Peak elution time difference

[Peptidefanatics]

We have been unsuccessful in trying to transfer this method from one site to another. The peak elution time differs by 5 minutes, even though it's the same method, same buffer, same column and same product. The only difference is the manufacturer of the HPLCs.

Does anyone have an idea how to appproach this delimma? We think this might be related to the dwell volume. Does anyone have a clue how to approach this...?

[H2Oh]

Two things you might want to check...

First are the pressures similar between the two instruments? If not, that might contribute to the problem.

Also, check the tubing lengths and diameters (more lengths I think). If the lengths are different this surely will translate into a different retention time. Generally speaking shorter lengths are better all around unless using a post column reaction/derivitization. Your lengths after the injection valve to the detector should match between the two systems in each portion.

It's a quick mechanical check that might be the problem.

[Bruce Hamilton]

Assuming a flow rate of more than 0.2 ml/min, then 5 minutes seems way to long for dwell volume. You need to provide more information on the method ( gradient/isocratic, mobile phase, column, retention time, etc )

The fastest solution is to send the column and mobile phase from one site to another and repeat the separation. Assuming all else is correct, one of the columns is stuffed.

Provided the hardware is working correctly ( especially any mixing and individual channel flow rates ), such a large difference is most likely on an isocratic method that has either the mobile phase prepared incorrectly or the column and/or solvent temperatures different.

There is also the possibility that the method is faulty, eg the compound's chromatography is pH sensitive, but here is insufficient buffer capacity in the mobile phase, so the peak retention time is drifting in the wind.

My guesses:-
1. A column is stuffed ( compare preesure at both sites at several times during a run, especially if gradient, allowing for initial instrument differences ).
2. Pumps are not working properly at one site ( if isocratic, preblend mobile phase at each site, check any mixing valve. If gradient, check individual channels )
3. Mobile phase has no buffer capacity, or has been prepared differently.
4. Ambient or column temperature are different ( more important for isocratic than gradient ).
5. Somebody is dissolving the sample differently than specified.

Bruce Hamilton

[rhaefe]

If, as you say, everything is the same except the HPLC systems and if both of them work correctly and you are using a gradient separation then the cause of the problem might very well be the different dwell volumes in your systems.
Most manufacturers provide a SOP explaining to measure gradient accuracy. In short it involves programming a step gradient:
eluent A: 100% DI Water
eluent B: DI water with 0.1% acetone

Run a step gradient from 100% A (keep there 10 min), 90%A, 50% A, 10% A and back. Don't use a column but install a restriction capillary. From the steps in the chromatogram (UV detection at 254nm) you can determine gradient accuracy and dwell volume.

cheers

[H2Oh]

If you have multiple peaks a quick way to check if the dwell time is playing a role in the discrepency is to time the peaks after the water dip (assuming you have one).

If peak A comes out 2 minutes after the dip and peak B comes out 8 minutes after dip in the first system, compare these to the retention times relative to the water dip in the second system. If the peaks comes out, say 7 and 13 minutes it's likely a tubing length/dwell volume issue. For that matter, you could maybe just time the water dip differences. Are these also 5 minutes apart?

If the peaks in the second system come out at seemingly random points, 7 minutes for A and then 10 minutes for B something more elaborate is causing the problem, such as the pH, temp, proportioning valve, etc. mentioned above.

[Dan]

More information is needed.

Is the method isocratic? or gradient? Are the HPLCs of the same type? (both low-pressure mixing or both high pressure, etc?) Is the mobile phase pre-mixed or mixed on-line? etc...?

As Bruce noted, 5 minutes is a bit large to attribute to dwell volume for an isocratic method.

However, a 5 minute difference is possible for a gradient method, where the dwell volume can have a great affect. This is especially true if you are comparing low-pressure to high-pressure HPLC systems. High-pressure HPLCs are not common, but they are still in use. For the two system types, gradients need to be adjusted based upon the dwell volume difference.

So, please provide more info for us forum members to help troubleshoot. Otherwise, we just can speculate and postulate.

Regards,
Dan

[Uwe Neue]

Considering that the method was posted by "Peptidefanatics", it is a peptide method, thus a gradient method. The method may be running on a 2 mm column, and the flow rate could be 0.2 mL/min. Thus the difference in the gradient delay volume between the two instruments is in the order of 1 mL, not unusual.

The simplest way to fix this is to change the procedure to specify retention times relative to a standard.

Another way to fix this is to play with a delayed injection (if the instrument with the longer retention time lets you do that).