HPLC stands for high-performance liquid chromatography. It is a laboratory technique used to separate compounds within a liquid sample so that they can be detected and measured.
HPLC Explained for Non-Chemists
How to read a chromatogram, understand peak-area purity, recognize excluded peaks, and avoid being misled by an impressive-looking “99% pure” result.
```What Is HPLC?
```HPLC stands for high-performance liquid chromatography. It is a laboratory technique used to separate compounds within a liquid sample so that they can be detected and measured.
In simplified terms, a small amount of dissolved sample is injected into a flowing liquid called the mobile phase. That liquid carries the sample through a tightly packed column containing the stationary phase.
Different compounds interact with the column differently. Some move through quickly. Others remain in the column longer. This separation causes different components to reach the detector at different times.
The detector records its response as the separated components leave the column. The resulting graph is called a chromatogram.
HPLC can be used for several different purposes, including separation, impurity profiling, identification support, and quantitative analysis. What it proves depends on the method, detector, standards, calculations, and validation—not merely on the fact that an HPLC instrument was used.
```What Does a Chromatogram Show?
```A chromatogram is a graph of the detector’s response over time.
Time
Usually displayed in minutes. It shows when each detected component leaves the chromatography column.
Detector response
Shows the strength of the signal recorded by the detector. Depending on the system, this may be expressed in absorbance units, millivolts, counts, or another signal measurement.
Each visible rise and fall in detector response may appear as a peak. Ideally, one compound produces one well-separated peak. In practice, chromatograms may also contain:
- The intended target compound
- Synthesis-related impurities
- Degradation products
- Residual starting materials
- Sample-preparation artifacts
- Solvent-related signals
- System contamination or carryover
- Baseline noise
- Unidentified compounds
A chromatogram is therefore not a direct photograph of everything inside a vial. It is a record of what the selected analytical method and detector were capable of separating and detecting under the chosen conditions.
A chromatogram only shows what the method can see
Compounds that do not absorb strongly at the selected ultraviolet wavelength may generate a weak signal or no useful signal. Some nonvolatile salts, water, counterions, metals, microorganisms, and endotoxins may not be represented meaningfully in an ordinary reverse-phase HPLC purity chromatogram.
Main Peaks Versus Impurity Peaks
```On a typical peptide purity chromatogram, the largest integrated peak is often identified as the target peptide. Smaller peaks may be classified as impurities, related substances, degradation products, or unidentified components.
The main peak
The main peak is usually the peak assigned to the compound being tested. This assignment may be based on comparison with a reference standard, expected retention behavior, mass spectrometry, spectral information, or a combination of analytical evidence.
The largest peak is not automatically the correct compound. An unknown contaminant could theoretically produce the largest detector response. Proper identity confirmation requires more than simply choosing the tallest peak.
Impurity peaks
Smaller integrated peaks are commonly treated as impurities. For peptide-related samples, these may include:
- Truncated peptide sequences
- Deletion sequences
- Incomplete synthesis products
- Oxidized or deamidated variants
- Aggregation-related variants detectable by the method
- Residual protecting-group derivatives
- Degradation products created during storage or handling
- Other ultraviolet-absorbing compounds
However, one visible peak does not always equal one chemically pure substance. Two compounds may elute at nearly the same time and appear as one combined peak. This is called co-elution.
When co-elution occurs, the apparent main peak can contain more than one component. Additional tools such as photodiode-array spectral analysis, liquid chromatography–mass spectrometry, or a second chromatographic method may be needed to evaluate whether the main peak is chemically homogeneous. FDA validation guidance specifically notes that peak-purity tools may help assess whether a chromatographic peak represents more than one component.
Largest peak does not equal confirmed identity
A chromatogram may support identification, but retention behavior or peak size alone is generally not conclusive. FDA guidance states that identification based only on a single chromatographic retention time is not sufficiently specific.
What Is Retention Time?
```Retention time is the amount of time between sample injection and the point at which a particular compound is detected as it leaves the column.
For example, a COA might report that the main peak appeared at 8.42 minutes. That means the detector recorded the maximum response for that peak approximately 8.42 minutes after injection.
