We have been experienced a strange phenomena: We analyze a drug substance on Waters Alliance equipment with Supelcosil ABZ-Plus column, MeOH/Buffer gradient (buffer: NH4OAc, HOAc, Bu4NH4OH, H2O - without adjusting the pH). The product has 8 impurities. Seven impurities and the main product have stable retention time while one of the impurities moves little by little, day by day. For example with a new column it has RT 9.2 min (RRT 0.64) and after one week of continuous run it has RT 8.2 min (RRT 0.58), while - repeat - all other peaks have stable RT and RRT-s. (The mobile phase does not seem to be the problem, because changing it does not influence the RRT. It does influence a little bit the RT, but we are used to setting the flow rate in order to the main peak have constant RT.)

I thought that perhaps the column needed to be "saturated" with the impurity, therefore I injected several times big quantities of it. But it did not have any other influence than a normal, "impurity concentration level" injection. It seams that the important is the eluent to pass through the column and not the analyte.

What could be the reason of this "shifting peak"? Has anyone experienced something similar? What could we do to "stop" the peak?

1. What's the pH?
2. Is the migrating compound ionic, and the others are not? More specifically, is the migrating analyte acidic?

Uwe!

Thanks for your answer/questions!

1. The pH of the buffer is about 4.9, why is it important if we do not have to adjust it?

2.Yes, the analyte is acid (at least it has a carboxilacid group), we also thought that it had something to do with the problem, but I still can not explain. Can you explain the "theory" of the migration?

A side reaction in the preparation of this stationary phase leaves amino groups on the surface. An acidic analyte would - if ionized - interact with this group by an ion-exchange mechanism on top of the reversed-phase mechanism. As this amino group splits off, the retention is reduced.
I suspect that this is the cause of the problem.