HPLC Analysis Discussion Examples: How to Interpret Retention Time, Resolution, and Calibration Curves

What Is HPLC Analysis?

HPLC stands for high-performance liquid chromatography.
A sample solution is carried into a column by the mobile phase, and components are separated by utilizing differences in the strength of their interactions with the stationary phase.
The separated components are measured by a detector and appear as peaks on the chromatogram.

Whereas GC is suitable for analyzing components that are readily vaporized, HPLC can also be used easily for heat-sensitive components, nonvolatile components, relatively large molecules, polar compounds, and similar substances.
In HPLC, estimation of components using retention time, quantitation using peak area, and evaluation of separation using resolution are important.

Main Items to Include in the Results

In HPLC results, organize the sample name, analyte, standard substances, column used, mobile phase, flow rate, detection wavelength, injection volume, retention time, peak area, resolution, calibration curve, quantitative values, and other information.
Because differences in HPLC conditions can greatly change retention time and peak shape, it is important to clearly state the measurement conditions.

Main Items to Include in the Results

• Sample name
• Analyte
• Standard substance
• Internal standard substance
• Column used
• Type of stationary phase
• Mobile-phase composition
• Mobile-phase pH
• Flow rate
• Column temperature
• Type of detector
• Detection wavelength
• Injection volume
• Retention time
• Peak area
• Peak height
• Peak width
• Resolution
• Calibration-curve equation
• Correlation coefficient
• Quantitative value of the unknown sample
• Sources of error and points for improvement

Reference Experimental Values and Calculation Examples for HPLC Analysis

Here, the process of identifying and quantifying components using retention time, peak area, resolution, and calibration curves obtained by HPLC analysis is organized using reference experimental values.

In HPLC, components in a sample are separated by a column and recorded as peaks by a detector.
Retention time provides a clue for component identification, while peak area provides an approximate indication of the amount of a component.
In addition, the concentration of an unknown sample can be determined by preparing a calibration curve from standard solutions.

Reference Experimental Conditions

HPLC Measurement Results for a Standard Mixture

A reference example is shown for measurement of a standard mixture containing caffeine, benzoic acid, and acetaminophen.

If the retention time of a peak in an unknown sample is close to the retention time of a standard substance measured under the same conditions, that component may be present.
However, identification should not be based on retention time alone, and standard addition or confirmation under different conditions should be performed when necessary.

Example of Component Identification Using Retention Time

Suppose unknown sample A was measured under the same HPLC conditions and the following peaks were obtained.

Because the retention times of the peaks in unknown sample A are close to those of the standard substances, the sample is considered highly likely to contain all three components.

Concept of Resolution

Resolution is an indicator of how well two adjacent peaks are separated.
The greater the resolution, the less likely the peaks are to overlap and the easier they are to quantify.

Resolution Rs = 2(tR2 − tR1) ÷ (w1 + w2)

Here, tR1 and tR2 are the retention times of the two peaks, and w1 and w2 are their respective peak widths.

Example Calculation of Resolution

The resolution between acetaminophen and caffeine is calculated.
The retention time of acetaminophen is 2.35 min and its peak width is 0.18 min, while the retention time of caffeine is 4.80 min and its peak width is 0.22 min.

Rs = 2(4.80 − 2.35) ÷ (0.18 + 0.22)

Rs = 4.90 ÷ 0.40 = 12.25

This value is sufficiently large, so the two peaks can be judged to be well separated.

Comparison of Resolution Between Peaks

If the resolution is low, errors are more likely to occur when reading peak areas, making accurate quantitation difficult.

Calibration Curve for Caffeine Standard Solutions

A reference example is shown in which caffeine standard solutions at multiple concentrations were measured and the relationship between concentration and peak area was examined.

In this reference example, the relationship between caffeine concentration and peak area is treated as the following calibration curve.

Peak area = 18250 × Concentration (mg/L)

Therefore, the concentration of an unknown sample is determined using the following equation.

Concentration (mg/L) = Peak area ÷ 18250

Example Calculation of Caffeine Concentration in an Unknown Sample

If the caffeine peak area of unknown sample A is 456000, the concentration is determined using the calibration curve.

Caffeine concentration = 456000 ÷ 18250 = 25.0 mg/L

Therefore, the caffeine concentration in unknown sample A is determined to be 25.0 mg/L.

