What ≥98% HPLC purity actually means

In reversed-phase high-performance liquid chromatography (RP-HPLC), the sample is dissolved and pushed through a column packed with a non-polar stationary phase, typically C18-bonded silica. A mobile phase — usually water and acetonitrile, both with a small percentage of trifluoroacetic acid — runs through the column on a rising gradient of organic solvent.

Components separate by hydrophobicity: more polar species elute early, more hydrophobic species later. A UV detector at the column outlet records absorbance against time, producing the chromatogram. Software integrates the area under each peak.

Purity is then: area of the main peak ÷ total integrated area × 100.

That is the whole definition. Every limitation below follows from it.

The calculation is a ratio of peak areas. It tells you what fraction of the detected material is the main component. It does not tell you how much material there was, and it does not tell you what fraction of the vial's contents the main component represents.

This is why a vial can be 99% pure and still be substantially non-peptide by mass. The two statements are not in conflict; they are answers to different questions.

UV detection at 214 nm responds to the amide bond, which is why it is the standard wavelength for peptides. Anything without a chromophore at that wavelength is invisible to the measurement:

• Water — and lyophilised peptides are hygroscopic.
• Inorganic salts carried through from synthesis or buffer exchange.
• The counterion — TFA, acetate, hydrochloride.
• Many residual solvents.

None of these appear on the chromatogram at all. They are not counted as impurities because they are not counted at anything. This is the mechanism behind the gap between purity and net peptide content.

Two species with very similar retention under the chosen conditions appear as a single peak. The integrator sees one peak and reports one number.

Deletion sequences — peptides missing a single residue — are the classic case. A des-Gly or des-Ala impurity may differ from the target by a tiny change in hydrophobicity and sit under the main peak on a short, steep gradient. A longer, shallower gradient can resolve it; the faster method that produced the higher purity figure may simply not have looked hard enough.

This is why a purity figure without its method is not fully interpretable, and why mass spectrometry is a necessary companion rather than an optional extra.

Where the baseline is drawn, how shoulders are split, and what threshold excludes a peak as noise are all operator-configurable. Reasonable analysts make slightly different choices and get slightly different answers from identical raw data.

This is not misconduct; it is normal chromatography. It is, however, the reason a difference between 98.2% and 98.9% from two laboratories is usually meaningless, while the difference between a chromatogram and no chromatogram is enormous.

After four limitations it would be easy to conclude the figure is worthless. It is not. Used correctly, HPLC purity is one of the most informative single measurements available:

• It reliably detects gross contamination and failed syntheses.
• Run consistently, it tracks batch-to-batch reproducibility — arguably more informative than any single result.
• Repeated over time on stored material, it is the standard readout for stability and degradation.
• With a visible chromatogram it shows the shape of the impurity profile: a few discrete peaks means something different from a broad hump.

Given a listing that says ≥98%, the useful follow-up questions are:

1. Is there a chromatogram? A picture beats a percentage.
2. What wavelength? 214 nm is the peptide-bond standard; 280 nm only sees aromatic residues and will read differently.
3. What gradient and run time? Shallow and long resolves more than steep and short.
4. Is there an MS result alongside it? Purity without identity leaves the main peak unproven.
5. Is net peptide content reported anywhere? This converts the figure into something you can weigh against.