No thanks! I would like to know more about CHROMacademy

 Over 3000 E-Learning topics / 5000 Articles & Applications
 

thermo The CHROMacademy Essential Guide:
HPLC Troubleshooting Masterclass

Thursday 20th June 2013 11am EST / 4pm GMT

Dr Merlin Bicking (President, ACCTA) and Scott Fletcher (Technical manager at Crawford Scientific) demonstrate practical HPLC separation troubleshooting.  Using real world chromatograms submitted by CHROMacademy members from many different application areas, they illustrate how to recognize problems and explain further tests and investigations necessary to properly qualify the issue.  They then offer advice on how the problem may be resolved to help you gain valuable insight into
recognizing, characterizing and fixing problems in HPLC to improve your data quality and reduce instrument downtime.

Topics covered include:

  • GC-MS/MSLearn how experts troubleshoot HPLC chromatography problems
  • A series of real chromatograms are used to demonstrate logical ways to approach various issues in HPLC
  • Learn what problems and symptoms are important and how to further investigate issues
  • Learn what further tests can be done to better understand problems and fix issues more quickly
  • Learn to recognize problems quickly and achieve better data quality and higher throughput

Who Should Attend:

  • Anyone working with Liquid Chromatography
  • Anyone wishing to reduce their instrument downtime or become more effective at troubleshooting HPLC separation problems

Do you have an interesting HPLC separation or instrument problem?
Send us your chromatographic problems and have them solved by our HPLCexperts in this forthcoming Webcast - upload my HPLC chromatogram »

 

Find out more about this Month's Essential Guide Webcast »


The Camtasia Studio video content presented here requires JavaScript to be enabled and the latest version of the Adobe Flash Player. If you are using a browser with JavaScript disabled please enable it now. Otherwise, please update your version of the free Adobe Flash Player by downloading here.

 

If you have any more questions on this or any other topic - please post them on our forum >>

The CHROMacademy Essential Guide Tutorial
HPLC Troubleshooting Masterclass

Dr Merlin Bicking (President, ACCTA) and Tony Taylor (Technical Director, Crawford Scientific) demonstrate practical HPLC separation troubleshooting. 

Using real world chromatograms submitted by CHROMacademy members from many different application areas, they illustrate how to recognize problems and explain further tests and investigations necessary to properly qualify the issue. 

They then offer advice on how the problem may be resolved to help you gain valuable insight into recognizing, characterizing and fixing problems in HPLC to improve your data quality and reduce instrument downtime.

 

  ask the CHROMacademy experts

Share this tutorial

In this first problem the analysis of an α2 adrenergic agonist and its main impurity are showing peak splitting that wasn’t previous obvious!

  • Sample is an α2 adrenergic agonist and it’s main impurity
  • Fairly  hydrophobic analyte with a log P value of 2.0
  • Possess a number of basic nitrogen's with only significant pKa at 7.49

Column:    Modern C18 consisting of type B silica and end-capped – 250 x 4.6mm, 5μm

M.P. A:    Aq. Buffer pH 3.6 : MeCN, 95 : 5

M.P. B:    Aq. Buffer pH 3.6 : MeCN, 30 : 70

Gradient:

Solvent:    Mobile Phase B

Inj. Vol.:    10μL

 
 

Figure 1: Normal Analysis prior to issues.

 
 

Figure 2: Problematic analysis showing a pronounced peak should on the leading edge of the peak.

 
 

Questions which our experts consider when dealing with split peaks:

  • Are all peaks affected or just one?
  • Is this a chemical or physical issue?
  • Is the sample solvent too strong and / or the injection volume too large?
  • Is this problem observed when injecting standards or only after samples have been injected?
  • What is the sample matrix?
  • What is the sample extraction / dilution procedure?
  • Is a guard column or pre-column filter employed?
  • Once observed does the peak splitting / shouldering remain for all future injections?
  • Is this a new method or one that has been used for some time?
  • Has a fresh column been tried or the column orientation reversed?
 

The expert opinions of Dr Bicking:

  • As all peaks are split / shouldering it is unlikely to be a chemical issue and more likely to be physical problem
  • Relatively large amount of MeCN in sample solvent indicates this could be eluotropically too strong
  • Check by increasing Aq content of sample solvent or by reducing injection volume

  • The guard column or pre-column filter may have become partially blocked
  • Check by removing from the flow path.  Please do remember that contamination could have migrated onto the analytical column

  • The column inlet frit could have been partially blocked
  • Check by reversing column orientation or replacing with a fresh column

Conclusions:

  • Reducing the injection volume from 10µL to 5µL did not improve peak shape
  • Reversing the column orientation improved peak shape
  • Replacing the column with a new one improved the situation  - indicating the inlet frit on the column was partially blocked
  • This can be cleaned by reversing the orientation of the column and flushing through, at low flow rate, with a range of solvent polarities
  • Tips and Tricks
    If the problem persists either filter the samples prior to analysis or incorporate an in-line filter immediately before the column

The ‘classical’ representation of split peaks can be seen in figure 3 which shows the separation of a series of phenones

 
 

Figure 3: Peak shape issues created by a blocked inlet frit in a separation of a range of phenone test compounds.

