Top Troubleshooting in HPLC :How to Fix Peak Shouldering in HPLC Systems

1. Peak Issues:

- Peak shouldering, tailing, splitting, fronting, or broadening

- Loss of resolution (some/all peaks)

- Additional (ghost) peaks or negative peaks

- Unexpected chromatograms

- Variable peak height/area

2. Baseline Issues:

- Baseline drift (positive/negative/irregular)

- Cycling or erratic baseline (short/long term)

- Noise (regular/irregular) or spikes

3. Retention Time Issues:

- Decreasing/increasing retention times

- Irreproducible retention times

- Loss of retention or too much retention

- Change in selectivity

4. Pressure and Leak Issues:

- Low/high/variable pressure

- Continuous pressure changes (increasing/decreasing)

- Leaks (pump head, valves, tubing, column, detector, reservoir, degasser)

5. Sensitivity and Quantitative Issues:

- Loss of sensitivity

- Poor reproducibility or linearity

6. Carryover Issues:

- Peaks in blanks matching/not matching analyte retention times

In this series, we will break down the top troubleshooting tips for HPLC into parts.

Part1️⃣ : Peak shouldering (one or more peaks)

Possible Causes:

1. Column end frit(s) blocked/contaminated (Problem reporting cases ⭐⭐⭐⭐⭐)

Solution:

- Column end frit blockage can lead to problems with increased system back pressure but also, crucially, can affect peak shape, leading to tailing, shouldered, and split peaks depending upon the severity of the contamination.

- Peak shape issues arise from the disruption of the sample plug flow onto the packed column bed.

- Typically the problem can be solved by reverse flushing the column (disconnect from the detector first) in a solvent which will either flush away trapped particulate materials or dissolve sorbed contaminants, in which case one must choose a flushing solution in which the suspected contaminants are highly soluble.

- Filter mobile phases to remove particulate matter.

- Consider the use of an in-line filter to remove particulate materials (0.22 and 0.45 µm filter materials are readily available).

- Consider use of a guard column to remove contaminants prior to the column.

- Care should be taken that the guard column does not irreversibly adsorb analyte species or affect the efficiency of analyte peaks through the introduction of large extra column volume.

- Employ more effective sample cleanup techniques such as solid phase extraction to selectively remove unwanted sample matrix components.

2. Void at column inlet (Problem reporting cases ⭐⭐⭐⭐⭐)

Solution:

- A void is caused by mechanical collapse of the stationary phase material which, under system pressure, compresses the bed materials, leaving a void at the head of the analytical column.

- The void often results in loss of efficiency and poor peak shape (fronting, tailing, and shouldering are typical). Voids are typically related to silica collapse which occurs under high pressure or high pH conditions.

- Avoid pressure shock - increase flow (pressure) gradually (some modern pumps can be programmed to achieve this automatically) and use ‘make before break’ injection valves when possible.

- Operate at pH < 7.5 where possible, unless your stationary phase has been designed for use at high pH.

- It may be possible to temporarily fix the problem by running the column in the reversed direction and then return to the normal flow direction, however, the void will ultimately return after extended operation.

3. Large extra column dead volume (Problem reporting cases ⭐⭐⭐⭐⭐)

Solution:

- There are many aspects of the system which add dead volume, leading to loss of efficiency and/or resolution – typically these include:

- Unnecessarily long tubing with i.d. larger than required.

- Incorrect fitting type or poorly made connection.

- Use of a guard column.

- Use of unnecessary unions within tubing.

- Large volume accessories or components (flow cell, injection loop etc.).

- Use the references given here for a host of information on reducing system extra column dead volume.

4. Column contamination - analytes or sample components irreversibly bound to the stationary phase (Problem reporting cases ⭐⭐⭐)

Solution:

- Amongst the most common causes of column failure, this problem typically results in loss of efficiency accompanied often by loss of resolution.

- Peak shape effects may also occur such as peak tailing or shouldering.

- As the contaminants build up within the column, they form a modification of the original phase and separation selectivity may be affected.

- Consider protecting the column with a guard column which will preferentially scavenge the contaminating species and which can be changed more frequently and less expensively than the analytical column.

