely related peptide impurities, in particular, are exactly the substances most likely to co-elute, because they are structurally similar by definition.

So the honest reading of ninety-nine per cent by area is: of the material that dissolved, ran, absorbed at that wavelength and separated from the main peak, ninety-nine per cent of the signal was the main peak. That is a real and useful result. It is much narrower than what people take it to mean.

Purity is not content

This is the distinction that costs people the most and appears on the fewest certificates. Purity asks what proportion of the peptide material is the right peptide. Content — net peptide content — asks how much peptide is in the powder you weighed.

Lyophilised peptide is not pure peptide. It carries a counterion from purification, commonly trifluoroacetate or acetate, and it carries residual water, and it may carry residual solvent. Depending on the peptide and the process, the powder in the vial can be substantially less peptide by mass than the label weight suggests, and there is nothing anomalous about that — it is ordinary chemistry. Established practice for research peptides is to state net peptide content separately, determined by amino acid analysis or by nitrogen determination, precisely because purity cannot answer it.

A vial can therefore be genuinely ninety-nine per cent pure and deliver materially less active substance than you calculated. Both statements are true at once. If you are reconstituting and dosing from a stated milligram figure, this is not an academic point: it is a systematic error in your arithmetic that no amount of careful measuring will reveal. Ask whether a figure is salt-corrected. Most of the time nobody will be able to tell you.

What mass spectrometry does and does not prove

Mass spectrometry ionises the molecule and measures mass-to-charge ratio very precisely. When a certificate says identity confirmed by MS, it usually means the measured mass matched the theoretical mass of the intended peptide within a stated tolerance. That is genuinely meaningful — it rules out a great many substitutions, because a different molecule generally has a different mass.

What it does not establish is more interesting. Mass alone does not fix sequence order: two peptides with the same amino acid composition in a different order have the same mass, and distinguishing them needs tandem mass spectrometry, which fragments the molecule and reads the pieces. Mass does not distinguish stereochemistry at all — a D-amino acid substitution weighs exactly what the L form weighs, and racemisation during synthesis is a known failure mode. Mass says nothing about aggregation state, about whether the peptide has already partly degraded into fragments that were filtered out before analysis, or about anything that is not a peptide.

So identity by MS plus purity by HPLC is a reasonable pair of questions to have answered, and it is still two questions out of a dozen that matter.

The long list nobody puts on the certificate

Write this list somewhere, because it is the honest inventory of what a purity panel leaves untouched.

• Sterility. Not tested, not implied, not achievable at home. Sterility testing is a defined microbiological procedure and it is not part of any chromatography.
• Bacterial endotoxin. A separate assay entirely. Endotoxin survives sterilisation, passes through a filter, and is the reason a technically sterile injection can still cause fever and rigors.
• Residual solvents. Determined by gas chromatography against defined limits in licensed manufacture. Absent here, and invisible to the ultraviolet detector.
• Elemental impurities. Heavy metals from catalysts, reagents and equipment. A different technique again, and one nobody in this market commissions.
• Fill accuracy. Whether the vial contains the stated mass at all. Weighing variation between vials in a hand-filled run is not hypothetical.
• Water content. Which changes both mass calculations and stability.
• Storage and stability history. The report describes a sample at a moment. It cannot describe a warehouse, a courier, or a fortnight in a customs shed.
• Provenance. Whether your vial came from the tested lot at all.

Reference materials, and why the comparator matters

Analytical identity is comparative. To say a peak is semaglutide, you compare its behaviour with something you already know is semaglutide — a certified reference material with documented identity, purity and uncertainty. Suppliers of certified reference materials, Sigma-Aldrich among them, exist for exactly this reason, and a laboratory’s access to appropriate reference standards is part of what separates a competent result from a plausible one.

The practical consequence for you as a reader is a question to ask of any certificate: against what was this identified? A retention time on its own is only meaningful relative to a standard run on the same system under the same conditions. Where no reference standard is stated, identity is inference from mass and expectation rather than a comparison.

This is also why laboratory accreditation is worth understanding rather than treating as a badge — third-party testing, who does what goes through it, including the uncomfortable detail that accreditation is granted for specific methods within a defined scope and does not automatically cover everything a laboratory offers.

How to hold a purity number honestly

Not "this is meaningless" — that is its own kind of laziness, and it is the position people retreat to when they realise the number is narrower than they hoped. Analysis is real information. It just has edges.

Hold it like this. A purity figure from a competent laboratory, on a sample drawn from the lot you are actually receiving, commissioned by somebody with no interest in the answer, is a reasonable piece of evidence that the material is what it says it is and is not grossly adulterated. Every clause in that sentence is doing work, and in practice at least one of them is usually missing.

And even with all of them present, the list two sections above is unchanged. Sterility, endotoxin, solvents, metals, fill accuracy, storage, provenance: still open, still unanswerable from a certificate. That is not scepticism for its own sake. It is the difference between a claim you can rely on and a claim you have decided to rely on, and only one of those is a decision made with open eyes.