 
 

Webcast Q&A Question

Can reversing the column damage the column?

Tony replied:
Column are packed using a constant (high) pressure rig in which a solvent based slurry of the packing material is forced through the column at high pressure with the outlet frit and fitting in place but the 'inlet' end connected to the packing rig. Whilst manufacturers do a great job in ensuring the packing is homogenous from top to bottom (inlet to outlet) there is always the possibility the material at the 'outlet' is more densely packed that the material at the inlet. So, when using a new column, one should observe the 'directionalty' of the column - i.e. one should obey the 'flow' arrow on the column. If one does not follow the direction of flow the packed bed may 'settle' in the other direction and one runs the risk of losing some efficiency.

Over time, the packed bed may 'settle' even further and, sometimes in combination with silica loss' a void may occur at the column head - often leading to split or shouldered peaks.

Turning the column around to clear a blocked frit or to overcome voiding effects should only be used a measure of last resort, however with most modern columns this should give us many injections before the column needs to be replaced. Reversing the direction as a temporary measure to clear a blocked frit should be fine and one should be able then to turn the column back and use in the direction indicated by the manufacturer unless a significant void was the original problem.


Merlin replied:

If the column is rated to 400 bar, it is usually packed at a higher pressure, so I do not observe pressure-related compression of the bed in modern columns. Twenty five years ago this was a bigger problem. The bonded phases are also much better now, so it is more difficult to chemically remove them during normal use.

However, it is still possible to damage the column if you really try! (But you should not.)

When UHPLC columns were first introduced, some manufacturers installed larger frits on the inlet side of the column, to reduce pressure. Those columns should not be reversed, as the packing may plug to frit when pushed in the wrong direction. I do not know if this practice is still common or not. Tony, do know which columns might have this configuration? I have successfully reversed Agilent columns, so they are not included in this category.

 
 

Figure 4: Peak shape recovery after column reversal to unblock an inlet frit.

 
 

Webcast Q&A Question

Is it always a safe practice to filter samples prior to LC analysis?

Tony replied:

Yes - because the benefits to the HPLC system, especially where dirty samples are being analysed, are innumerable.

A simple response but which does hold true in practice. However - there is one caveat.

One needs to ensure that the analytes do not adsorb to the surface of the filter - giving erroneous, irrepeatable results. This can be checked by running filtered and non-filtered samples and comparing analyte peak areas. If analyte adorption is suspected, one might try a different filter material to see if the problem can be overcome.

I know that this topic is one of Merlin Bickings favourites and so I will let him elaborate with his response to this question.

 

Merlin replied:

There are three critical steps for a long column lifetime:
1. Filter!
2. Filter!
3. Filter!

Yes, it is inconvenient, but the cost you spend on filters is less than the cost of replacing a new column sooner than expected.

As Tony indicated, you must verify that the filter does not retain an analytes. One way to minimize this problem is to discard the first 10 - 20 % of the sample that exits the filter, and then transfer the next portion to your vial.

The uptake varies with the filter type, solvent, and analyte. There is a very good discussion of this for some pharmaceuticals in an older article:
Carlson et al., J. Chrom. Sci., 38, 77 (2000).

 

If you do not see two peaks, but you do see severe tailing as the column gets older, can this also be caused by blocked frit? Or is this caused by the free silica which loses its endcapping??

Tony replied:
Nice question!

Tailing peaks are usually a sign that the silica is interacting with the analyte via a secondary (unwanted) mechanism of retention. You are correct that this may indeed be the interaction of surface silanol species which have lost their end capping reagent through reverse hydrolysis. A sign of an ageing column.

However, blocked frits and eluent/diluent mismatches can also give some strange effects and my advice would be to turn the column round - pump a few column volumes to waste - then hook up to your detector and make and injection. Now, if this gets you a few more analyses until a new column arrives great! At this stage you won't really know if the problem is associated with a blocked frit or an ageing stationary phase because the 'fresh' silica at the column outlet (now the inlet if you follow me) will most probably show less signs of tailing. However, what you do get is the option to carry on for a few more samples until help arrives.

