HPLC purity: what 99% means, and what it leaves out

Purity is the number most people read first. It is a precise answer to a narrow question.

1How HPLC separates a sample

Chromatography separates the components of a sample between two phases: one stays still, the other moves^[1]. In high-performance liquid chromatography (HPLC), a dissolved sample is pumped through a packed column. Each component moves at its own speed and leaves the column at its own time. A detector at the end records a signal, and each component draws a peak.

2From peaks to a percentage

The harmonised chromatography chapter of the European, US and Japanese pharmacopoeias describes the normalisation procedure. In it, a component's percentage is its peak area as a share of the total area of all the peaks. Peaks from solvents, reagents, the mobile phase or the sample matrix are left out, and so are peaks at or below the reporting threshold^[1].

Purity (%) = main peak area ÷ total area of the counted peaks × 100

3What the number leaves out

• What hides under the main peak. The pharmacopoeia defines a peak as the detector's response to one component, or to two or more components that were not separated^[1]. An impurity that leaves the column with the main peak (4 in Figure 1) is counted as main peak.
• What falls below the threshold. Peaks at or below the reporting threshold are not counted^[1].
• Counter-ions, salts and water. Synthetic peptides can carry counter-ions such as trifluoroacetate from their synthesis and purification^[2]. These are not part of the peptide peak, which is one reason purity is not the same as content.
• Identity. Purity assumes the main peak is the right compound. Identity is tested with other methods, such as mass spectrometry^[3]. The identity article explains how.

4The impurities a synthetic peptide can carry

Most peptides are made by solid-phase peptide synthesis, which builds the chain one amino acid at a time. A review of peptide impurities sorts what can go wrong into two families^[2].

• From synthesis: missing or extra amino acids (deletion and insertion sequences) and amino acids turned into their mirror image (racemization). Also leftover protecting groups, oxidation, and chains joined together (dimers).
• From degradation: chemical changes such as beta-elimination, diketopiperazine, pyroglutamate and succinimide formation.

Many of these molecules are close relatives of the peptide itself. Some, such as mirror-image forms, need specific methods to tell apart^[4]. This is why a good purity figure comes from a method that separates them, and why it is read together with an identity test.

5Purity on this site

9 of the 11 lots on this site have an HPLC purity figure. They run from 97.502% (3P1TH4L0N) to 99.903% (T1RZ3P4T1D3). The two blends (8PC Wolverine Stack, C1C-1295 + 1P4M0R3L1N) are reported as measured content for each peptide instead of one purity figure.

6How to read a purity figure

• Check that it is HPLC purity, for your lot, from an independent lab, with a report you can open at the lab.
• Read it with the identity result and the measured content: the three answer different questions.
• A purity number without a method, a lot or a report number is a claim, not a measurement.