Why retention time is useful
When a properly prepared sample and an authenticated reference standard are analyzed under the same conditions, similar retention times can support the conclusion that the same compound may be present.
Laboratories may compare:
- The sample’s retention time
- A reference standard’s retention time
- A system-suitability standard
- A previously established relative retention time
Why retention time is not a unique fingerprint
Retention time depends on many method conditions, including:
- Column chemistry
- Column dimensions
- Particle size
- Mobile-phase composition
- Gradient program
- Flow rate
- Column temperature
- Sample solvent
- Instrument dwell volume
- Column age and condition
Two different compounds may also have similar retention times. For this reason, retention time is considered characteristic under a defined method but not necessarily unique to a single substance. USP materials likewise caution that chromatographic retention information is not conclusive by itself for impurity identification.
What Is Peak-Area Percentage?
```Chromatography software can calculate the area under each integrated peak. This value is called the peak area.
Peak area is generally more useful than peak height because it represents the detector response across the entire width of the peak rather than only at its highest point.
Basic area-percent calculation
Imagine a chromatogram with the following integrated areas:
| Peak | Integrated area | Area percentage |
|---|---|---|
| Main peak | 990,000 | 99.00% |
| Impurity A | 6,000 | 0.60% |
| Impurity B | 4,000 | 0.40% |
| Total | 1,000,000 | 100.00% |
Under this calculation, the main peak would be reported as 99.00% by area.
What that percentage actually means
A 99% peak-area result generally means:
Of the detector response assigned to the peaks included in the calculation, approximately 99% of the integrated area was attributed to the main peak.
It does not automatically mean:
- 99% of the powder’s physical weight is the target peptide
- The vial contains 99% active ingredient by mass
- The vial contains 99% of the amount stated on the label
- All impurities were detected
- The largest peak was conclusively identified
- The sample is sterile
- The sample is free of endotoxin
- The sample is free of residual solvents, metals, salts, or water
Detector response is not always equal across compounds
Area-percent calculations often assume that the target compound and the impurities produce comparable detector responses. That assumption may not always be valid.
At a particular ultraviolet wavelength, one impurity may absorb much more strongly than another. Equal physical amounts of two compounds can therefore produce different peak areas. Accurate impurity quantitation may require individual response factors, calibrated standards, or validated assumptions.
FDA analytical-validation guidance distinguishes impurity calculations expressed as area percentage from quantitative approaches expressed by weight or concentration. The reporting method and calculation procedure should therefore be disclosed.
```Baseline Noise, Drift, and Solvent Fronts
```The line running across the chromatogram between peaks is called the baseline. In an ideal system, it would be perfectly flat. Real chromatograms rarely have a completely motionless baseline.
Baseline noise
Baseline noise refers to small, irregular fluctuations in the detector signal that are not caused by a meaningful sample component.
Noise can come from:
- Electronic detector noise
- Temperature fluctuations
- Mobile-phase mixing
- Air bubbles or outgassing
- Contaminated solvents
- Dirty detector cells
- Pump pulsation
- Column bleed
- Insufficient system equilibration
Baseline noise matters because integration software must decide whether a small signal is a real peak or meaningless fluctuation. Higher noise can make it difficult to determine where a peak begins and ends, affecting calculated peak area and the ability to detect low-level impurities.
Baseline drift
Baseline drift occurs when the baseline gradually rises or falls during the run. Drift is especially common in gradient methods because the mobile-phase composition changes over time, potentially changing detector response.
Some drift may be expected and manageable. Severe drift can conceal small peaks or cause improper integration.
The solvent front
Shortly after injection, a large disturbance may appear near the beginning of the chromatogram. This is often called the solvent front, injection front, or void-volume disturbance.
It may be caused by:
- The sample solvent
- Unretained compounds
- Salts and buffer components
- Differences between the injection solvent and mobile phase
- Pressure or refractive-index disturbances
A solvent-front region is often excluded from purity integration because it may not represent meaningful retained analytes. However, the exclusion window should be scientifically justified and clearly reported.