 

Merlin Replied:
The cause is not important. If your peak becomes much worse over time, then you have damaged your column. Time to replace it.

Remember to observe all the normal procedures to extend column lifetime, such as :
high purity solvents
flush the system to remove buffers before storage
clean the column with a strong solvent
filter, filter, filter.

 
 

For further information on this issue:

HPLC Troubleshooting Separations Retention Time, Efficiency and Peak Shape

Peak Shape Problems

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 


Like CHROMacademy on Facebook and keep up-to-date with the latest webcast, tutorial and eLearning release schedules.

 

Facebook

Here we see an example of a different kind of peak splitting issue in this reversed phase separation of short and long lasting β2-agonists

  • Sample is a combination of short and long lasting β2-agonists
  • One is mildly hydrophobic with a log P value of 0.8 and the second is more hydrophobic with log P 2.5
  • Peaks started to exhibit splitting after a column change

Column:    Modern C18 consisting of type B silica, non-endcapped – 50 x 4.6mm, 1.8μm

M.P. A:    Aq. TFA, (0.1%)

M.P. B:    MeCN TFA, (0.1%)

Gradient:

Solvent:    Mobile Phase B

Inj. Vol.:    100μL

 

Figure 5: Normal chromatography of two β2 – agonists – note the baseline disturbance at 1.65 minutes is due to a detector wavelength change.

 
 

Figure 6: Problematic chromatography noted after a column change.

 
 

Again – here are the questions that our experts would typically ask when considering the causal factors for peak splitting phenomena:

  • What is causing the splitting / shouldering of the peaks?
  • Are all peaks affected or just one?
  • Is this a chemical or physical issue?
  • Is the sample solvent too strong and / or the injection volume too large?
  • Is this problem observed when injecting standards or only after samples have been injected?
  • What is the sample matrix?
  • What is the sample extraction / dilution procedure?
  • Is a guard column or pre-column filter employed?
  • Once observed does the peak splitting / shouldering remain for all future injections?
  • Is this a new method or one that has been used for some time?
  • Has a fresh column been tried or the column orientation reversed?
 

The expert opinions of Dr Bicking:

  • As all peaks are split / shouldering it is unlikely to be a chemical issue and more likely to be physical problem
  • Large amount of MeCN in sample solvent indicates this could be too strong
  • The injection volume is also very large (100ml)
  • Check by removing from the flow path.  Please do remember that contamination could have migrated onto the analytical column

  • The guard column or pre-column filter may have become partially blocked
  • Check by removing from the flow path.  Please do remember that contamination could have migrated onto the analytical column

  • The column inlet frit could have been partially blocked
  • Check by reversing column orientation or replacing with a fresh column

Conclusions:

  • Reversing the column orientation had no effect on improving peak shape
  • Replacing the column with a new one did not improve the situation either
  • Reducing the injection volume from 100μL to 75μL to 50μL improved peak shape and dropping to 25μL removed the split completely
  • As this affected the method  sensitivity, the sample solvent strength was reduced – 1:1 mixture of eluents A &B with 100μL injection
  • Tips and Tricks
    When the sample solvent strength is too strong, early eluting peaks are affected to a much greater extent as compared to later eluting analytes

 

Webcast Q&A Question

If you think the problem is that the injection volume is too large, why it was OK with previous column? The question said it split after change column.

Tony replied:
Great question – we were hoping you wouldn’t notice!
In reality this method is poor and is operating at the edges of failure – too large and injection, too strong a solvent as we have explained.  In reality the problem took an awfully long time to solve because with several old column batches the chromatography appeared to be ‘reasonable’ as seen in figure 5.  On further investigation by the column manufacturer (using a principal component analysis) it was discovered that a very small change in particle size distribution had led to the fallibility of the method being highlighted.  The problem was ultimately solved by reducing the volume of injection, eluortropic strength of the diluent and identification of column batches which had a specific particle size distribution.

 
 

Figure 7: Investigation into changing eluent strength and injection volume on speak splitting phenomena.

 

Figure 7 demonstrates some further investigative work to help with sample loadings and shows the compromise solution of decreasing the eluotropic strength of the sample diluent whilst also reducing injection volume.  One should note the drastic improvement in peak shape of the analyte as differential partitioning effects are reduced at the head of the analytical column.

 
 

Figure 8 demonstrates the principle nicely with some fungicide compounds.

Figure 8: Fungicide compounds (containing sulfur, halogens and weakly basic nitrogen) separation and the effects on peak shape of altering injection volume and eluotropic strength of the sample diluent.

