← Back to blog

How to Interpret an HPLC Chromatogram: A Practical Guide

November 10, 2026
How to Interpret an HPLC Chromatogram: A Practical Guide

An HPLC chromatogram is a plot of detector signal against time. To interpret one, find the baseline, identify each peak by its retention time against a reference standard run under the same conditions, check that the peaks are well shaped and properly separated, and only then use peak areas to say how much of each component is present. The sections below take those steps in order, with a worked example.

This guide is written for laboratory staff, students and researchers who need to read a chromatogram with confidence, whether it is their own run or one attached to a supplier's certificate.

Bioforgeuk
Source Peptides for Research
BioForgeUK supplies high-purity research peptides, with 99%+ HPLC purity, for laboratory and scientific research.
Explore research peptides

TL;DR

  • The horizontal axis is time and the vertical axis is detector response. Each peak is the response as one component leaves the column.
  • Retention time identifies a peak only when it matches a reference standard run on the same system. Peak area, not height, is the usual basis for quantitation.
  • Check the baseline, peak symmetry and resolution before trusting any number taken from the trace.
  • An area percentage is a useful purity estimate, but it depends on the detector wavelength and on whether impurities are resolved from the main peak.
  • Always record the method conditions alongside the chromatogram, because the trace means little without them.

Table of Contents

What a chromatogram actually shows

A chromatogram is a two-dimensional plot. According to Shimadzu's HPLC basics, the vertical axis shows detector signal intensity, which tracks concentration, and the horizontal axis shows analysis time. When nothing is eluting from the column, the trace is a flat line parallel to the time axis: the baseline. A peak is the bell-shaped response produced as an analyte passes through the detector.

Waters describes the same idea as a time-axis plot of detector responses in which each peak represents a different compound. That is the sentence to hold on to: one peak, one component, at least in principle. The exceptions (two components hidden in one peak) are covered further down.

The British Pharmacopoeia gives the formal definition. In its chapter on chromatographic separation techniques, a peak is the part of the chromatogram recording the detector response while a single component, or two or more unresolved components, elutes. The same text says a peak can be described by its area, or by its height and its width at half height, or by its height and its width between the points of inflection.

The definition already contains a warning. A single peak may be more than one substance, so a tidy-looking peak is not proof of a pure one.

Before you read any peak: the method header

A chromatogram printed without its conditions is hard to interpret. Before looking at the trace, note:

  1. The column (chemistry, length, particle size) and the mobile phases.
  2. Whether the elution is isocratic or a gradient, and the gradient programme.
  3. The flow rate and column temperature.
  4. The detector and wavelength, for example UV at a stated nanometre value.
  5. The injection volume and sample concentration.

Retention time and peak size both depend on these. Two chromatograms of the same material can look very different if the wavelength or gradient differs.

Reading retention time

Retention time (tR) is the time from the injection point to the apex of the peak, as Shimadzu UK explains. The same page defines dead time (t0) as the time a non-retained compound takes to travel from the injector to the detector. Anything that appears at t0 has not interacted with the column at all, so an early peak near t0 usually means unretained material such as solvent front or very polar species.

The Sepscience HPLC Back to Basics guide measures retention from injection to the highest point of the peak, and uses t0 to work out the retention factor, k. It describes the best chromatography as lying between k values of 2 and 10, with 1 to 20 usually acceptable for isocratic separations. Put simply, a peak that elutes too close to t0 is poorly retained and risks overlapping with unretained material, while one that elutes very late wastes time and broadens.

What retention time can and cannot tell you

Retention time is a property of a compound on a specific system. Mobile phase composition, temperature, flow rate, pH and column condition all shift it. For that reason a retention time quoted from a different laboratory or column is a guide at best. The retention time of a peak in your own chromatogram becomes meaningful only when you compare it with a standard run under the same conditions.

The European Pharmacopoeia chapter on chromatography makes a related point. EDQM notes that the revised chapter 2.2.46 clarifies that retention times and relative retentions in monographs are given for information rather than as requirements, while compliance with the system suitability test is always required.

Peak height versus peak area

Both measure how much of a component reached the detector, but they behave differently.

  • Peak height is the vertical distance from the baseline to the apex. Waters notes that a more concentrated compound gives a greater peak height above the baseline.
  • Peak area is the area enclosed by the peak and the baseline. Shimadzu describes it as used for both qualitative and quantitative analysis, and Waters treats it as the measure of the concentration of the compound the peak represents.

Area is generally preferred for quantitation because it accounts for the whole peak, including its width. If two injections give peaks of the same height but one is broader, they do not contain the same amount, and only area shows that. Height can be reasonable when peaks are narrow, symmetrical and consistently shaped, but it becomes unreliable as soon as peak width changes between runs.

Integration matters

The area your software reports depends on where it places the baseline and where it starts and ends each peak. The WHO guidance on pharmaceutical quality control laboratories (Annex 4) says peak areas should be integrated accurately, consistently and in a scientifically sound manner. It adds that where unresolved peaks must be quantified, a laboratory should have clear policies on how they are integrated, with the type of integration described and justified. Atypical peak shapes should be investigated.