Not every early signal is an impurity—but not every early signal should be ignored
An excluded early region may contain only an expected injection disturbance. It could also conceal unretained sample components. The chromatogram, blank injection, method, and integration report should be reviewed together.
Why Excluded Peaks Matter
```HPLC purity is calculated only from the peaks included in the integration and purity calculation.
A chromatogram can contain visible signals that are:
- Not integrated
- Integrated but excluded from the total
- Classified as solvent-related
- Ignored because they fall below a reporting threshold
- Removed through blank subtraction
- Excluded because they occur outside a selected time window
These decisions can be scientifically appropriate. They can also substantially change the reported purity percentage.
Example: how exclusions change the result
All relevant peaks included
Main peak area: 990,000
Other peak area: 20,000
Reported purity: 98.02%
Half the other area excluded
Main peak area: 990,000
Included other area: 10,000
Reported purity: 99.00%
The main peak did not change. The sample did not change. Only the denominator changed.
This is why a purity number should not be evaluated without asking:
- Which peaks were included?
- Which peaks were excluded?
- What was the integration start time?
- What was the integration stop time?
- Was the solvent front excluded?
- Was a blank chromatogram used?
- Were peaks below a certain threshold ignored?
- Were any peaks manually integrated?
Reporting thresholds
A validated method may define a reporting threshold below which very small peaks are not reported as impurities. This may be justified when the method’s noise, detection limit, quantitation limit, and intended purpose support the threshold.
The problem is not necessarily that a threshold exists. The problem occurs when the threshold, exclusion rules, or integration settings are hidden while the final percentage is presented as an absolute measurement of everything in the sample.
```Why HPLC Purity Is Not Assay
```This is one of the most important distinctions in analytical testing.
HPLC area purity
Estimates how much of the included chromatographic response belongs to the main peak compared with the total included response.
Assay or content
Measures how much of the target substance is actually present, usually by comparing the sample response with a calibrated reference standard.
A sample can be highly pure but underfilled
Suppose a vial is expected to contain 10 milligrams of a peptide but actually contains only 6 milligrams.
If nearly all ultraviolet-detectable peptide-related material in the sample is the intended peptide, the chromatogram could still report 99% area purity. The purity test describes the relative composition of the detected material—not whether the vial contains the stated quantity.
A sample can be highly pure but diluted with invisible material
Imagine a powder containing:
- 5 milligrams of highly purified peptide
- 45 milligrams of water, buffer salts, or another material that does not produce a meaningful peak under the selected HPLC conditions
The peptide-related chromatogram might still show one dominant peak and report very high area purity. That does not mean 99% of the total powder weight is peptide.
Assay requires calibration
A quantitative assay generally requires:
- An appropriate reference standard
- A known standard concentration
- Accurate sample preparation
- A validated calibration model
- Defined calculations
- Demonstrated linearity, accuracy, and precision
- Correction for standard potency or water content when applicable
The sample’s response is compared with the reference standard’s response to calculate the amount or concentration of target analyte.
Purity asks: “How much of the included chromatographic signal belongs to the main peak?”
Assay asks: “How much target compound is actually present?”
Regulatory analytical guidance treats assay and impurity testing as related but distinct analytical purposes, each requiring an appropriate validated approach.
```How Chromatogram Manipulation Can Mislead the Reader
```Chromatograms require processing. Software must determine where peaks begin, where they end, and which signals should be counted. Integration is not automatically dishonest; it is a necessary part of chromatography.
The concern is that processing choices can be altered in ways that improve the apparent result without improving the sample.
1. Changing the integration threshold
Integration software uses thresholds to distinguish peaks from baseline noise. Raising the threshold can cause smaller impurity peaks to be ignored.
When fewer impurity peaks are counted, the main peak’s area percentage increases.
2. Moving the integration start time
Starting the calculation after an early group of peaks can remove those peaks from the total. This may be appropriate for a verified solvent front, but it can be misleading when the excluded region also contains sample-related compounds.