 

Webcast Q&A Question

If you inject a large volume that is different from starting mobile phase in gradient can you get partitioning of analyte between injection slug and mobile phase?

Tony replied:

The short answer to the question is YES!

As the slug of analyte moves toward the column it will begin to partition at the eluent / diluent interface. At the head of the column, especially in gradient analysis, everything comes to a standstill. Those analyte molecules associated with the most highly eluting solvent (usually, and incorrectly, the sample diluent) will partition down the bed somewhat, giving two 'discrete' regions of analyte at the column head - those associated mainly with eluent and those associated mainly with the diluent as they entered the column - leading to a mis-shaped peak.

As we covered in the webcast - one should get the eluotropic strength of the sample diluent as low as possible and try to match the ionic strength of the eluent also (buffer strength). Where this is not possible, one should use the lowest injection volume possible in order to allow effective mixing prior to, and at the head of, the column which will help to mitigate some of the effects mentioned above.

Merlin replied:

Tony provides a very good explanation. In most modern LC systems, the volume between the injection point and the column is relatively short. Some mixing will always occur when two different solutions are moving through a tube, due to laminar flow inside the tube. However, this mixing is not complete when the injection solvent reaches the column. So, there is still a "slug" of the original solvent entering the column. Once this injection solvent is in the column, it will cause elution of whatever is dissolved in it. If the solvent is a "stronger" mobile phase, then those molecules in that region will move faster than the molecules in a weaker solution.

Mixing inside the column will dilute this slug very quickly, but the damage to peak shape is already complete, since part of the peak moved faster than the rest of the peak in the first few mm of the column. The problem is usually worse for early eluting peaks. Later eluting peaks may not be affected at all.

 
 

Further information on this subject

HPLC Troubleshooting Separations Retention Time, Efficiency and Peak Shape

Key Factors in Sample Diluent Selection for HPLC Assays of Active Pharmaceutical Ingredients

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 

In this example a coenzyme nucleotide base is causing issues with noisy baselines and disappearing peaks!

  • Sample is a coenzyme nucleotide base with three phosphate functional groups
  • Highly polar with a calculated log P value of -11.5

Column:    C18with low level surface charge – 50 x 2.1mm, 2.5μm

M.P. A:    10mM Tributylamine (aq.) : MeOH, 95 : 5

M.P. B:    Propan-2-ol, 100%

Gradient:

 
 

Figure 9: Chromatography of the co-enzyme nucleotide base (ADP+) shown – the legend indicates the type of sample (SST – system suitability test standard) and the number of samples of standards which have been analyzed at that point.

 

To summarize Figure 9 – there was no peak obvious from the test standard after two blank injections and the baseline is obviously noisy.  This continues for up to 30 blank injections, with no peak at the expected retention time in the MRM chromatogram.

After the analysis of many samples, the peak appears and whilst the shape is initially poor it improves with further sample numbers.

Having changed to a fresh batch of mobile phase – the peak disappeared again...

Problems observed

  • Changes in peak shape
  • Changes in retention times
  • Loss of sensitivity

Problems observed when

  • Addition of new mobile phase
  • Column equilibration
  • During the course of a run
 
 

Questions which our experts considered when dealing with this issue:

  • There are a large number of problems observed and causes of the observed problems – are they all interlinked?
  • Is this a new method or one that has been in operation for a decent period of time?
  • Why is propan-2-ol used as mobile phase B?
  • Why does the flow rate drop during the run?
  • Why is there no ion-pairing reagent in mobile phase B?

The expert opinions of Dr Bicking:

  • There are a few issues with this method – but a single overriding cause
  • Ion-pair methods are not designed to be operated under gradient conditions
  • The concentration of ion-pair reagent bound onto the column is at equilibrium
  • Equilibration is dependent upon the mobile phase polarity / IPR concentration / IPR type and stationary phase hydrophobicity
  • Between 50 and 100 column volumes may be required for equilibration
  • Column with low level surface charge may cause a de-stabilizing  affect on this equilibrium
  • Flow rate gradient may be necessary due to viscosity of propan-2-ol, and the potential for azeotrope formation – potentially highly irreproducible!

