Baseline flactuations using Ion Pairing Reagent

[Omar]

Hi all,

I am running a gradient method on a C18 column 250 x 4.6mm, 5um using a phosphate buffer containg hexane sulfonic acid 1.88g/L pH adjusted to 2.5, with an increasing concentration of acetonitrile with UV detection. As you can see from the embedded picture, (if it works!) I am having some irregular and irreprodicible baseline flactuations which are not acceptable for my application. At the moment I am not considering a detector problem as this is a brand new detector which has been used for only about for 1 month and has always worked perfectly.

I consider myself new to the use of ion paring reagents and since i've never seen such flactuations in normal separations my question is if this could be due to the hexane sulfonic acid since furthermore it is being used in a gradient? Can you please give any suggestions on how can reduce the occurance of such flactuations?

PS: In the above image I consider the bottom chromatogram as representatinve of how it should really look.

Thanks and Regards

[Kostas Petritis]

Can you give some more details on the UV wavelenght used, exact gradient conditions and equilibration time?

[Omar]

The wavelength is 238nm
The Gradient programme is as follows:

flow: 1.2ml/min

Line A: Buffer
Line B: ACN
Line C: MeOH

00mins: A:80 B:15 C:5
60mins: A:60 B:35 C:5
65mins: A:80 B:15 C:5
80mins: A:80 B:15 C:5

[zokitano]

Dear Omar,

As I can see the scaling of the y-axis is very detailed. From one number to another (on the y-axis, absorbance axis) there is a difference of 0.2mAU, which is too small. Therefore, the change in the mobile phase composition during the gradient is clearly seen (the increase of the baseline).
Try to view the chromatogram in less detailed fashion. If the peaks of interest are small, then you may increase the injection volume or better increase the initial concentration of the injected sample.

Also there are many peaks in your chromatogram. If all these small peaks are peaks of interest ok, but if there are not then you may have issues with the purity of the reagents you're using.

Regards

[Omar]

Hi zokitano,

The test is a chromatographic purity and so I am interested in all the peaks in the chromatogram. I have worked with such detail with other methods and I have never seen such flactuations which are so irreproducible. I don't think its a problem of reagent purity as apart from the fact that I am using chromatographic grade reagents, I would expect that impure reagents would produce interference but this should be somewhat reproducible. Furthermore the flactuations I am observing are not peak shaped as I would expect with impurities but rather square shaped involving sudden increases and decreases in absorbance.

[zokitano]

Omar,

This seems to me like a regular case. During the gradient program, an increase of the ACN content and decrease of the buffer content in the mobile phase occurs. Therefore, the molar absorption coefficient of the mobile phase changes during the gradient program. So the background absorption changes too, and that's why you're seeing "irregular" baseline.
If I may add, the baseline rise and fall is the same, for me, for all chromatograms. And I don't see any irreproducibility in this case. I must say that this is something to be expected. The gradient change of the composition of the mobile phase in the first 60 minutes corresponds with the slight rise of the baseline in chromatograms. The rapid change of the composition of the mobile phase after the 60th minute back to the initial composition, makes the fall of the baseline (as a result of the rapid change of its absorption coefficient).

Regards

[Omar]

Hi Zokitano,

I know what you are talking about but that is not my issue. The above pic is a portion from the above. What I was saying is that these chromatograms represent multiple injections of the same solution and the factuations that I am talking about are the ones shown in this pic. These are problematic to me because if I integrate these flactuations as peaks, they are too big of the order of about 10,000 µV*sec (Area units).

[zokitano]

Dear Omar,

Did you try to equilibrate the column with the initial mobile phase composition longer, prior the subsequent injection. Due to the presence of the ion-pair reagent in your mobile phase it is possible that the required equilibration time after a gradient is longer than in the regular RP separations.

[danko]

The problem Omar experiences, is not fluctuations but irrelevant peaks.
They are most probably caused by impurities in the mobile phase (i.e. eluent A).
It could be the water or the added chemicals (e.g. the ion pairing reagent).
Very similar topics have been discussed a number of times in this forum.
So if you search (here in the forum) for “peaks + gradientâ€

[dsduncan]

I agree with zokitano. Your fluctuations may be due to inadequate column equilibration. When doing IPC with a gradient I've found it best to prepare MPA and MPB which contain the same concentration of ion pair reagent. It takes a bit longer to prepare mobile phase but it will help to re-equilibrate the column much faster, and you will avoid the issues that may occur with on-line mixing of organic solvent and buffers. Once you have a consistent chromatogram from injection to injection you can then troubleshoot any ghost peaks that still exist as discussed by Danko.

[Kostas Petritis]

Hi Omar,

Without entering in too many details, when using ion-pairing chromatography it is a good idea to keep the mobile phase as simple as possible as each component of the mobile phase (under certain circumstances) can create additional peaks when the system is perturbed (i.e. from an injection).

Let me ask you this: After cleaning your column and equilibrating it, does your first or couple of injections look good and then you start getting peaks at different places? If yes, it means that you get additional peaks (system/ghosts peaks) later. You can verify this by injecting (the higher the amount the higher the system peaks) and let your system go through several consequtive gradients where you do not inject anything...

Your choices? Either allow the column to equilibrate enough so that all the post injection system peaks have been eluted (you can see how much you need to wait with the above experiment). You can also try to eliminate MeOH and phosphate buffer in order to see if the system peaks decrease or completely eliminated.

[Uwe Neue]

Some of these baseline disturbances look more like soap bubbles or air bubbles to me. My suggestion is to look at your degassing protocol.

[Bruce Hamilton]

The 15 minute equilibrium time at the end of each run should be adequate, but I would definitely premix your solvents (80:15:5, and 60:35:5 ), and place them in A and B after degassing, and go from 0-100 A-B.

Because hexane sulfonic acid is very surface active, also ensure that any instrument degasser has been thoroughly flushed with a highly aqueous solvent before adding the components.

Uwe is correct, the variable baseline shifts seem to be from the detector. I assume that blank runs also show similar baseline misbehaviour, if not, consider your sample solvent.

Please keep having fun,

Bruce Hamilton

[unmgvar]

i tend to agree with Uwe. it look like bubbles, maybe huge presure drops.
can you monitor your pressure?
it is either the degassing like he says or due to the fact that your are mixing in the pump.

you are using a low pressure pump.
it is also possible that because you are mixing using the pump that you are damaging the proportion valves and check valves. this makes your gradient non reproducable at all, valves are not behaving correctly and also you probably have "nice" retention time shifts.

see what you get if you use 2 solvents only
composition of A= 80% buffer, 15% ACN, 5% MeOH
composition of B= 60% buffer, 35% ACN, 5% MeOH

do a simple gradient that goes from 0% A to 100% B and back. the retention times will change to a new value but should remain stable and your problem shoud be solved