3. Ending the run too early
Late-eluting impurities may not appear if the chromatographic run ends before they leave the column. A short run can create a cleaner-looking chromatogram while leaving strongly retained compounds unreported.
4. Cropping the chromatogram
A COA may show only a selected portion of the full run. Signals before or after the displayed window may be hidden from the reader.
5. Compressing the vertical scale
Changing the y-axis scale can make small impurity peaks appear nearly invisible. The underlying peak areas may remain unchanged, but the visual impression becomes much cleaner.
6. Expanding the main peak
Enlarging or zooming into the main peak can make it dominate the page while surrounding impurities receive little visual attention.
7. Manual baseline placement
Software calculates peak area by drawing a baseline beneath the peak. Moving this baseline changes the area assigned to the peak.
Manual integration may be necessary when automatic integration fails, but repeated or unexplained manual baseline changes can alter reported purity.
8. Combining partially separated peaks
A shoulder or neighboring impurity may be integrated as part of the main peak instead of as a separate component. This increases the main peak’s reported area.
9. Splitting a peak selectively
The reverse can also happen. An analyst may split one broad signal into multiple peaks or classify certain portions differently depending on the desired calculation.
10. Removing peaks as “unknown” or “system-related”
Labeling a peak as unknown does not automatically justify excluding it. An unknown sample-related peak may still represent an impurity.
11. Using blank subtraction without showing the blank
Blank subtraction can remove legitimate background signals from solvents, containers, or the chromatography system. However, inappropriate subtraction may also remove real sample peaks.
12. Selecting an advantageous wavelength
Ultraviolet detectors measure absorbance at selected wavelengths. A wavelength may strongly detect the intended compound while responding weakly to certain impurities—or the reverse.
Selecting an unsuitable wavelength can make some compounds effectively less visible.
13. Reporting only the best injection
Multiple injections may produce slightly different results. Showing only the cleanest chromatogram without reporting replicate performance can conceal variability, carryover, poor repeatability, or sample-preparation problems.
14. Reusing a chromatogram
A chromatogram can look legitimate while being unrelated to the product or batch listed on the COA. Readers should look for matching sample identifiers, batch numbers, file names, injection dates, laboratory information, and traceable report metadata.
Integration settings can change the answer
Chromatography data systems use parameters such as peak width, detection thresholds, baseline rules, shoulder detection, timed integration events, and manual integration. These settings directly affect which peaks are recognized and how their areas are calculated.
What a Strong HPLC Report Should Include
```A credible report should provide enough information for a qualified reviewer to understand what was tested and how the result was produced.
| Report element | Why it matters |
|---|---|
| Sample name and batch number | Connects the analytical result to a specific submitted sample. |
| Test date | Shows when the sample was analyzed. |
| Method identification | Explains which analytical procedure was used. |
| Column information | Helps define the separation conditions. |
| Mobile phases | Shows which solvents and additives carried the sample. |
| Gradient or isocratic program | Explains how solvent composition changed during the run. |
| Flow rate and temperature | These parameters affect retention and separation. |
| Detection wavelength | Shows what type of ultraviolet response was measured. |
| Injection volume | Provides context for signal strength and peak shape. |
| Full chromatogram | Allows review of the complete run rather than a cropped image. |
| Peak table | Lists retention times, areas, heights, and area percentages. |
| Integration range | Shows which portion of the chromatogram was included. |
| Excluded peaks | Explains what was removed from the purity calculation. |
| Reference-standard information | Supports identity or quantitative assay calculations. |
| System-suitability results | Shows whether the system performed acceptably during analysis. |
| Analyst and laboratory authorization | Provides accountability and traceability. |
How to Evaluate an HPLC COA Without Being Misled
```Match the batch
Confirm that the batch or lot number on the COA matches the batch identifier on the product.
Look beyond the summary
Do not rely only on a line that says “Purity: 99.7%.” Review the chromatogram and peak table.
Check the main-peak assignment
Determine how the laboratory concluded that the main peak represented the claimed compound.
Review the full time range
Look for cropping, unusually short run times, or missing early and late portions of the chromatogram.