Conclusions:

  • All issues, irreproducible retention times, peaks shapes (and therefore sensitivity), can be traced back to the concentration of ion-pair reagent on the stationary phase
  • Change to an isocratic method and fully equilibrate OR make injections of concentrated IPR solution to accelerate the equilibration process OR increase hydrophobicity of IPR
  • May need to be repeated for each fresh batch of mobile phase
  • The column may be changed from one that contains a low level surface charge to a standard RP HPLC column or a Mixed Mode column employed which may negate the requirement for ion-pair reagents
  • Tips and Tricks
    Although not recommended, a gradient ion-pair method may be run but IPR concentration must be maintained
    (have same concentration of IPR is both solvents)

 
 

Further information on this subject

Troubleshooting your LC Chromatogram Part II Selectivity, Resolution and Baseline Issues

Translating and Transferring HPLC Method

HPLC Troubleshooting - Eluents and Solvent Delivery Systems

HPLC Troubleshooting Separations Retention Time, Efficiency and Peak Shape

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 
  • Main component is a local anaesthetic drug
  • This compound is hydrophobic with a Log P value of 4.5
  • Co-administered with a rapid acting opioid analgesic
  • This lesser component is also hydrophobic with a log P value of 3.8
 
Column:   Modern C8 consisting of type B silica and end-capped – 150 x 4.6mm, 3μm
Mobile Phase:   Heptane sulfonic acid, 60%, Methanol, 40%
Solvent:    Water
Inj. Vol.:   10μL
Flow:    1.5mL/min
λ:    255nm
 
 
Figure 10: Separation of a local anaesthetic drug and a rapid acting opioid analgesic – analyte 2 is showing peak fronting whilst analyte 1 shows poor sensitivity.
 
 

Figure 11: Zoom in from the chromatogram shown in Figure 10 with peak asymmetry values indicated.

 
 

Questions which our experts considered when dealing with this issue:

  • The large peak shows considerable fronting (As<<1)
  • The smaller peak does not show the same fronting and appears fairly symmetrical
  • Is this a new method or one that has been employed for some time?
  • Does the large peak always front?
  • What is the sample matrix?
  • What is the sample preparation procedure?
  • Are the samples ever analyzed at different concentrations?
  • What is the history of this specific column?
  • Has another column ever been used?
  • Is it possible that the column has been dropped or without and liquid passing through it?

The expert opinions of Dr Bicking:

  • Only the large peak seems to be fronting
  • Fronting peaks caused when a portion of the analyte band moves ahead of the main analyte band
  • Main cause of peak fronting is CONCENTRATION overload of the COLUMN
  • On a local level all active sites are occupied with analyte molecules therefore necessitating a portion of the analyte band to move to the next region of unsaturated surface
  • Easily tested by diluting sample or reducing injection volume
  • Also consider ‘channeling’ – a void created over the column length by pumping an air bubble under pressure
  • A portion of the analyte band can diffuse through the column with the mobile phase as it will not come into contact with stationary phase coated particles (path of least resistance)

Conclusions:

  • Reducing the injection volume alleviated the fronting  of the large peak
  • Unfortunately this meant the injected concentration of the small peak was too low for accurate quantitation
  • There are two solutions for this;
    1. Perform separate analyses for each analyte with optimized injection volumes
    2. Change the detector wavelength in order to optimize peak intensity of the minor peak – a dual / multi wavelength or photodiode array detector can monitor more than wavelength

    Tips and Tricks
    Fronting peaks are observed for column concentration overload whereas broad peaks will be seen when the column has been volume overloaded.  Detector overload is seen as flat-top peaks.

 
 

Figure 12: Effect of wavelength choice on baseline drift when operating in gradient elution mode with eluent additives which adsorb UV radiation.

 
 

A short note on related subject from Dr Bicking:

When employing a wavelength switch to improve sensitivity for analytes within a chromatographic separation or simply when selecting a suitable wavelength for your analysis – remember that the eluent solvents and additives have the ability to cause strange baseline effects, especially in gradient analysis.  In figure 12 we see the effect of the formic acid / water mixture, which absorbs at 210nm, causing a falling baseline as the amount of aqueous reduces during the gradient.  The baseline position can be affected by changes in both the absorptivity and refractive index of the eluent solution.  By choosing a higher wavelength for the analysis the baseline position shift are mitigated.

This also goes for choosing the correct reference wavelength and bandwidth in Diode Array Detection, which can help to reduce gradient baseline drift at any wavelength.

 
 

For further information on reducing gradient baseline drift:

Instrumentation of HPLC - Detectors

HPLC Troubleshooting - Autosampler, Column & Detector Issues

 

Further information on peak overloading

HPLC Troubleshooting Separations Retention Time, Efficiency and Peak Shape

Key Factors in Sample Diluent Selection for HPLC Assays of Active Pharmaceutical Ingredients

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 

In this example we look at the high and lows of baseline fluctuations….