- Care should be taken that the guard column does not irreversibly adsorb analyte species or affect the efficiency of analyte peaks through the introduction of large extra column volume.

- There are many column flushing techniques depending upon the mode of chromatography and the degree/type of column contamination – a summary of some useful techniques is given here and the references also contain a host of useful suggestions for column flushing, including the removal of protein residues.

- When flushing, it is recommended that the column be disconnected from the detector to avoid contamination.

- Sometimes it is necessary to reverse the flow direction of a column to effectively/quickly clean it - this is a justifiable approach, however, remember that column flow direction indicates the way in which they are packed and pumping in the opposite direction (especially if the column contains a void at the inlet), may cause settling in the reverse direction leading to a potential reduction in column efficiency.

- Reversed Phase HPLC Flushing:

- 1st Stage– flush with 20 column volumes of 90–100% strong solvent (take care to increase the organic strength slowly to avoid buffer precipitation).

- Check column cleanliness.

- 2nd Stage (if required) – flush with 10 column volumes of the following solvents:

- 100% methanol, 100% acetonitrile, 75% acetonitrile–25% isopropanol, 100% isopropanol, 100% methylene chloride, and 100% hexane.

- Return to the original mobile phase via isopropanol (20 volumes).

- Check column cleanliness.

- 3rd Stage (if required) – flush with 20 column volumes of 0.05 M EDTA, followed by 20 column volumes of 95% water-5% acetonitrile.

- Check column cleanliness.

- Column volumes for some typical column geometries:

- Column size and void volume: 250 x 4.6 mm - 2.5(mL), 150 x 4.6 mm - 1.5 mL, 150 x 3.0 mm - 0.64 mL, 150 x 2.1 mm - 0.28 mL, 50 x 4.6 mm - 0.50 mL, 30 x 4.6 mm - 0.30 mL, 15 x 4.6 mm - 0.15 mL.

5. In-line filter or guard column blocked/contaminated (Problem reporting cases ⭐⭐⭐)

Solution:

- This issue can lead to increased system backpressure, but also to degraded separation characteristics such as loss of efficiency or poor peak shape.

- Change the filter or guard column as appropriate.

- Filter mobile phases to remove particulate matter.

- Consider the use of an in-line filter to remove particulate materials (0.22 and 0.45 µm filter materials are readily available).

- Consider the use of different packing material to alter matrix component removal capability.

6. Sample diluent mismatched with eluent (Problem reporting cases ⭐⭐⭐)

Solution:

- Diluent solution should contain the same or less of the strong solvent than the mobile phase.

- If analyte solubility is the issue, try making a concentrated solution in the strong solvent and dilute to the required concentration with the eluent.

- Adjust the injection volume to minimize effects – use less than 10% of loop volume when the diluent is the strong solvent, less than 25% of the loop volume when the diluent is ‘stronger’ (contains a greater proportion of the strong solvent) than the eluent.

- Wherever possible the mobile phase should be used as the sample diluent.

- pH and buffer strength differences between the eluent and diluent may also affect peak shape.

7. Incorrect/non-optimal mobile phase pH (Problem reporting cases ⭐⭐⭐)

Solution:

- Mobile phase pH affects retention of ionizable analytes and the selectivity of separations in reversed phase HPLC.

- Some methods require a very precise pH in order to remain robust.

- Low pH tends to increase retention of acidic analytes and decrease retention of basic analytes.

- High pH tends to decrease retention of acidic species and increase retention of basic species.

- Low pH (< 3) tends to improve peak tailing, especially for basic compounds due to the decrease in ionization of acidic silanol species on the stationary phase surface.

- Buffers are used to prevent pH changes, and it is important that the correct buffer system is used in order to prevent pH shifts (usually due to the introduction of the analyte in a diluent which does not match the mobile phase pH).

- pH should be measured using a calibrated pH meter on the aqueous portion of the mobile phase only and accuracy to within ±0.05 pH units is recommended.

- The actual mobile phase pH may alter significantly (typically 1 to 1.5 pH units closer to neutral) in the presence of larger amounts of organic modifier.

- Mobile phase pH may alter on standing through the ingress of CO2 (making the mobile phase more acidic) and via evaporation of the organic portion when pre-mixed mobile phases are used.