In practice: look at the integration marks on the chromatogram, not just the table of numbers. A baseline drawn through a shoulder or a small impurity peak that has not been integrated at all can change a purity figure noticeably.

Identifying peaks with reference standards

A peak has no name until you give it one, and the evidence for that name is a reference standard. Waters recommends comparing the retention times of sample peaks with those of reference standards injected on the same system, meaning the same mobile and stationary phases.

A practical routine:

  1. Inject the reference standard alone and record its retention time.
  2. Inject the sample under identical conditions.
  3. Check whether the main sample peak has the same retention time, within your laboratory's acceptance window.
  4. For extra confidence, spike the sample with a small amount of the standard. One peak that simply grows is consistent with the same compound. A new peak or a split peak suggests they are different.

Matching retention time is necessary but not sufficient. Different compounds can co-elute. For that reason identity is usually confirmed with an orthogonal technique such as mass spectrometry. Our guides on LC-MS peptide analysis and reading mass spectra of peptides explain how that confirmation works.

External standards and quantitation

For quantitation, a validation reference describes external standard quantitation as comparing the peak area or height of the sample with a reference standard analysed separately, as set out in this chromatographic validation document. The same document recommends including a purity correction factor, where known, for non-compendial reference standards. If your standard is itself 98% pure, ignoring that will bias every result by the same amount.

For guidance on choosing and checking standards for peptide work, see our article on HPLC calibration peptides.

Estimating purity from the chromatogram

The simplest purity estimate is area percent: the area of the main peak divided by the total area of all integrated peaks, multiplied by 100.

A worked example

Imagine a UV chromatogram with three integrated peaks (illustrative numbers, not a real batch):

PeakRetention time (min)Area (mAU*s)Area %
1 (impurity)8.4200.4
2 (main peak)9.14,95099.0
3 (impurity)10.3300.6
Total5,000100.0

Main peak area divided by total area gives 4,950 / 5,000 = 0.99, so 99.0% by area at that wavelength. The figure means that, at the detection wavelength used and among the peaks that were integrated, the main component accounts for 99.0% of the response.

That qualifier matters, for the reasons below.

Why area percent is an estimate, not a verdict

Different compounds respond differently. ChemGuide explains that you cannot compare the concentrations of two different substances from their peak areas, because a smaller peak may reflect a lower concentration or simply weaker UV absorption at the chosen wavelength. It also notes that for two samples of the same substance, the ratio of peak areas does reflect the ratio of concentrations. So comparing batches of one compound is more reliable than comparing one compound with another.

Hidden impurities. Anything that co-elutes with the main peak, or does not absorb at the detection wavelength, is invisible in the area percentage.

Integration choices. Ignoring peaks below a threshold, or merging a shoulder into the main peak, shifts the result.

For a fuller treatment aimed at peptides, read how to read and judge an HPLC purity number.

Judging peak shape

Shape tells you whether the separation was behaving well, and whether the area can be trusted.

The British Pharmacopoeia defines the symmetry factor (As) using the peak width at one-twentieth of the peak height. A value of 1.0 means a symmetrical peak, above 1.0 means tailing and below 1.0 means fronting. For a quantification peak in a related substances test or an assay, the pharmacopoeial text requires 0.8 to 1.5 unless otherwise prescribed.

Newer European rules are slightly wider. EDQM announced that from 1 January 2023, with the 11th Edition, the default symmetry factor range moved from 0.8 to 1.5 to 0.8 to 1.8.

Peak symmetry categories and regulatory ranges

Tailing and fronting

Chromatography Online explains that only computer-generated Gaussian peaks are perfectly symmetrical. Real peaks always have some asymmetry, caused by extracolumn dead volume, mass overload and reversible adsorption. Excess dead volume and adsorption make peaks tail, with a wider back half. Overloading the column produces fronting, with a wider front.

Tailing and fronting

A separate troubleshooting article gives rules of thumb. Column makers typically release columns at a tailing factor between 0.9 and 1.2, a tailing factor up to 1.5 is acceptable for many applications, and at 2 or above corrective action is advisable. Tailing peaks are harder to integrate, take longer to elute, and their height limits detection.

What to do with that:

  • Tailing: check connections and tubing for dead volume, and consider secondary interactions with the column.
  • Fronting: dilute the sample or inject less, since overload is the usual cause.
  • Split or shouldered peaks: consider co-elution or a problem at the column inlet before you reach any conclusion about the sample.

Resolution, separation and the baseline

Two peaks can be far apart in time yet still overlap if they are wide. Chromatography Online draws this distinction in the same article: separation is the time gap between retention times and says nothing about width, while resolution takes width into account and indicates how pure each peak is on detection. Baseline resolution is a resolution of 1.5, where the first peak returns close to the baseline before the second begins and overlap is under 2%.

When an impurity peak sits on the tail of the main peak, resolution is low, and its area is partly lost to the main peak or mis-assigned by the integrator.