  • Sample consists of a series of hydrophobic preservatives
  • All three compounds are of the same family and posses log P values from 1.9 – 3.1
  • A short frequency cycling baseline is observed
 
Column:   Modern C18 core-shell type column, end-capped – 150 x 4.6mm, 2.7µm
Mobile Phase:   Aqueous Buffer pH 3.2, 75% / Acetonitrile, 15% / Methanol, 10%
Solvent:    Water : Methanol, 75 : 25
Inj. Vol.:   20μL
Flow:    1.0mL/min
λ:    265nm
 

Figure 13: Undulating baseline with both long and short frequency issues.

 
 

Questions which our experts considered when dealing with this issue:

  • Analytical peaks look good but how does their retention time compare to expected?
  • Is the cycling baseline observed when making injection or all the time the pumps are pumping?
  • Does the pressure output match the detector output?
  • Has the system been maintained recently?
  • Was the system thoroughly purged to remove trapped air prior to use?
  • Has an alternative mobile phase, ideally completely different solvents, been prepared and tested?
  • Has a different column, been installed on the system and tested?
  • Are any leaks observable around the pump head or other system components (liquid pooling or white buffer precipitate)
 

The expert opinions of Dr Bicking:

  • Short frequency cycling can result from a number of system areas and each system area should be examined independently
  • Always check the easiest and most convenient (i.e. leaks!) first

  • One of the primary causes is an air bubble trapped in the pump head
  • Air can enter the pump head by
    1. Not leaving the solvent lines immersed in solvent
    2. Always leave the solvent lines in a solvent bottle

    3. Running out of mobile phase during a run
    4. Set a minimum pressure of 10 bar in your method

    5. Following maintenance of the pumphead
    6. Always purge the pumphead fully following opening

    7. A leak somewhere in the pumphead assembly
    8. ‘Blue Roll’ is a great way to check for small leaks

Conclusions:

  • It was observed that a small leak was coming from the bottom of the pumphead
  • This indicates that the seals have perished and need replacing
  • Piston seals are designed to deform around the piston and  as such wear and replacement is expected.
    Always check the pistons for scratches and pitting when the pumphead has been opened. 
    Pistons can be cleaned with solvent and even mildly abrasive pastes!

  • The piston seals were replaced and we can now relax with coffee, basking in our brilliance - having fixing our system without having to call an engineer
 
 

Figure 14: OOPS!  Perhaps we aren’t as smart as we think we are…

 
 
  • As we said - piston seals need to deform around the piston in order to create a liquid tight seal
  • Most instrument manufacturers recommend a ‘wear-in’ procedure after installation of new seals
  • Consists of flushing a solvent through the system under relatively high pressure for 40 mins – 2 hours.
  • Propan-2-ol is a good solvent to use as not only does it soften the seals to help them deform but it also increases the back pressure due to its high viscosity
  • A long length of restriction capillary (narrow I.D.) or an old ‘dead’ column can be used to help increase the pressure – typically to above 200bar
 
 

Figure 15: Good baseline after seal replacement and wear in.

 
 

A further note on this subject from Dr Bicking:

Figure 16: Deciding if the baselines fluctuations are pressure or detector related.

  • Normally, regular pulsations in the baseline indicate a pump problem
  • However, the pressure plot is stable and normal for this chromatogram
  • The problem is actually caused by a defective detector lamp

Hint: if you change the flow, or set the flow to 0, the baseline pattern would not change.  So, the problem is not in the pump.

 
 

Further information on cycling baselines

Troubleshooting your LC Chromatogram Part II Selectivity, Resolution and Baseline Issues

HPLC Troubleshooting - Eluents and Solvent Delivery Systems

Instrumentation of HPLC - Solvent Pumping Systems
 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 

In this example we consider the underlying causes of changes in separation selectivity…

  • Purity method for a mildly hydrophobic (log P = 1.0) NSAID
  • The method was recently developed and validated
  • Upon installing a new column - analyte retention and selectivity changed

Column:    Modern alkyl-phenyl consisting of type B silica and end-capped – 50 x 2.1mm, 1.9μm

M.P. A:    Aq. Buffer pH 7.6 : MeCN, 95 : 5

M.P. B:    Aq. Buffer pH 7.6 : MeCN, 5: 95

Gradient:

Solvent:    100 % Aq

Inj. Vol.:    40μL

 
Figure 17: ‘normal’ chromatography of an NSAID compound and related impurities.
 
 

Figure 18: Chromatogram shows changed selectivity after changing column to a new version of that used for method development and validation of the method.

