HPLC purity is the area of the main peak in a chromatogram expressed as a percentage of the total area of all peaks — so "≥99% by HPLC" means related impurities together account for less than 1% of what the detector saw. It is the standard purity metric for research peptides because reversed-phase HPLC separates a target sequence from its closest chemical relatives — deletion sequences, oxidized variants, truncations — better than any other routine method. But the number has a precise scope, and most misuse of it comes from stretching the definition. This guide explains how the measurement works, what ≥98% versus ≥99% means in practice, and what an HPLC purity figure deliberately does not claim.
How the measurement works
In reversed-phase HPLC, the dissolved peptide is pumped through a column packed with hydrophobic (typically C18) particles while the mobile phase gradually shifts from water toward acetonitrile. Each species elutes when the solvent becomes strong enough to pull it off the packing — more hydrophobic species later — and passes a UV detector, usually set near 214–220 nm where the peptide backbone absorbs. The result is a chromatogram: signal versus time. Software integrates each peak's area, and purity = main-peak area ÷ total peak area × 100. Because retention depends on the column, gradient, temperature and mobile-phase additives, a purity number is only strictly comparable within one method — which is why a proper COA states its conditions.
What the impurities actually are
Peptides from solid-phase synthesis are built one residue at a time, and every coupling step is slightly imperfect. The characteristic by-products are deletion sequences (one residue missed), truncated sequences (synthesis capped early), incompletely deprotected species, and post-synthesis degradation products such as methionine oxidation (+16 Da) or deamidation. These relatives differ from the target by tiny amounts of hydrophobicity, which is exactly the axis reversed-phase HPLC separates on — hence its role as the purity workhorse. Longer sequences accumulate more error opportunities: holding a 31-residue chain like semaglutide to ≥99% is a materially harder synthesis task than a pentadecapeptide like BPC-157, and harder still than a tripeptide like GHK-Cu.
≥98% vs ≥99%: what the extra point buys
| Specification | Max related-impurity load | Typical context | Practical meaning |
|---|---|---|---|
| ≥95% | 5% | Screening-grade material, some catalog research peptides | Acceptable where impurity tolerance is high |
| ≥98% | 2% | Standard research grade | Suitable for most cell and animal model work |
| ≥99% | 1% | Our standard catalog specification | Halves the impurity budget vs 98%; preferred for quantitative and receptor work |
| ≥99.5%+ | 0.5% | Reference standards | Analytical comparator territory |
The jump from 98% to 99% sounds like one point but is a halving of the allowed impurity load — from 2% to 1%. Whether that matters depends on the research question: in a binding assay where a deletion sequence might be a partial agonist, the impurity budget is the experiment's noise floor; in a gross-effect rodent model it may be negligible. Catalog vials here are specified at ≥99% by HPLC with lot-matched COA, across lyophilized peptide vials and the raw material used in finished formats.
What HPLC purity does not tell you
- Identity. A single clean peak proves homogeneity, not correctness — the wrong sequence can also be 99% pure. Identity is the job of mass spectrometry, and the two sections of a COA are read together for exactly this reason.
- Net peptide content. Purity is a percentage of the peptide fraction. Water, counter-ion and residual solvent in the lyophilized powder sit outside the calculation, so 99% purity coexists with, say, 80% peptide by gross mass.
- Invisible-at-UV contaminants. Inorganic salts and anything that does not absorb at the detection wavelength or elute in the window will not appear as a peak. Endotoxin, likewise, is a separate assay, not an HPLC output.
- Aggregation state and biological activity. A chemically pure peptide can still be aggregated or misfolded; activity is established in the assay, not the chromatogram.
Reading a chromatogram like a reviewer
- One dominant, symmetric main peak. Shoulders or splitting suggest a co-eluting relative the integration may be hiding.
- A flat baseline. A drifting or humped baseline inflates or deflates area ratios.
- Small peaks acknowledged, not cropped. An honest report integrates the minor peaks; a cropped time axis is a classic way to flatter a number.
- Stated method. Column, gradient, wavelength, injection mass. No method, no comparison.
- Lot linkage. The trace belongs to your lot, verifiable against the vial — and independently checkable, as described in third-party testing explained.
Purity in practice: from number to experiment
Purity feeds three practical decisions. First, sourcing: comparing vendors on purity only makes sense when methods are comparable and lots are verifiable — the broader diligence list is in the vendor checklist. Second, quantification: for molarity-critical work, combine purity with net peptide content per molecular weight, moles and molarity before weighing anything into a calculation. Third, stability tracking: purity is also the metric that degrades — a re-run chromatogram after months of storage is how loss is quantified, which is why storage discipline per how to store peptides is ultimately an HPLC-visible variable. Purity is a snapshot at analysis time; handling decides how long it stays true.