Baseline behaviour

A flat, quiet baseline makes small peaks believable. Drift or wander can come from the method itself. The LCGC Blog notes that gradients with a significant change in ionic strength can cause baseline instability, particularly with UV detection, and that careful choice of the reference wavelength and minimising ionic strength changes through the gradient help.

The EDQM update also changed how signal-to-noise is judged: it is now based on a baseline of 20 times the peak width at half height, with a minimum of 5 times if that is not obtainable. In plain terms, noise should be measured over a long enough stretch of baseline for the figure to be representative.

When reading a gradient run, remember that rising baseline and small bumps can come from the gradient itself. Comparing against a blank injection is the standard way to separate those system peaks from real sample peaks.

Checking peak purity with a diode array detector

A single symmetrical peak is not proof of a single compound. With a diode array detector (DAD), software can compare spectra across the peak. Chromatography Online describes a common approach: take a reference spectrum at the apex and evaluate how similar the spectra are across the rest of the peak.

The same article cautions that a low similarity value can indicate an impurity, but a high value does not guarantee purity. Impurities at much lower concentration, with very similar spectra, or co-eluting with the same peak profile can still go unnoticed.

For stability-indicating methods, one method validation procedure says peak purity should be demonstrated, for example by peak slicing with a photodiode array and comparison with reference and stressed material, and that no co-elution should occur on the peaks of interest.

A step-by-step routine for reading any chromatogram

  1. Read the header. Method, column, wavelength, sample ID, injection volume.
  2. Find t0 and the solvent front. Ignore peaks that belong to the system.
  3. Compare with a blank. Remove any peaks that also appear without sample.
  4. Identify the main peak against a reference standard on the same system, then confirm identity with mass spectrometry where it matters.
  5. Inspect shape. Note symmetry and any shoulders.
  6. Check resolution between the main peak and its neighbours.
  7. Check integration. Look at the baseline marks and the thresholds.
  8. Calculate area percent, and record what wavelength and which peaks it covers.
  9. Record everything with the raw data so another analyst can reproduce your reading.

Our perspective

When a researcher opens a chromatogram from a supplier, the question we hear most is whether the number can be relied on. Our answer is that the number is only as good as the trace and method behind it, which is why we put weight on 99%+ HPLC purity supported by verification processes and a Certificate of Analysis.

We would rather you read the peaks yourself than take a figure on trust. A good certificate lets you do that: a labelled trace, stated conditions and a clear link between the batch and the data. That is the standard we work to for every research peptide we supply, strictly for laboratory use.

If you want to go deeper on the certificate itself, our guide to reading a peptide Certificate of Analysis walks through each field.

Neil

Where your peptide research materials come in

If you are comparing a chromatogram against your own run, it helps to start from material with documented purity. BioForgeUK supplies research peptides with 99%+ HPLC purity, with Certificates of Analysis and a secure checkout, for scientists, institutions and laboratories.

Bioforgeuk

All our products are supplied strictly for laboratory and scientific research, not for human or veterinary consumption. You can read the terms on our research use only page, or see the full range on the BioForgeUK homepage.

Common mistakes when reading chromatograms

  • Treating retention time as an identity on its own. It is a match to a standard on one system, nothing more.
  • Quoting area percent without the wavelength. The figure changes with detection conditions.
  • Comparing different compounds by area. Response factors differ.
  • Ignoring early and late peaks. Impurities elute before and after the main peak, and a run that ends too soon can miss late-eluting material.
  • Forgetting the blank. Gradient and system peaks are easily mistaken for impurities.
  • Accepting auto-integration unchecked. Inspect the baseline placement every time.
  • Not accounting for counterions or residual salts. Peptide samples often contain them, and they affect weighing rather than the UV trace. See our notes on TFA removal and desalting peptides.

Storing and handling also affect what you see in a later run, so keep to good practice, as set out in our peptide handling guidelines for UK research labs. If you are weighing instrument options, HPLC versus UPLC explains how peak widths and run times change.

FAQ

What does each peak on an HPLC chromatogram represent?

Each peak is the detector response as one component, or sometimes several unresolved components, leaves the column. Its position along the time axis is the retention time, and its size reflects how much of the component reached the detector. Naming the peak requires comparison with a reference standard.

Is peak area or peak height better for quantitation?

Peak area is usually preferred because it captures the whole peak, including its width. Height can work for narrow, consistent, symmetrical peaks, but it misleads when peak width varies between injections. Whichever you use, apply it the same way to standards and samples.

Does a single sharp peak mean the sample is pure?

Not necessarily. Co-eluting impurities, impurities that do not absorb at the chosen wavelength, and very low-level species can all hide inside one peak. Diode array peak purity checks and an orthogonal method such as LC-MS give stronger evidence.

What is an acceptable peak symmetry?

Pharmacopoeial text requires a symmetry factor of 0.8 to 1.5 for quantification peaks in related substances tests and assays unless otherwise prescribed, and the European Pharmacopoeia default range has been 0.8 to 1.8 since 1 January 2023. Your own method or laboratory procedure may set tighter limits.

Sources