Example Calculation Including Dilution Factor

If the sample was diluted before HPLC measurement, the concentration in the measured solution determined from the calibration curve is multiplied by the dilution factor to obtain the concentration in the original sample.

For diluted samples, forgetting to apply the dilution factor results in a substantial underestimation of the concentration.

Calibration Curve for Benzoic Acid and Quantitation of an Unknown Sample

A reference example is also shown in which a calibration curve was prepared for benzoic acid using standard solutions.

In this reference example, the calibration curve for benzoic acid is treated as follows.

Peak area = 21000 × Concentration (mg/L)

If the peak area of benzoic acid in the unknown sample is 315000,

Benzoic acid concentration = 315000 ÷ 21000 = 15.0 mg/L

Example Measurement Outside the Calibration-Curve Range

If the peak area of an unknown sample exceeds the range of the calibration curve, extrapolating directly may produce a large error.

Quantitation is basically performed within the linear range confirmed using the standard solutions.
High-concentration samples are appropriately diluted before measurement.

Changes in Retention Time Caused by Mobile-Phase Composition

In reversed-phase HPLC, increasing the proportion of organic solvent generally shortens the retention time of hydrophobic components.
A reference example is shown for changing the water:methanol ratio.

If too much organic solvent is added in order to shorten the analysis time, separation between peaks may become poorer.

Changes in Retention Time Caused by Flow Rate

Increasing the flow rate causes components to pass through the column more quickly, shortening the retention time.
However, an excessively high flow rate may affect separation and peak shape.

Because flow rate affects both analysis time and separation, simply increasing it is not always preferable.

Examples of Abnormal Peak Shapes

Peak shape is also important in HPLC.
If peaks broaden to the left or right or trail before or after the peak, area calculation and separation are affected.

When peak area is used for quantitation, poor peak shape makes the result more dependent on how the area is integrated.

Example Confirmation by Standard Addition

A standard substance can be added to an unknown sample and measured to confirm whether a peak in the unknown sample truly corresponds to the target component.
If it is the target component, the area of the peak at the same retention time increases.

If only the peak at the same retention time increases after addition of the standard substance, the peak is more likely to correspond to the target component.

Example of How to Write the Results

When the standard mixture was measured by HPLC, peaks were obtained at retention times of 2.35, 4.80, and 7.15 min.
Based on the measurement results for the standard substances, these peaks were considered to correspond to acetaminophen, caffeine, and benzoic acid, respectively.
Because peaks were also observed at 2.36, 4.82, and 7.12 min in unknown sample A, the sample is considered highly likely to contain these three components.

A calibration curve was prepared using caffeine standard solutions, and the relationship between peak area and concentration was expressed as “Peak area = 18250 × Concentration.”
Because the caffeine peak area of unknown sample A was 456000, the caffeine concentration was calculated as 456000 ÷ 18250 = 25.0 mg/L.

The resolution between acetaminophen and caffeine was 12.25, while that between caffeine and benzoic acid was 9.79.
Both values were sufficiently large, so the peaks were judged to be well separated.
Therefore, under these conditions, quantitation using peak areas is considered relatively reliable.

Points for Connecting the Results to the Discussion

In a discussion of HPLC analysis, it is important to explain retention time, peak area, resolution, calibration curves, and peak shape in relation to one another.

• Can retention time be compared with standard substances to provide a basis for component identification?
• Can the need for confirmation by standard addition or different conditions be explained rather than making a definitive judgment from retention time alone?
• Can a calibration curve be prepared from the relationship between peak area and concentration?
• Can the concentration be determined by substituting the peak area of the unknown sample into the calibration curve?
• If the sample was diluted before measurement, has the dilution factor been correctly taken into account?
• Can it be explained that low resolution increases errors in peak-area determination?
• Can the effects of mobile-phase composition and flow rate on retention time and separation be discussed?
• Can it be explained that peak tailing, fronting, and splitting can lead to quantitative errors?
• Can it be explained that samples outside the calibration-curve range must be diluted and remeasured?

Example Discussion

In this experiment, a standard mixture and unknown sample A were analyzed using reversed-phase HPLC.
In the standard mixture, acetaminophen, caffeine, and benzoic acid eluted at 2.35, 4.80, and 7.15 min, respectively.
Because peaks with retention times close to these were observed in unknown sample A, the same components may have been present.