Inspect every visible peak
Compare the chromatogram with the peak table. Visible peaks should not disappear from the calculation without explanation.
Ask what was excluded
Check whether solvent fronts, blank peaks, peaks below a threshold, or unidentified signals were excluded.
Separate purity from quantity
Look for a separate assay, content, or mass test if the stated amount per vial is important.
Check identity testing
Look for mass spectrometry or another complementary identity method rather than relying solely on retention time.
Look for method details
A percentage without meaningful method information is difficult to evaluate independently.
Verify the laboratory
Confirm that the laboratory exists and that the report contains traceable identifiers, dates, and authorization.
HPLC Cannot Answer Every Quality Question
```HPLC is valuable, but it is only one part of a broader analytical picture.
| Quality question | Is an ordinary HPLC purity test enough? | Potential additional testing |
|---|---|---|
| Is the main compound chemically pure? | Provides useful evidence, subject to method limitations | Orthogonal HPLC method, LC-MS, peak-purity analysis |
| Is it the claimed compound? | No, not conclusively by retention time alone | Mass spectrometry, reference-standard comparison, NMR |
| How much compound is in the vial? | No | Quantitative assay, amino-acid analysis, calibrated HPLC |
| Is the sample sterile? | No | Sterility testing |
| Does it contain bacterial endotoxin? | No | Bacterial endotoxin testing |
| Does it contain water? | Not reliably through ordinary area purity | Karl Fischer water analysis |
| Does it contain residual solvents? | Usually not adequately | Gas chromatography |
| Does it contain elemental impurities? | No | ICP-MS or another elemental-analysis method |
| Is the powder visually identifiable? | No | Validated chemical identity testing |
Frequently Asked Questions
```Does one large peak mean the sample is pure?
Not necessarily. The method may not detect every component, multiple compounds may co-elute, and some signals may have been excluded from integration. One large peak is useful evidence, but it must be interpreted in context.
Does a 99% HPLC result mean the vial is 99% peptide by weight?
No. It usually means approximately 99% of the included chromatographic detector area was assigned to the main peak. Water, salts, counterions, buffers, and other weakly detected or undetected materials may not be represented by the area percentage.
Can a 99% pure vial still be underfilled?
Yes. Purity and quantity are different measurements. A vial may contain less material than claimed while the material that is present produces a high-purity chromatogram.
Is retention time enough to confirm identity?
No. Matching retention time can support identity when compared under controlled conditions, but unrelated compounds can have similar retention behavior. Complementary testing such as mass spectrometry provides stronger evidence.
Why are some peaks excluded?
Peaks may be excluded because they come from the solvent, blank, injection disturbance, system contamination, or signals below an established reporting threshold. Every exclusion should have a documented scientific reason.
Can integration settings change the purity result?
Yes. Peak-detection thresholds, integration windows, baseline placement, shoulder detection, and manual integration decisions can change which peaks are counted and how their areas are calculated.
Can HPLC determine whether a sample is sterile?
No. Sterility and bacterial endotoxin require separate microbiological or biological testing procedures.
What is the difference between HPLC and LC-MS?
HPLC separates compounds and records detector responses. LC-MS combines liquid chromatography with mass spectrometry, allowing separated components to be evaluated according to their mass-to-charge characteristics. LC-MS can provide stronger identity information, although it also has limitations and requires appropriate interpretation.
A Purity Percentage Is Only as Reliable as the Method Behind It
```HPLC is one of the most useful tools available for examining the chemical composition of a sample. A properly designed method can separate a target compound from related impurities and provide meaningful evidence about relative chromatographic purity.
But the number printed at the top of a COA is not the entire result.
To understand what “99% pure” actually means, the reader must consider:
- What the detector measured
- How the main peak was identified
- Which peaks were integrated
- Which peaks were excluded
- Whether compounds may have co-eluted
- Whether the full chromatogram was shown
- Whether the method was suitable and validated
- Whether a separate quantitative assay was performed
A chromatogram can provide valuable evidence. It should not be treated as a complete guarantee of identity, quantity, sterility, safety, or overall product quality.
```