Purity and Identity Are Not the Same Claim

Two questions that look like one

Purity asks: of everything in this vial, what proportion is a single component? Identity asks: what is that component?

In ordinary use the two collapse together, because a buyer reading '99% pure semaglutide' reasonably hears both claims at once. Analytically they are independent, and they are established by different measurements. A high purity figure is fully compatible with the wrong molecule, provided the wrong molecule is what is overwhelmingly present.

A sample that is 99% one substance is 99% pure. Whether that substance is the one on the label is a different measurement.

How the two come apart in practice

The cleanest example is a product name that does not fix a chemical identity. TB-500 is sold both as full-length thymosin β-4 and as a short fragment based on the LKKTETQ actin-binding motif. Their molecular weights differ by roughly a factor of five. Each can be manufactured to a high chromatographic purity, and a certificate reporting purity alone cannot tell you which one you received.

The same gap appears with truncated or deletion sequences, where a synthesis has dropped a residue. The product is a peptide, absorbs at the same detection wavelength, and may elute close to the intended compound. If it co-elutes it is not even resolved as a separate peak.

None of this requires anyone to be dishonest. A purity figure can be accurate, reproducible and correctly reported while the identity question has simply never been asked.

What establishes identity

Mass spectrometry is the usual first answer. An observed molecular weight matching the calculated mass for the expected formula is strong evidence that the intended molecule is present, and it is what separates the two TB-500 products immediately.

Intact mass has a known limit: molecules of the same mass are indistinguishable by it. Sequence isomers share a molecular weight, so a mass match does not by itself confirm the order of residues. Where that matters, fragmentation methods read sequence rather than only mass.

Retention time against a characterised reference standard, run under the same method, adds supporting evidence. On its own it is corroborative rather than conclusive.

What a certificate should let you check

A batch identifier, so the document corresponds to material rather than to a product line. A date. The methods used, with enough conditions stated that the purity figure means something. A purity result. A mass result. The name of the laboratory that produced it.

If a certificate carries a purity figure and nothing addressing identity, that is not necessarily a defect in the testing. It is a limit on what the document supports, and it is worth knowing which of the two questions you have an answer to.

Where this site holds a transcribed purity figure it prints it and calls it certified. Where it holds a published certificate without one, it says so rather than substituting the supplier's specification. Where there is no certificate, the figure shown is the supplier specification and is labelled as such on the same line.