For caffeine, a calibration curve was prepared from the concentrations and peak areas of the standard solutions.
Peak area was proportional to concentration, and the calibration curve was expressed as “Peak area = 18250 × Concentration.”
Substituting the caffeine peak area of 456000 for unknown sample A into this equation gave a concentration of 25.0 mg/L.
In this way, HPLC can be analyzed by combining identification based on retention time with quantitation based on peak area.

The calculated resolution between acetaminophen and caffeine was 12.25, while that between caffeine and benzoic acid was 9.79.
These values were sufficiently large, and overlap between the peaks was considered small.
Therefore, under these conditions, errors in reading peak area were considered relatively small and the separation was judged suitable for quantitation.

Changing the mobile-phase composition caused the retention time to change greatly.
Increasing the proportion of methanol weakened retention on the reversed-phase column and caused each component to elute more quickly.
However, if the proportion of organic solvent is increased too much, the spacing between peaks becomes smaller and separation may become poorer.
Therefore, it is necessary to balance shortening the analysis time with maintaining sufficient separation.

Possible sources of error include variation in injection volume, deviations in mobile-phase composition, changes in flow rate, column deterioration, and deterioration of peak shape.
In particular, if a peak tails, the quantitative value may change depending on the integration range used for the peak area.
In addition, if a high-concentration sample exceeding the calibration-curve range is measured directly, linearity may not be maintained and the error may become large, so the sample must be appropriately diluted and remeasured.

Summary

In HPLC analysis, candidate components are judged from retention time, and concentration is determined by substituting peak area into a calibration curve.
If resolution is sufficient, overlap between peaks is small and the reliability of quantitation increases.

This reference example used a standard mixture of acetaminophen, caffeine, and benzoic acid and an unknown sample to examine retention time, resolution, calibration curves, concentration calculations, mobile-phase conditions, and the effects of peak shape.
In a report, it is useful to discuss retention time, comparison with standard substances, peak area, calibration curves, dilution factor, resolution, and errors caused by measurement conditions in relation to one another.

How to Read a Chromatogram

A chromatogram obtained by HPLC is a graph with time on the horizontal axis and detector signal intensity on the vertical axis.
When a component reaches the detector, the signal increases and appears as a peak.
The position of the peak indicates retention time, while the size of the peak is related to the amount of the component.

When examining a chromatogram, check the number of peaks, retention times, peak areas, peak shapes, separation, and stability of the baseline.
If peaks overlap, errors are more likely to occur in identification and quantitation.
In addition, if a peak trails, is broad, or is asymmetric, there may be a problem with the column, mobile phase, or sample conditions.

What Is Retention Time?

Retention time is the time from injection of the sample until a component reaches the detector and appears as a peak.
In HPLC, the more strongly a component is retained by the stationary phase, the longer its retention time.
Components that are readily dissolved in the mobile phase and interact weakly with the stationary phase elute more quickly.

Retention time depends not only on the properties of the component but also on the type of column, mobile-phase composition, pH, flow rate, column temperature, and other conditions.
Therefore, when a component is estimated using retention time, the standard substance and unknown sample must be measured under the same conditions.

Causes of Changes in Retention Time

Retention time is readily affected by HPLC conditions.
If the composition of the mobile phase changes, the ease of component elution also changes.
In reversed-phase HPLC, increasing the proportion of organic solvent tends to cause many hydrophobic components to elute more quickly and shorten their retention times.

Mobile-phase pH, flow rate, column temperature, column deterioration, contamination inside the column, and differences in sample solvent also affect retention time.
If the retention time differs from that of the standard substance, slight differences in the measurement conditions must be considered.

What Is Peak Area?

Peak area is the total area of a peak on a chromatogram.
In many cases, peak area is proportional to the amount of the detected component.
In HPLC quantitation, a calibration curve is prepared from the relationship between standard-solution concentration and peak area, and the concentration of an unknown sample is determined by substituting its peak area into the calibration curve.

Correct determination of peak area requires good peak separation and appropriate baseline setting.
If peaks overlap or the baseline is unstable, errors occur in the area values.
In addition, if the detection wavelength is not suitable for the target component, the peak area becomes small and quantitative accuracy may decrease.