 
 

Questions which our experts considered when dealing with this issue:

  • Is the problem chemical or physical?
  • Are all peaks affected to the same extent - physical
  • Or are some analytes affected more than others - chemical
  • Is there anything linking the affected analytes?
  • The method was recently developed and validated – was a new column used for the development?
  • If no, what was the history of the column used?
  • Were any other column batches used in validation?
  • Has the original column been re-installed – how did it perform?
  • Has the method been run on another system?
  • What is the age of the mobile phase, in particular the buffer portion?
  • Is it possible to re-measure the pH of the buffer portion?
  • Has a fresh batch of mobile phase been prepared?
  • Has the column oven been used – if available?
 

The expert opinions of Dr Bicking:

  • As only certain peaks show altered retention and selectivity this indicates it is not a hardware issue (either column or instrument)
  • If all shifting / drifting analytes were ionizable this would indicate a potential pH drift issue
  • Column temperature should always be controlled when available – ionizable analytes are affected most
  • If a method states it should be run at ‘Room Temperature’ your local procedures should state to set the column oven to 20oC or 25oC etc.

  • Where possible method development should be carried out using a new column.  If this is not possible then the full history of the column used should be known and documented
  • Some mobile phases can irreversibly modify a column

  • As a minimum, different batches of column should be investigated during method validation.
  • This can be widened to look at differing manufacturers of the same (nominal) stationary phase type
  • Aqueous mobile phases should be changed at least every week due to bacterial growth
  • The pH of mobile phases can also change over time due to ingress of CO2 for instance
  • It is also good practice to have a  spare unused column so that the column performance can be investigated

Conclusions:

  • The method had been developed using a an old column whose history was not fully known
  • No other columns had been investigated during either Intermediate Precision or Robustness validation
  • Additional columns supplied by the manufacturer generated identical selectivity as compared to the new column
  • No matter how many columns were screened, the original selectivity could not be reproduced
  • Tips and Tricks
    Ion-pair reagents, extremes of pH, highly polar and / or highly hydrophobic analytes can all irreversibly modify an HPLC column creating a unique stationary phase and surface which can never be fully emulated.
    GOLDEN RULE
     New Method - New Column!

 
A further note on this subject from Dr Bicking:

Figure 19: Deciding if the baselines fluctuations are pressure or detector related.

 
 

In this example we see retention times change in both direction –earlier eluting peaks elute earlier than the reference chromatogram (blue) and later eluting peaks move to even later retention.  There is also a notable change in peak area which is not reproducible.

These are all symptoms of a change in solvent  - i.e. the incorrect organic modifier has been used, which is affecting the portioning of different analytes to different extents, hence there is no pattern in the retention time changes.  Irreproducible changes across all of the analytes in peak area are also indicative of the changes in detector response in different solvents.

A further and final clue is in the detail of the vacancy or solvent peak caused as the solvent plug moves through the detector flow cell.  This is significantly different between the two chromatograms and this does indicate a change in the composition of the eluent or the sample diluent between the two injections.

 
 
 
 
 

Figure 20: Retention and selectivity changes due to changes in the eluent pH with acidic test compounds.

 

We have talked many times in CHROMacademy broadcasts about the susceptibility to changes in eluent pH when ionizable compounds are present in the sample, especially when the eluent pH is at or near to the pKa of the analyte molecule.

You will find a lot more on the basic detail of why this is the case in the links at the end of this section – however, let’s briefly look at why pH may change during an HPLC analytical campaign.  On standing, air may ingress into the phase if not properly capped, primarily because CO2 is acidic.  Volatile eluent additives (such as TFA) may volatilize and thus alter the mobile phase pH over time.  Differences in sample matrix or diluent composition may cause changes in pH at the head of the analytical column, which is the primary reason why we use buffers in HPLC eluent systems.

 
 

Further information on the causes of selectivity change in HPLC

Troubleshooting your LC Chromatogram Part II Selectivity, Resolution and Baseline Issues

HPLC Troubleshooting - Eluents and Solvent Delivery Systems

The Theory Of HPLC Reverse phase (partition) chromatography

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 

In this final example of this session, we investigate the reasons behind some strange retention time and baseline position shifts in the analysis of a glycoprotein dimer.

  • Glycoprotein dimer of approx. 60kDa
  • pI is 5.3 – 6.5
  • Loading – 10mg / injection
  • Obvious baseline position shifts on repeat injections

Column:   Modern C8 consisting of type B silica, endcapped – 100 x 2.1mm, 5μm

M.P. A:    Aq. TFA, (0.1%)

M.P. B:    MeCN TFA, (0.085%)

Gradient:

Detection:    214nm

Flow:    0.25 mL/min.

Temp:    40oC

 

Figure 21: Reduction in retention time and baseline position shifts in the reversed phase gradient HPLC analysis of a glycoprotein dimer – eluent compositions are marked.

