Peak Fronting (Co elution) Troubleshooting

[NO89]

Hi all,

I am currently running an HPLC method for the analysis of carbohydrate deficient transferrin (CDT) that has been validated for use on a Acuity UPLC H-Class system.

I have now been running this method for almost a year, but all of a sudden I am now experiencing peak fronting on only a single peak.

I am running blank and test solutions prior to analysis and they go through fine, no issues what so ever, but when I inject the control sample this is when I experience the issue. So I am only seeing this peak fronting on my control samples and only on the one peak, the other peaks in the chromatography are fine.

I have used a competitors control and see the same issue, I have changed buffer LOT numbers, changed the pre filter, preped fresh control, used new HPLC grade water to reconstitute the control and still seeing this issue.

I was wondering if any of you could have any further information on how i can further approach this to try and resolve this issue.

Many Thanks

[danko]

Do you see the same peak fronting when you inject standards and unknown samples?
Do you inject the same amount and the same volume as previously?
Have you changed the diluent – f. ex. Organic content or pH?
Is the peak an early eluting one – what is the retention time?

Best Regards

[NO89]

Thank you for your reply,

The peak fronting is not seen on any of the test solution injection or unknown samples. The fronting is only seen on the one specific peak (tetra) and only in control one.

When the fronting gets rather bad and begins to look like co elution, it then seems to be affecting control 2, which is injected directly after control 1.

And yes, the injection volumes have not changed.

All the diluents and buffers used in this process are pre made by the company from which we buy the products from. I have carried out several runs now using different LOT numbers of each, incase it could have been a contamination that was causing the issue, and i even changed the column.

With regards to the retention time, the test solution produces the Tetra peak at 6.3 minutes, with control 1 producing the tetra peak at 2.63 minutes.

I hope I've provided the information to answer your questions, and again thank for replying.

Im really at a dead end with this at the moment.

[danko]

I’m not sure I understand the entire concept here but I’m in doubt that the method of analysis you are working with is straight forward. I’m sure that 99.9% of the members of this board will agree with me that the retention time of a certain compound in the control sample should be at least very similar (roughly identical) to that of the same compound in the standard/reference sample and not least in the unknowns.
So here you’re dealing with a different elution profiles and the elution of your control is not fully thought through
Please shed a light on the matter if you think I’ve misunderstood something. Then maybe we’ll be able to find a solution to your problem.

Best Regards

[NO89]

Sorry about that, it would be much easier for me to explain if i was able to post the images of the chromatograms on here for you to see.

With regards to the method:

It is a gradient method, using pre made (by RECIPE) Buffers A, B and C for CDT separations.
The CDT peaks elute whilst buffer A is ran through, whilst buffer B and C clear the system through and preparing it for the next injection.

The sample set consists of a test solution injection to ensure that the system is working as we would expect it to be, followed by one injection of control one and one injection of control two and finally the unknown samples.

The controls are prepared in exactly the same way as the unknown samples, from start to finish.

My main concern, and confusion is as to what could possibly be causing the peak fronting to only happen in one sample (control one) and only affect the one peak.

If there was fronting on all peaks, or in all samples then id be more inclined to believe that there is an issue with the system or the actual set up, but unfortunately that is not the case as it seems.

Again, thank you for your help. I have tried contacting the suppliers of the system and product, but not had much luck with them at the moment, so your help here is very much appreciated.

[danko]

OK, gradient….
Maybe you have an equilibration issue.
Please alter the sequence so that another sample is injected instead of the first control.
Also, try injecting a blank (0 µL) injection right after the “famous” control 1.
Then we’ll take it from there.

Best Regards

[danko]

I forgot another yet test: Inject a blank before the control 1.
Btw. There is a possibility of posting a chromatogram here but it’s not a 2 seconds procedure.
Please follow the link below:

http://chromforum.org/viewtopic.php?f=1&t=2617

Best Regards

[NO89]

Here is a chromatogram of the test solution:




here is a chromatogram of the Control 1, with the fronting in the tetra peak:





I hope these work.


Our system is now currently being set up to run a different analytical test so I will not be able to carry out your suggest troubleshooting advice until next week unfortunately.

I will keep you updated.

[danko]

It looks like ”sample solvent/mobile phase” inconsistency – perhaps the control sample’s pH is higher or lower than the mobile phase’ and the rest of the samples.
You can also see that the peaks if the control sample elute slightly earlier than those of the test sample.
Maybe you’d like to post some more info for use in the further trouble-shooting process.
1. What is the mobile phase pH?
2. What is the pH of the rinsing eluents.
3. What is the pH of the sample solutions (both the control and the test)
4. How does the gradient table look like?
5. How long is each run?
6. How much volume do you load/inject?

Best Reagrds

[Klaus I.]

I am currently running an HPLC method for the analysis of carbohydrate deficient transferrin (CDT) that has been validated for use on a Acuity UPLC H-Class system.

Please add information about the used systems.

[Hollow]

...and about the column and gradient table

you've said the peak is eluted whilst A is running through and B&C are used for column cleaning. So in fact the separation is in isocratic mode?