Difference Between Peak Height and Peak Area

Peak height represents the maximum signal value of a peak.
Peak area, on the other hand, represents the total signal from the entire peak.
If a peak is sharp and symmetrical, peak height can also serve as an approximate indicator of component amount, but if peak width or shape changes, height alone does not represent component amount accurately.

In HPLC quantitation, using peak area generally provides a more stable evaluation.
However, peak area is also affected by separation and baseline conditions, so it is important to check peak shape before analysis.

What Is Resolution?

Resolution is an indicator of how well two adjacent peaks are separated.
The higher the resolution, the less overlap there is between peaks and the greater the reliability of identification and quantitation.
If the resolution is low, it becomes difficult to determine peak areas accurately, causing errors in quantitative values.

Resolution is related to the difference in retention time between two peaks and their peak widths.
Even if the retention times are sufficiently separated, broad peaks are more likely to overlap.
Conversely, even if the retention times are close, sharp peaks are easier to separate.

Resolution Rs is evaluated from the difference in retention time and peak widths of two peaks.

Discussion When Resolution Is Low

If the resolution is low, the peak of the target component may be close to or overlapping with another component.
In this condition, identification based on retention time becomes uncertain and peak areas are difficult to determine accurately.
Quantitative values may be overestimated or underestimated.

To improve resolution, mobile-phase composition, pH, flow rate, column temperature, column type, and gradient conditions are reviewed.
In particular, the organic-solvent ratio and pH of the mobile phase greatly affect component retention and selectivity.

What Is a Calibration Curve?

A calibration curve is a graph prepared by measuring standard solutions of known concentration and plotting the relationship between concentration and peak area or peak height.
In HPLC quantitation, a calibration curve is often prepared from the peak areas of standard solutions, and the concentration of an unknown sample is determined from its peak area.

To use a calibration curve, the unknown sample must be evaluated within the concentration range of the standard solutions.
If the peak area of the unknown sample is outside the calibration-curve range, it is desirable to dilute or concentrate the sample and remeasure it.
The correlation coefficient, intercept, outliers, and accuracy of standard-solution preparation are also checked.

Discussion of Calibration-Curve Linearity

The linearity of a calibration curve is closely related to the reliability of HPLC quantitation.
If standard-solution concentration and peak area show a linear relationship, the detector response is considered proportional to concentration within that range.
The closer the correlation coefficient is to 1, the easier it is to judge the linearity as high.

However, it is insufficient to judge only from the correlation coefficient.
The concentration range of the standard solutions, variation at low concentrations, saturation of detector response at high concentrations, outliers, and the size of the intercept must also be checked.
Particularly at high concentrations, the detector may exceed its linear range and peak area may no longer be proportional to concentration.

Discussion of the External-Standard Method

The external-standard method is a method in which standard solutions and unknown samples are measured separately, and the concentration of the unknown sample is determined using a calibration curve prepared from the standard solutions.
The procedure is relatively simple and is used in many HPLC experiments.
However, it has the disadvantage of being susceptible to variation in injection volume and changes in instrument condition.

For accurate quantitation using the external-standard method, standard solutions and unknown samples must be measured under the same conditions and variation caused by injection volume and measurement order must be minimized.
Using an autosampler provides better reproducibility than manual injection.

Discussion of the Internal-Standard Method

The internal-standard method is a method in which a fixed amount of an internal-standard substance is added to the sample and standard solutions, and quantitation is performed using the ratio of the peak area of the target component to that of the internal standard.
It has the advantage of making it easier to correct for variation in injection volume and losses during sample preparation.

The internal-standard substance must not already be present in the sample, must be sufficiently separated from the target component, and must be stable under the measurement conditions.
If the internal-standard peak overlaps with the target component or another component, an accurate area ratio cannot be determined.

Effect of the Mobile Phase

The mobile phase greatly affects retention time and resolution in HPLC.
In reversed-phase HPLC, water is often combined with organic solvents such as methanol or acetonitrile.
As the proportion of organic solvent increases, hydrophobic components become more soluble in the mobile phase and their retention times may become shorter.

The pH of the mobile phase is also important.
For acidic and basic compounds, the ionization state changes with pH, which changes interactions with the stationary phase and retention time.
Therefore, it is important to keep the composition and pH of the mobile phase constant.

Effect of the Stationary Phase and Column

The stationary phase of an HPLC column is closely related to component separation.
In reversed-phase columns, hydrophobic stationary phases such as C18 and C8 are commonly used, and more hydrophobic components tend to be retained more strongly.
On the other hand, normal-phase columns, ion-exchange columns, and size-exclusion columns use different separation principles.