- The sensitivity of some detection systems (such as atmospheric pressure MS and fluorescence detection) may be affected by mobile phase pH changes which affect the degree of analyte ionization.

8. Inadequate/inappropriate sample preparation (Problem reporting cases⭐⭐)

Solution:

- Proper sample preparation is crucial for successful HPLC and inattention to this aspect of the analysis can lead to a multitude of retention time and quantitative reproducibility issues.

- Instrument problems can also result from poor sample preparation, such as blockages, high back pressure, and shortened column lifetimes.

- Aspects of sample preparation which need to be considered include: sample diluent strength and solubility, sample concentration, possibility of interfering species within the sample matrix or from other analytes, sample diluent pH and buffer strength, requirements for filtration if particulate matter is present.

- Use the references to explore options for sample preparation and appropriate sample preparation techniques.

9. Silica degradation through high pH use (Problem reporting cases ⭐⭐)

Solution:

- At high pH, the silica matrix itself becomes susceptible to hydrolysis and, therefore, to chemical (and subsequent mechanical) breakdown of the support material.

- As the silica matrix continues to break down, fines are formed which can often migrate to the column outlet frit and cause back pressure increases.

- Consider lowering the mobile phase pH to below 7.5 when using traditional silica based columns.

- Be aware that above temperatures of around 60 °C the upper pH limit for traditional silica support materials is 6.5.

- Many columns are available which can operate at elevated pH – use the references to further investigate the properties required for high pH operation – be aware that selectivity may alter when changing phase type, even though the bonded phase may be nominally identical.

- A measure of last resort would be the use of a silica saturation column – a sacrificial column placed prior to the analytical column which degrades in preference to the analytical column and works by saturating the eluent passing through the column with silica.

10. Two or more poorly resolved compounds per peak (Problem reporting cases ⭐⭐)

Solution:

- Many manifestations of poor peak shape such as peak tailing, peak splitting, peak shouldering, and peak broadening may, in some instances, be due to the co-elution of two peaks.

- The relative size of each peak will affect the appearance of the peak.

- Determine if all peaks in the run are split – this may indicate a blocked column frit or void at the head of the column.

- If only one or two peaks show the splitting behavior then a separation issue should be suspected.

- Inject a different (usually lower) concentration in a solvent which more closely matches the elution strength of the mobile phase to investigate solvent effects.

- If the ratio of the main peak to shoulder height changes on injecting a lower mass, the presence of two separate peaks is indicated.

- Consider using a different stationary phase to investigate an orthogonal selectivity.

- Check possible secondary causes such as a blocked guard column or incorrectly made/degraded mobile phase.

11. Channeling in column packing (Problem reporting cases⭐)

Solution:

- Channeling is the phenomenon which occurs when large areas of packing material (typically running longitudinally through the column), are created when agglomerated air bubbles are forced through the bed, destroying the silica particles in their path.

- Channels within the column have limited sample capacity and the stationary phase surface quickly becomes (locally) saturated with analyte, causing the remaining analyte molecules to run ahead of the main band, ultimately resulting in peak fronting.

- To avoid this issue, store columns with end caps in place and slowly ramp column flow (pressure) when priming the column after installation.

- Ensure solvents are properly degassed.

12. Sample degraded (Problem reporting cases ⭐)

Solution:

- Sample degradation typically occurs due to standing in the diluents over a protracted time or via thermal degradation on standing in the autosampler.

- Analyze samples as soon as possible after preparation where degradation is possible.

- Employ a thermostatic autosampler where thermal degradation is possible.

13. Inconsistent sample volume injected (Problem reporting cases ⭐)

Solution:

- There are a variety of reasons for inconsistent sample injection amount which include: cavitation of the sample solution during aspiration due to overfilled vial, incorrect loop size for sample volume injected, blockages in the sample needle, leaking sample valve, cross-port scratches on the injection valve rotor seal causing sample loss during sample loop filling and injection into the mobile phase, leaks around high pressure seals.

- Running an autosampler linearity test (see references) is a good way to identify this problem and get an idea of what may be going

wrong with injection precision.

- Use an injection volume that is optimal for your system.

Next Part Additional (ghost) peaks or negative peaks ..