Reading an HPLC Chromatogram

The Two Axes

The x axis is time, almost always in minutes, running from injection to the end of the run. The y axis is detector response, typically shown in milli-absorbance units (mAU) for a UV detector, or in arbitrary intensity units if the trace comes off a charged aerosol or evaporative light scattering detector instead. The y axis is not concentration and it is not mass. It reflects however much material absorbs light at the wavelength the detector is set to, or however much material the detector responds to by whatever physical principle it uses.

Peptide bonds absorb UV light around 214 nm, which is why most peptide methods run there rather than at 280 nm, a wavelength that only picks up aromatic side chains from tryptophan, tyrosine, and phenylalanine. A peptide with no aromatic residues can look nearly invisible at 280 nm and perfectly normal at 214 nm. If a report does not state the wavelength, ask for it before comparing two chromatograms against each other, because a peak that looks small at one wavelength can be the dominant feature at another.

Retention Time and What Sets It

Retention time is when a compound elutes off the column, measured from the moment of injection. It is set by the interaction between the compound and the stationary phase, usually a C18-bonded silica packing for peptide work, under a given mobile phase composition, flow rate, and column temperature. Change any of those and retention time shifts, even for the exact same material run on the exact same instrument the following week.

Because of that sensitivity, retention time alone does not confirm what a peak is. It confirms that something eluted at a given point under a given method. Identity confirmation comes from mass spectrometry run alongside or after the HPLC trace, matching an observed mass to the expected mass of the sequence. A chromatogram tells you how clean the material is and roughly where the main component sits relative to its neighbors; a mass spectrum tells you what those neighbors actually are. Neither one alone gives the full picture, which is why a complete report includes both.

Peak Height Versus Peak Area

Peak height is the single highest point of detector response for a given peak. Peak area is the integral of the response curve over the time window the peak occupies, essentially the space enclosed between the peak and the baseline. Area accounts for both how tall a peak is and how wide it is, which is why quantitation is done on area rather than height in almost every peptide method.

Height is more sensitive to small shifts in peak shape that have nothing to do with how much material is present. A slightly wider, shorter peak and a slightly narrower, taller peak can carry the same area and represent the same amount of material, while their heights would suggest something different. A well-run report bases its purity calculation on area, and most chromatography software defaults to area percent unless someone has deliberately changed the method settings, so seeing area percent listed on a report is the expected, standard convention rather than an unusual choice.

Integration and Where the Baseline Is Drawn

Integration is the process by which chromatography software decides where a peak begins, where it ends, and where the baseline sits underneath it. Most instruments run an automatic integration algorithm that looks for a slope change to mark a peak's start and end, then draws a baseline connecting those two points. That baseline placement is what counts as background noise and excluded from the peak's area, and everything above it is counted toward that peak.

Analysts can override automatic integration manually, and sometimes need to, especially around partially resolved peaks. This is also where a report can be quietly shaped. Extending a baseline under a small shoulder peak, rather than giving that shoulder its own start and end points, folds it into the main peak's area and inflates the purity number without changing anything about the underlying material. A raw, unprocessed trace with visible tick marks at each integration boundary is the way to check that the baseline was drawn where it should have been, and a supplier willing to share that raw trace, not just the summary table, is a reasonable thing to ask for.

Shoulders and Tailing Peaks

A shoulder is a smaller peak that partially overlaps the main peak without fully separating from it, usually because a related impurity, often a deletion or truncation sequence close in structure to the target, has nearly the same retention behavior under the method. Shoulders are one of the clearest visual signs that a reported purity number understates the actual impurity load, since automatic integration can absorb a shoulder into the main peak rather than flagging it separately.

Tailing is different: it is an asymmetry in the main peak itself, where the back half trails off more slowly than the front half rises. Tailing usually points to interaction between the compound and residual, unbonded silanol groups on the column packing, to a column that is aging or overloaded, or to a mismatch between the sample's solvent and the mobile phase at the point of injection. A tailing factor, calculated at 5 percent of peak height, is often reported alongside the trace, and values much above 1.5 are worth asking about, since severe tailing can also hide a co-eluting impurity underneath the broadened trailing edge of the main peak.

Why Area Percent Is Not Mass Percent

Area percent purity assumes that every component in the mixture absorbs light at the detection wavelength in direct proportion to how much of it is present, and at the same rate as every other component. That assumption rarely holds exactly. UV response depends on a compound's extinction coefficient, which is driven by its chromophores, and a deletion sequence missing an aromatic residue will absorb less light per unit mass than the parent peptide, even at 214 nm, since the peptide bond itself is the dominant chromophore there but aromatic side chains still contribute.

The practical effect is that area percent is a good, standard proxy for purity, not a direct measurement of mass fraction. Two peaks of equal area are not guaranteed to represent equal amounts of material by weight. Where mass-accurate quantitation matters, it is done against a reference standard with a known extinction coefficient, or by an orthogonal method such as amino acid analysis, not by reading area percent off a single HPLC trace. This gap between area and mass is small for most peptides and most impurities, but it is not zero, and it grows larger the more a given impurity's aromatic content differs from the target sequence's own composition.

Gradient Versus Isocratic Runs

An isocratic run holds mobile phase composition constant for the entire method, for example a fixed ratio of water to acetonitrile the whole way through. A gradient run instead ramps the percentage of organic solvent up over time, moving from a mostly aqueous mobile phase toward a higher-acetonitrile mobile phase as the run progresses.

Most peptide methods use a gradient because peptides span a wide range of hydrophobicity depending on sequence, and a single fixed solvent strength would either elute early components too fast to resolve them or hold late components on the column far longer than practical. A typical peptide gradient might run from 5 percent to 65 percent acetonitrile with 0.1 percent trifluoroacetic acid as an ion-pairing modifier, over 20 to 30 minutes. The gradient profile, expressed as percent organic per minute, belongs on the method line because it changes where and how sharply everything elutes, and a shallower gradient across the region where the main peak sits will generally give closely related impurities more room to separate from the target than a steep one will.

Reading the Method Line

The method line on a report, sometimes only a single sentence at the bottom of the chromatogram page, is where the analytical conditions live. A complete one states the column (dimensions, particle size, pore size, and bonded phase, for example 4.6 x 250 mm, 5 micron, 300 Angstrom C18), the flow rate, the detection wavelength, the mobile phase composition and modifier, and the gradient profile or isocratic hold time.

• Column: chemistry and dimensions set what resolves from what
• Flow rate: usually 1.0 to 1.5 mL/min for an analytical column
• Wavelength: 214 nm is standard for peptide backbone detection
• Mobile phase and modifier: water and acetonitrile with a trifluoroacetic acid ion-pairing agent, most commonly
• Gradient or isocratic profile: how solvent strength changes over the run

A report missing any of these lines is not automatically wrong, but it removes your ability to compare two lots run under different conditions, or to judge whether a shoulder that shows up on one instrument's method would resolve better on another. Instruments from different manufacturers, such as a Waters Alliance system versus an Agilent 1260 Infinity, can produce slightly different traces for the same material even under nominally identical methods, which is one more reason the method line, not just the instrument name, is the detail worth checking.