If a column deteriorates or becomes contaminated, retention-time shifts, increased peak width, tailing, and reduced resolution may occur.
The condition of the column directly affects the reliability of HPLC results.

Effect of Flow Rate

Flow rate is the speed at which the mobile phase moves through the column.
As the flow rate increases, components reach the detector more quickly and retention times become shorter.
However, if the flow rate is too high, components may elute before sufficient separation occurs and the resolution may decrease.

Conversely, if the flow rate is too low, the analysis time becomes longer and peaks may broaden.
In HPLC, the flow rate is set while considering the balance between analysis time and resolution.

Effect of Column Temperature

Column temperature affects retention time, peak shape, and resolution.
As temperature increases, the viscosity of the mobile phase decreases, and diffusion of components and interactions with the stationary phase also change.
In many cases, an increase in temperature tends to shorten retention time.

If the temperature is not constant, retention time fluctuates and comparison between standard substances and unknown samples becomes difficult.
Particularly when precise quantitation or retention-time comparison is performed, it is important to keep the column temperature constant.

Effect of Detection Wavelength

In HPLC using a UV detector or photodiode-array detector, selection of the detection wavelength is important.
If measurement is performed at a wavelength where the target component absorbs strongly, sensitivity increases and the peak area becomes larger.
On the other hand, measurement at a wavelength where absorption is weak produces smaller peaks and lowers quantitative accuracy.

However, if other components in the sample also absorb at the same wavelength, interfering peaks may become larger.
It is important to select a wavelength at which absorption by the target component is strong and absorption by interfering components is small.

Effect of Injection Volume

In HPLC, injection volume affects peak area and peak shape.
A larger injection volume increases peak area, but if the volume is too large, the column becomes overloaded, peaks broaden or show fronting, and resolution decreases.
If the injection volume is too small, peaks become small and more susceptible to noise.

In quantitative analysis, it is important to use the same injection volume for standard solutions and unknown samples.
Even when an autosampler is used, bubbles or precipitates in the sample may affect injection volume and peak area.

Discussion When Peaks Overlap

If peaks overlap in HPLC, multiple components may be eluting at similar retention times.
When peaks overlap, it becomes difficult to accurately determine the area of the target component and errors occur in quantitative values.
The reliability of identification based on retention time also decreases.

To improve peak overlap, the mobile-phase composition, pH, flow rate, column temperature, column type, gradient conditions, and detection wavelength are reviewed.
Adding a standard substance and checking the position at which the peak increases can also provide a clue for identification.

Discussion When Peaks Are Broad

Causes of broad peaks include reduced column efficiency, excessive injection volume, unsuitable sample solvent, inappropriate flow rate, column deterioration, diffusion, and dead volume in instrument tubing.
Broad peaks reduce resolution and also reduce the accuracy of reading peak area and retention time.

If peak width is broad, the column condition, injection volume, sample solvent, flow rate, and instrument tubing are checked.
In particular, if the sample solvent is stronger than the mobile phase, the peak may broaden or become distorted.

Discussion When Peaks Tail

The phenomenon in which a peak trails toward the back is called tailing.
Tailing may occur when the component adsorbs to active sites inside the column, the mobile-phase pH is inappropriate, the column has deteriorated, or the interaction between the sample and stationary phase is too strong.

When tailing occurs, the endpoint of the peak becomes unclear and errors occur in area calculation.
For acidic and basic compounds, adjustment of pH or use of a buffer may improve tailing.

Discussion of the Baseline

The baseline is the reference signal line in regions where no peaks are present.
The more stable the baseline, the easier it is to determine peak area accurately.
If the baseline slopes, fluctuates, drifts, or contains large noise, errors occur in peak-area calculation.

Causes of baseline disturbance in HPLC include insufficient degassing of the mobile phase, temperature changes, insufficient detector stability, changes in mobile-phase composition, substances eluting from the column, contamination, and bubbles.
Particularly in UV detection, absorption by the mobile phase and bubbles affect the baseline.

Discussion of Noise

Noise is unwanted fluctuation contained in the detector signal.
If noise is large, detection of small peaks and area calculation become difficult.
In low-concentration samples, the peak signal is small, so the relative effect of noise becomes larger and quantitative accuracy decreases.