 
 

Questions which our experts considered when dealing with this issue:

  • Instrument performance has been verified as being OK – we will explain in the next few slides some nice quick checks
  • Obvious retention time shifts which are additive on subsequent sample injections
  • Retention time is decreasing with subsequent injections
  • The gradients at the end of the run are to ‘flush’ the system
  • Note that the TFA in this case will be acting as an ion pair reagent – associated with analyte in solution PLUS bound to the stationary phase surface (C8 more so than C18)
 

The expert opinions of Dr Bicking:

  • This appears to be a problem with poor column equilibration and perhaps some matrix / protein binding
  • Equilibration time should be increased to 20 x column volume
    (4.6 mL) PLUS the gradient dwell volume of the system
    (needs to be determined) (10 minutes plus in total for this example)
  • The ballistic gradients at the end of the analysis have a ‘lag’ time (volume) corresponding to the gradient dwell time of the instrument and as stated the gradient segments are not given enough time to ‘complete’
  • As with any gradient - marked compositions will not correspond to the actual column mobile phase content and the very rapid gradient will not have completed when the next gradient is invoked – leading to a very confused system!
  • If the system is not properly equilibrated the ionic and eluotropic strength will be decreasing / increasing respectively with each injection – especially true for ionic equilibria
  • If the issue also involves the strong binding of analyte or matrix proteins or peptides to the column surface, this may also explain the continuously reducing retention time of the analyte.

 

Conclusions:

  • Increasing shift in baseline position indicates a systematic change in either [A] or RI also accompanied in this case by a consistently lower ionic strength due to the issues with the poor gradient equilibration
  • TFA concentration (UV cut-off of 0.1% TFA (aq) reported at 205nm – close to our 214nm)
  • Systematic retention changes with constant void marker indicate change in eluotropic strength of mobile phase
  • Recommend an increase in system equilibration time to nullify differences in system dwell volume and ensure column is fully equilibrated prior to the next analysis
  • Consider altering the system flush to reach high % organic and/or ionic strength prior to complete equilibration between injections
  • If binding of analyte or matrix is suspected the column may be ‘conditioned’ with an injection of a concentrated solution of Bovine Serum Albumin.
 
 

Further information on proper gradient equilibration

Protein Loss on HPLC Columns

HPLC Troubleshooting Separations Retention Time, Efficiency and Peak Shape

Gradient HPLC for Reversed-Phase Separations

 
 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now
 
 

Learn chromatography from the experts

Whether you work in a lab or manage a lab, you will benefit from being a member of CHROMacademy.

As a member of CHROMacademy, you will get access to our vast e-Learning archive full of great interactive content and animations.
All our Essential Guide Webcasts and tutorials and LCGCs archive of magazine articles and webcasts from your favourite authors - John Dolan, John Hinshaw, Mike Balough, and Ron Majors. Plus vendor application notes, electronic laboratory tools and calculators and with our 'Ask the Expert' function - help is always at hand.

 

Analytical Chemists

  • I feel empowered to fix things
  • I can troubleshoot effectively
  • I know where to go for help
  • I understand my analyses
  • I know where to get applications
  • I’m up to date
  • I’m more employable
  • My career is progressing
 

Laboratory Managers

  • Improved equipment utilization
  • Faster method development/problem solving
  • Flexible workforce with a common standard
  • Better quality data
  • Get up to speed quicker
  • Lower T&E
  • Less reliant on me
  • I spend less time on training
 

Subscribe for $399 per/year and access:

  • The entire e-Learning archive
  • All Essential Guide Webcasts and Tutorials
  • LCGCs archive of articles and webcasts
  • Expert troubleshooting advice when needed
subscribe now

 


Like CHROMacademy on Facebook and keep up-to-date with the latest webcast, tutorial and eLearning release schedules.

 

Facebook

loading data
loading data
loading data
loading data
loading data
Home | About UsContact Us | SubscribeTerms and Conditions | Advertise | Privacy Policy |

loading data

loading data

loading data

 

loading data


loading data

In this session, Dr Merlin Bicking and Scott Fletcher demonstrate practical HPLC separation troubleshooting. 

Dr Peter Tranchida

Merlin Bicking
President
ACCTA

Josep Miquel Serret

Scott Fletcher
Technical Manager
Crawford Scientific

Key Learning Objectives:

  • Learn from experts how to recognize problems in HPLC separations
  • Understand the steps which can be taken to better characterize problems and investigate issues
  • Get advice on how to fix or avoid problems in ‘real-world’ chromatography
  • Become more proficient in recognizing and fixing problems with HPLC separations or instrumentation