Reading an HPLC chromatogram: a visual walkthrough
A purity percentage on a Certificate of Analysis is a single number derived from a chromatogram, but the chromatogram itself carries information the percentage cannot. Two batches with identical 99.4% purity can have very different chromatographic shapes — and the shape often tells you more about what is actually in the vial than the headline number does. This article is the visual primer. The mechanics of how HPLC works and how the percentage is calculated live in the cornerstone on what ≥99% purity actually means.
What do the axes of an HPLC chromatogram show?
The x-axis of an HPLC chromatogram is time, typically reported in minutes from the moment of sample injection. Each compound that elutes from the column reaches the detector at a characteristic moment called the retention time. The y-axis is detector response — for peptides analyzed under reverse-phase HPLC, this is almost always UV absorbance at a peptide-relevant wavelength (commonly 220 nm for the peptide bond, 280 nm for aromatic residues), reported in arbitrary absorbance units.
The plot is read left-to-right as time advances. Compounds that interact weakly with the C18 stationary phase elute first (left side of the chromatogram); compounds that interact strongly elute later (right side). The system suitability conditions defined in USP <621> — column type, mobile phase, gradient, flow rate, detection wavelength — are what make any specific retention time interpretable. Without the method parameters, the same peak shows up at different x-axis positions on different runs.
How do you identify the target peak vs. impurities?
On a clean chromatogram of a pure peptide, the target appears as the dominant peak — the tallest single peak on the plot, with substantially more area than any other peak. Smaller peaks at retention times earlier than the target are typically more polar impurities (truncated sequences, oxidation products, deletion sequences); peaks later than the target are typically more hydrophobic impurities (synthesis side products, residual protecting-group fragments). The retention-time pattern is one of the diagnostics a practiced reader uses to guess the impurity origin.
A complete COA pairs the chromatogram with an integration table listing every peak above the integration threshold — retention time, peak area, area percentage. The integration table is what the purity percentage is calculated from, and it is what makes the chromatogram cross-checkable: if the report says 99.4% but the integration table shows the target peak at only 95% of total area, something has gone wrong in the reporting. Identity confirmation by mass spectrometry on the same batch is what pairs with HPLC to confirm the dominant peak is actually the intended molecule.
What does a clean chromatogram look like?
A clean chromatogram has visual features that are quickly recognizable with practice. The target peak is tall, sharp, and roughly symmetrical — meaning the peak rises and falls at similar rates on either side of the apex. The baseline (the detector response between peaks) is flat and close to zero. Impurity peaks, where present, are well-separated from the target by clean baseline returns rather than running into the target as shoulders.
Key visual indicators of a clean run:
- Symmetric target peak. The peak rises sharply and falls sharply, with similar slopes on both sides. The asymmetry factor (a calculated metric in system-suitability data) is close to 1.0.
- Flat baseline between peaks. The detector response returns to near-zero between distinct compounds, allowing the integration software to clearly delimit each peak's start and end.
- Resolved impurity peaks. Where impurities exist, they appear as their own peaks separated from the target by a baseline return, not as shoulders or tails on the target peak.
- Reproducible retention time. Across replicate injections and across system-suitability standards, the target peak elutes within a tight retention-time window — a sign the method is performing within spec.
What are the visual flags of a problematic chromatogram?
A handful of visual patterns recur in problematic chromatograms; recognizing them is most of the chromatogram-reading skill. The table below summarizes the patterns, what each suggests about the run, and which complementary check the reader should look for next.
| Visual pattern | What it suggests | Where to check next |
|---|---|---|
| Broad, low target peak | Column degradation, sample overload, or method optimization needed | System-suitability data — was column performance within spec? |
| Tailing on the target peak | Poor peak shape from sample-column interactions or mobile-phase pH issues | Asymmetry factor in the integration table — should be near 1.0 |
| Shoulder peaks on the target | Co-eluting impurity not fully resolved from the target | Mass spec confirmation — is the shoulder a related sequence variant? |
| Multiple peaks of comparable height | Mixed sample, failed synthesis, or wrong-compound contamination | Identity by mass spec — what is each peak? |
| Drifting baseline | Detector drift, mobile-phase contamination, or temperature instability | System-suitability data and replicate injections |
| No clear target peak / chromatogram does not match COA peak | The reported purity is not derivable from the chromatogram shown | Request the actual integration table; treat the COA as suspect |
What about peaks below the integration threshold?
Modern HPLC software lets the analyst set an integration threshold — peaks with area below the threshold are not counted as impurities. The default threshold is typically around 0.1% of the total area, which corresponds to the practical resolution limit of most peptide HPLC methods. Peaks below 0.1% are often either real low-level impurities indistinguishable from baseline noise, or pure-noise events the integration software catches and (correctly) ignores.
The threshold matters because a peptide reported as 99.95% pure may simply be one whose lab set the integration threshold lower than another lab would. The methodology section of the COA should specify the integration threshold explicitly — without it, comparing percentages between labs is not perfectly apples-to-apples. This is part of why the reported number is method-bound rather than absolute. The full COA-field walkthrough is in how to read a Certificate of Analysis.
Frequently asked
- What does the y-axis on an HPLC chromatogram represent?
- Detector response — for peptides under reverse-phase HPLC, almost always UV absorbance at a peptide-relevant wavelength (commonly 220 nm for the peptide bond, 280 nm for aromatic residues). The units are typically arbitrary absorbance (mAU). Higher peaks correspond to more material reaching the detector at that moment of elution.
- How do I tell which peak is the target peptide?
- The target is almost always the tallest single peak with substantially more area than any other peak — typically integrating to 95–99.9% of total area on a clean chromatogram. The retention time of the target should match the system-suitability standard run alongside the sample, and the COA's integration table should explicitly identify the target peak by retention time. Identity confirmation via mass spectrometry is the independent check that the dominant peak is actually the intended molecule.
- What does it mean when a chromatogram shows a shoulder peak on the target?
- A shoulder peak indicates a co-eluting impurity that the method has not fully resolved from the target. Shoulders often correspond to closely related sequence variants — deletion sequences missing a small residue, deamidation products with subtle mass shifts, or partial-oxidation products. Mass spec on the same sample is what confirms what the shoulder actually is. Until that confirmation, the headline purity percentage may be slightly overstated.
- Why do some COAs include the chromatogram and others just give a number?
- The chromatogram is the evidence the test was actually performed; a number alone is a claim. Reputable analytical labs include the chromatogram and an integration table on the COA because reviewers need to be able to verify the percentage against the visible peak shape. A "report" that is just a number — particularly one with several decimal places — without the chromatogram should be treated as a marketing claim until the underlying data is produced.
- Can I tell whether a peptide is pure just from looking at the chromatogram?
- You can tell whether the chromatogram is consistent with the reported purity — the target peak should be tall, sharp, and dominant; impurity peaks should be small and well-resolved. What you cannot tell from the chromatogram alone is whether the dominant peak is the correct molecule (that needs mass spectrometry) or whether the sample is endotoxin-contaminated (that needs LAL or rFC). The chromatogram answers one question well; the full release framework needs three methods together.
Sources and further reading
- USP <621> Chromatography — pharmacopeial reference for HPLC system suitability requirements, peak resolution criteria, and reporting standards.
- USP <1086> Impurities in Drug Substances and Drug Products — guidance on classifying impurity peaks observed in chromatographic analysis.
- Snyder, Kirkland, & Dolan — Introduction to Modern Liquid Chromatography (Wiley, 3rd ed.) — the standard graduate-level reference on HPLC method development and chromatogram interpretation.