Causes of noise include detector instability, contamination of the mobile phase, bubbles, pump pulsation, temperature changes, column contamination, and electrical noise.
If noise is large, filtration and degassing of the mobile phase, stabilization of the instrument, and checking the column and detector are necessary.

Importance of Standard Substances

In HPLC analysis, standard substances can be used to confirm retention time and prepare calibration curves.
By measuring a standard substance under the same conditions and comparing its retention time with a peak in the unknown sample, the target component can be estimated.
However, retention time alone does not provide complete identification, so standard addition or measurement under other conditions is also performed when necessary.

In quantitative analysis, the accuracy of standard-solution concentration is extremely important.
If there is an error in preparation of the standard solution, the entire calibration curve shifts, causing a systematic error in the concentration of the unknown sample.

Discussion of the Standard-Addition Method

The standard-addition method is a method in which a known amount of standard substance is added to an unknown sample and the increase in the peak is checked.
If the matrix in the sample affects detection sensitivity or retention, the standard-addition method may be effective.
It can also provide a clue for confirming whether a peak in the unknown sample is the target component.

If only the peak area at the same retention time increases when the standard substance is added, the peak is more likely to originate from the target component.
However, care must be taken not to distort the peak shape or make the concentration excessively high through addition.

Errors Caused by Sample Preparation

In HPLC analysis, sample preparation greatly affects quantitative results.
Errors in dilution factor, mistakes in standard-solution preparation, adsorption during filtration, sample decomposition, incomplete dissolution, precipitation, bubbles, and matrix effects can all be sources of error.
In particular, if the target component adsorbs to the filtration membrane, the measured concentration may become low.

In addition, if the sample solvent differs greatly from the mobile phase, peak shape may deteriorate.
In HPLC, it may be desirable to prepare the sample in the mobile phase or in a solvent having a composition close to that of the mobile phase.

Discussion of Matrix Effects

Matrix effects refer to the influence of components other than the target component in a sample on detection, separation, or quantitation.
In complex samples such as foods, pharmaceuticals, biological samples, and environmental samples, coexisting components may cause peak overlap, changes in retention time, and changes in detection sensitivity.

If matrix effects are present, the response of an unknown sample may not agree with that of a calibration curve prepared using pure standard solutions.
Countermeasures include the standard-addition method, internal-standard method, pretreatment, extraction, solid-phase extraction, and dilution.

Causes of Quantitative Values Being Too High

Causes of an HPLC quantitative value being higher than the actual value include peak overlap, errors in baseline setting, mistakes in preparation of standard-solution concentration, errors in dilution-factor calculations, absorption by coexisting components, and excessive injection volume.
Particularly in UV detection, if a component other than the target component also absorbs at the same detection wavelength, the peak area or signal may become larger.

Causes of Quantitative Values Being Too Low

Causes of an HPLC quantitative value being lower than the actual value include loss during sample preparation, adsorption to filtration membranes or containers, decomposition, incomplete dissolution, insufficient injection volume, inappropriate detection wavelength, and underestimation of peak area.
If the sample is unstable, the target component may decompose between sample preparation and measurement, reducing the peak area.

In addition, if the target component is strongly adsorbed to the column or the peak tails, area calculation may become inaccurate.
Sample-storage conditions and pretreatment conditions also affect the quantitative value.

When HPLC Analysis Can Be Judged to Have Given Good Results

HPLC analysis can be judged to have given good results when the peak of the target component appears clearly at the same retention time as the standard substance, is sufficiently separated from adjacent peaks, has a good peak shape, and the calibration curve shows linearity.
It is also important that the peak area of the unknown sample be within the calibration-curve range and that repeated measurements show reproducibility.

Agreement of retention time is important for identification, while the validity of peak area and the calibration curve, resolution, and baseline stability are important for quantitation.
The less peak overlap and tailing there is, the easier it is to judge the results as reliable.

Example Discussion When the Experiment Did Not Go Well

When HPLC analysis does not go well, possible causes are considered from results such as shifted retention time, overlapping peaks, broad peaks, tailing, disturbed baseline, a nonlinear calibration curve, or a quantitative value that deviates greatly.
Organizing the possible causes into mobile phase, column, flow rate, pH, sample preparation, injection volume, detection wavelength, and baseline makes the discussion easier.

