Peak Splitting only in early part of Gradient Run

[Cody14]

Hello,

I'm currently validating a method and have ran this multiple times now. This is a gradient run, gradient starts at 0.8 min and rises to 11.5 min. changing from Aqueous to Organic. At ~2.5 min i have 2 peaks that come off within about 0.3 min of each other. This has been ran successfully with no splitting multiple times, then we will randomly see the 2 peaks split/ We have see this as a slight shoulder to a full split peak that shows up ~ 0.2 min down field. This does not always occur, and is not column dependent or sample solvent. I say this because we have run it with the same column and same system suitability solution after we have seen splitting and it returns to normal. I cannot seem to find the issue. This has been seen across multiple systems and columns, and again, is very random. We have also ran this and followed it by the same run and we will see splitting. The only true success I have had is re-seating the column and ensuring a very proper column installation. Any help would be great!

[krickos]

Hi
Without details regarding columntype, sample solvent, mobilphase..it will be harder for the forum to advice.

Resolving the issue by very careful re-installment of column could indicate that it is partly deadvolume related vs the short 0,8min initial isocratic time. But there could be a combination with sample solvent/initial mobilephase mismatch.

[Kreall]

The double peak could indicate various of problems. As Krikos mentioned partly dead volumne inside the system could idicate a double peak. Also a mechnical cloth inside the system from, for example, mobile phase could give such a result. Another problem which gives double peaks is damaged or colapsed stationary phase in column - you can check it by simple doing one run on the reversed flow on column.

Most common for me would be too high organic concentration in sample solvent or in solvent used to wash needle/sample loop. It will have different result in different HPLC system because of different dead volumnes of systems. Solution to this problem is just lowering the concentration of organic solvent as much as you can.

[Cody14]

Hello,

Thank you for the replies. To give you more information, we are using a CSH C18, 1.7 micorm, 2.1 x 100 mm column, sample solvent is 70:30 20 mM Ammonium Acetate Buffer, pH 5.0:ACN, and we are running M.P. A - 0.05%TFA in Water, M.P. B - 70:30 MeOH:ACN. Purge solvent is 90:10 Water:ACN, and wash is 10:90 Water:ACN. I'm not entirely sure it is the solvent as we have run injections of the same solution, on the same column, and the same system, and did not see any issues at first, then it split a few runs later. As well as the reverse, where we initially saw peak splitting and the obtained acceptable chromatography.

[tom jupille]

Any time your diluent is stronger than your initial mobile phase, that's a warning flag for peak shape problems. You may be on the ragged edge. Have you tried changing the injection volume (halve and double) to see if that has any impact?

[folkerta]

Hello,

Thank you for the replies. To give you more information, we are using a CSH C18, 1.7 micorm, 2.1 x 100 mm column, sample solvent is 70:30 20 mM Ammonium Acetate Buffer, pH 5.0:ACN, and we are running M.P. A - 0.05%TFA in Water, M.P. B - 70:30 MeOH:ACN. Purge solvent is 90:10 Water:ACN, and wash is 10:90 Water:ACN. I'm not entirely sure it is the solvent as we have run injections of the same solution, on the same column, and the same system, and did not see any issues at first, then it split a few runs later. As well as the reverse, where we initially saw peak splitting and the obtained acceptable chromatography.

Fancy meeting you here, small world. Also, thanks for not solving this before you left

[Rndirk]

What is the injection volume, and the volume of the sample loop? You can run into these kind of troubles if you inject a very small amount in a big loop, and vice versa. Does your instrument configuration have the right dead volume for the loop you're using?

As the posters above me said, and from my own experience, early eluters show this kind of behaviour if the solvent composition you're injecting is a lot different from the composition of the mobile phase at the start of the run.

What is the starting composition of the mobile phase (%A & %B)?

EDIT: Excuse me for replying to an old topic

[Gerhard Kratz]

You are running a gradient, so it seems that there is a problem with the equilibrium of the column. What is the flow rate? Did you check the pH with a calibrated pH meter?

[folkerta]

Wow. I didn't expect to receive such a response when I commented on this dead thread. I'm pretty new to LC so I will answer your questions to the best of my ability since the OP is no longer working on this.

At the time of the original post, we were really only seeing peak splitting in one of our products. At this time, two of our other products have also begun to show what we believe is splitting. We have had technicians come and replace part after part of our UPLCs. Waters is very concerned about our mobile phase bottle cleaning procedure (as analysts our hands are tied by the powers that be). We are required to soak all of our glassware in liquinox for a minimum of 1 hour, thoroughly rinse with tap water, thoroughly rinse with purified water, and then one rinse with alcohol. Waters completely replumbed one or two of our systems and used their own clean mobile phase bottle for reagents. Their tests didn't show any contamination or chromatography issues so they deemed the system good to go. However, when we ran our method on that system with their clean glassware, we again saw peak splitting.

At one point we were given a column and sample solution to inject that had been proven to produce good chromatography at another lab with the same method, but when we ran it on our systems we saw the same peak splitting. We tried a phosphoric acid clean but still saw peak splitting.

I was only a little bit involved in the investigation of the splitting, so I'm not exactly sure about all of the other remedies that we may have tried.

I'm not sure whether we've tried to change the injection volume. Most of our injections are 1µl or 1.5µl (those volumes are small, but our hands are kind of tied in that area as well).

We tried different µl gradient pre-injection runs (0, 100, and 600µl I believe). The splitting was more pronounced in the 600µl and much less (if at all) in the 0µl.

I don't know much about the UPLC system itself so I can only tell you that we don't have an extension so our sample loop is just the default volume for our system. We have a 50µl syringe and I believe a 15µl needle (if that information is relevant at all). Dead volume and system configuration are relatively new concepts to me so I'm no good in responding to your questions in that area unfortunately.

The starting composition for our most troublesome splitting is 98% MPA and 2% MPB at the start and during the splitting (solution information is in OP's post above).

Flow rate is 0.43 mL/minutes.

The sample solvent pH was checked with a calibrated pH meter.

We keep going back and forth an whether it's a system issue or a method issue because there are no real patterns for us to cling to.

We have had a lot of different people doing a lot of different things to try to figure this out so all suggestions are welcome, and I will do my best to figure out whether we have explored those options or not.

[mattmullaney]

Hi folkerta,

I think an earlier post brought this up, we know the column, column dimensions, sample solvent, needle wash and seal wash, mobile phases A and B...but we still don't know the initial composition of A and B at the beginning of the gradient program. This info may help us to help you.

I do agree, column installation may effect peak splitting behavior, my feeling in this case is a variety of injection solvent mismatch. More info will help confirm this.