And what about the detection? Any possibility to spectroscopically (PDA,MS) check the front shoulder if there are any differencies to the peak apex?

[M Farooq]

Your "tetra" peak is not fronting at all but there is a co-eluting minor component, the so-called shoulder in your component. If possible, do a LC-MS of this shoulder, other simply alter the mobile phase (make it slightly weaker) so that the minor component is separated from this "tetra" peak.

When a peak fronts in the true sense (due to departure from the linearity of adsorption isotherms), it always fronts. Your standard shows a perfectly nice peak.

here is a chromatogram of the Control 1, with the fronting in the tetra peak:

[danko]

Farooq, it is fronting! Remember that right after the 1. Control another one is injected (control 2) which is just a replicate. That control (2) doesn't exhibit the same fronting. Ergo a mobile phase composition inconsistancy is most probably the issue. More thorogh review of the method parameters including the gradiet table is the way forward.

Best regasrds

[NO89]

Firstly, thank you all for taking your time to contribute to this discussion, I very much appreciate the time you have taken to respond.

I will try and answer / provide more information as I go along.

Right with regards to

Farooq, it is fronting! Remember that right after the 1. Control another one is injected (control 2) which is just a replicate. That control (2) doesn't exhibit the same fronting. Ergo a mobile phase composition inconsistancy is most probably the issue. More thorough review of the method parameters including the gradiet table is the way forward.

Control 2 inst exactly a replicate. Control 2 has a higher percentage of disialotransferrin, so in a sense control 1 is like a negative control where as control 2 is a positive control.

Farooq,

If possible, do a LC-MS of this shoulder, other simply alter the mobile phase (make it slightly weaker) so that the minor component is separated from this "tetra" peak.

Unfortunately here in the lab we do not have the equipment to perform MS, but this is something i will suggest to my superiors. As for weakening the buffer, this is a pre made buffer, it comes in already made, with no information regarding its pH or anything, however I will contact the suppliers and see if they can provide me with any further information.

The fronting / co eluting gets worse as the sample set progresses (the control 1 injection at the end of the analytical run is much worse that the control 1 injection at the beginning of the sample set).. please see attached image of the control 1 injection at the end of the analytical run:


Klaus I, could you please specify which type oif information you would like regarding the system? It uses a Quaternary solvent manager, with a pre column heater in which the analytes are detected via UV PDA detector. Current running Empower Pro software.

It looks like ”sample solvent/mobile phase” inconsistency – perhaps the control sample’s pH is higher or lower than the mobile phase’ and the rest of the samples.
You can also see that the peaks if the control sample elute slightly earlier than those of the test sample.
Maybe you’d like to post some more info for use in the further trouble-shooting process.
1. What is the mobile phase pH?
2. What is the pH of the rinsing eluents.
3. What is the pH of the sample solutions (both the control and the test)
4. How does the gradient table look like?
5. How long is each run?
6. How much volume do you load/inject?

After analysing yesterdays analytical run this fronting/co elution issue is now affecting my control 2, but this could be due to carry over from control one.
I do not know what the pH for the mobile phase is or the controls as this is not disclosed by the company that provides the buffer, but I will contact them to try and gather this information.

The gradient table is as follows : Start to 10 minutes is 100% Buffer A, from 10 to 10.1 minutes it is 60% buffer A and 40% buffer B. from 10.1 minutes to 14 minutes it is 100% buffer B. from 14.1 to 20 minutes it is buffer C.

The flow rate is 0.8 ml/min which buffer A runs through and 1.5ml/min when the other buffers are running through.

Each run is 20 minutes long, however we only use the first ~10 minutes of data.

The sample volume that we inject is 250ul.

Hollow,

you've said the peak is eluted whilst A is running through and B&C are used for column cleaning. So in fact the separation is in isocratic mode?

Yes that would be correct, when the peaks are eluting this only happens when A is running through.

And what about the detection? Any possibility to spectroscopically (PDA,MS) check the front shoulder if there are any differencies to the peak apex?

Again, Unfortunately this is something we do not have readily available here, it is something I would have to bring up with my superior.


Again, thank you all for your input. I will not be back in lab now until Tuesday, so apologies in advance if i can not respond to any more of you question until then.

[Hollow]

Still missing column information (dimensions, type)

you're saing that you have a PDA detector, so if not yet done, activate the 3D data collection in the instrument method. Then you'll have some spectoscopic information for a first check (don't know how distinctiv the UV spectra of your compounds will be).

you don't state anything about the reequilibration with eluent A after C? how does the gradient table exactly looks like? Is ther a "next injection delay" programmed in the sample set?
If I'm correct, the H-class has a delay volume of about 700ul, together with the t0 from your chromatogramm, I would like to have running A through the system for about 4 min before injecting the next sample. (But this would affect all inj...)

If this isn't a typo, 250ul is quite large (ca. 50% of your column volume). many effects can occure under such conditions.

Regarding the commercial buffer, have a look at its MSDS, will possibly give some hints about the composition.