How to Write Points for Improvement

In a discussion of HPLC analysis, including not only sources of error but also points for improvement makes the report easier to organize.
Improvements can be organized by dividing them into sample preparation, separation conditions, measurement conditions, quantitative methods, and instrument management.

Improvements to Sample Preparation

• Prepare standard solutions accurately
• Manage dilution factors accurately
• Dissolve the sample completely
• Filter the sample when necessary
• Check for adsorption to the filtration membrane
• Prevent sample decomposition
• Use a sample solvent similar to the mobile phase

Improvements to Separation Conditions

• Adjust the mobile-phase composition
• Optimize the mobile-phase pH
• Adjust the flow rate
• Keep the column temperature constant
• Consider another column
• Review the gradient conditions
• Find conditions that avoid peak overlap

Improvements to Measurement and Analysis

• Keep the injection volume constant
• Select an appropriate detection wavelength
• Degas and filter the mobile phase
• Confirm baseline stability
• Confirm the linear range of the calibration curve
• Confirm that the unknown sample is within the calibration-curve range
• Consider the internal-standard method or standard-addition method
• Perform multiple measurements and show the average value and variation

Difference Between a Superficial Discussion and a Good Discussion

In a discussion of HPLC analysis, simply writing that “a peak appeared,” “the retention times were the same,” or “a calibration curve was obtained” results in a superficial discussion.
A good discussion relates retention time, resolution, peak area, calibration curves, mobile-phase conditions, and sources of error.

Examples of Expressions That Can Be Used in Reports

The following expressions can be used when writing the results and discussion of HPLC analysis.
Adjust the necessary parts according to your own experimental results.

• Retention time indicates the time for a component to be retained in the column and reach the detector.
• Because the retention time of the peak in the unknown sample agreed with that of the standard substance, the presence of the target component was suggested.
• Retention time changes depending on mobile-phase composition, pH, flow rate, column temperature, and column condition.
• Peak area corresponds to the amount of the detected component and can be used for quantitative analysis.
• If resolution is low, signals from other components are included in the peak area and errors occur in quantitative values.
• Because the calibration curve showed linearity, quantitation using peak area was possible within this concentration range.
• If the peak area of an unknown sample is outside the calibration-curve range, the sample must be diluted or concentrated and remeasured.
• Baseline disturbance affects calculation of peak area.
• Matrix effects may cause detector response to differ between standard solutions and unknown samples.
• Using the internal-standard method or standard-addition method makes it easier to correct for variation in injection volume and matrix effects.

Points to Check When Discussing HPLC Analysis

Checking the following points before writing the report makes the discussion easier to write.

• Are the HPLC conditions clearly stated?
• Has retention time been compared with that of a standard substance?
• Has identification avoided relying on retention time alone?
• Has the resolution been checked?
• Have the effects of peak overlap been considered?
• Has the basis for using peak area for quantitation been explained?
• Have the calibration-curve equation and correlation coefficient been shown?
• Has it been confirmed that the unknown sample is within the calibration-curve range?
• Have the effects of mobile-phase composition and pH been considered?
• Have the effects of flow rate, column temperature, and detection wavelength been considered?
• Have sample preparation and matrix effects been considered?
• Do the points for improvement correspond to the sources of error?

Summary

In HPLC analysis, components in a sample are separated by a column, and retention time, peak area, resolution, and calibration curves are used for component identification and quantitation.
Retention time is the time required for a component to reach the detector and can be used to estimate components by comparison with standard substances.
Because peak area corresponds to the amount of the detected component, it can be used for quantitation with a calibration curve.

However, retention time changes depending on mobile-phase composition, pH, flow rate, column temperature, and column condition.
Peak area is also affected by peak separation, baseline, injection volume, detection wavelength, sample preparation, and matrix effects.
Therefore, in a discussion of HPLC analysis, it is important not to make a definitive identification based only on agreement of retention time, but also to check resolution, comparison with standard substances, and standard addition when necessary.

In a report, rather than simply writing that “a peak appeared” or “the concentration was determined from the calibration curve,” discuss retention time, resolution, peak area, calibration-curve linearity, mobile-phase conditions, and sources of error in relation to one another.
For accurate quantitation, it is necessary to confirm that the target peak is sufficiently separated, the unknown sample is within the calibration-curve range, and the baseline is stable.