what is hplc, and how it measures peptide purity

Chromatography works by pushing a sample through a system that different molecules travel through at different speeds. The molecules that interact more strongly with the system move slowly; the ones that interact weakly move quickly. By the time everything exits, the components have separated in time.

For peptides, the specific technique is almost always reverse-phase HPLC (RP-HPLC). The sample is dissolved and injected into a stream of liquid (the mobile phase, usually water and acetonitrile with a small amount of an acidic modifier) that flows through a column packed with a hydrophobic material (the stationary phase, typically a C18 carbon chain bonded to silica).

Peptides stick to that hydrophobic surface to varying degrees depending on their sequence and shape. As the proportion of acetonitrile is gradually increased (a "gradient"), each component releases from the column and flows past a detector. Most peptide analysis uses ultraviolet detection at 214 nm or 220 nm, wavelengths where the peptide bond itself absorbs light. The detector records a trace called a chromatogram: a series of peaks, each one a component of the sample, plotted against the time it took to emerge (its retention time).

The purity number on a COA is an area-percent calculation. The detector measures the area under each peak, which is proportional to how much of that component is present. Purity is the area of the main peak (the target peptide) divided by the total area of all peaks, expressed as a percentage.

purity (%) = (area of the target peak ÷ total area of all peaks) × 100

A batch reported at 98.5% means the main peak accounts for 98.5% of the total detected signal, and the remaining 1.5% is distributed across smaller peaks: synthesis byproducts, truncated sequences, or degradation products, collectively called related substances.

This is where careful reading matters. An HPLC purity figure is precise, but it is narrow.

Three specific limitations are worth internalizing:

It is a relative measure, not an absolute one. Area-percent assumes every component responds to UV light similarly. That is an approximation. An impurity that absorbs weakly at 214 nm contributes a smaller peak than its true abundance, and a component that does not absorb at all is invisible to a UV detector.

A single wavelength can miss things. Rigorous analysis sometimes runs detection at more than one wavelength, or pairs HPLC with mass spectrometry, precisely because co-eluting or non-absorbing species can hide under or beside the main peak.

Purity is not identity. A batch can be 99% pure and still be the wrong peptide, if the synthesis produced a clean but incorrect sequence. This is why a credible COA reports HPLC purity and a mass-spectrometry identity result. The two answer different questions: HPLC asks "how much of one thing," mass spec asks "what is that thing." See LC-MS and peptide identity confirmation.

If a vendor publishes the actual chromatogram rather than just a number, a few features are worth looking at:

• A single, sharp, symmetrical main peak. Broad or tailing peaks can indicate column or method issues; a clean peak is a good sign.
• Small, well-separated minor peaks rather than a cluster of shoulders fused to the main peak. Fused shoulders can inflate the apparent purity because the integration software may not resolve them.
• A stated method and wavelength. A purity figure reported with its analytical conditions (column, gradient, detection wavelength) is far more meaningful than a bare percentage.

For how this figure fits alongside the other parameters on a certificate, see the pillar guide, reading a Certificate of Analysis.