ABRF Friends

We have an ABI 421/172 amino acid analyzer that has been giving shifting peak retention times in the hydrophobic region sporadically over the last year. Specifically, the peaks in the latter part of the chromatogram creep forward as much as several minutes over the course of numerous runs. The creeping is proportional; the hydrophilic early peaks are stable, then about mid-run the hydrophobic peaks come off faster and faster, with L shifting the greatest percent in each run. Our first observation of this, several months ago, was between reps within single runs, which made data analysis a challenge. We had the instrument extensively serviced and improved some operator techniques with solvents and operation frequency, and the problem finally subsided. We recently started seeing this again, with all those fixes in place. The shifting is occuring more gradually now, but is very noticeable. We just changed columns again, and it seems to have stopped. We have had numerous problems with the PTC columns over the year, but we showed the column was NOT causing shifting RT's during the first episode (the same column on another instrument performed well). While we had several hypotheses, we never proved what precisely caused the shifting RT's (or what fixed them).

Have you seen shifting RT's in only one portion of chromatograms, when everything seems to be working well (eg no leaks, good seals, new column, well-mixed solvents)? What did you do to correct it? Are shifting RT's indicative of the system not being run for a week (as ABI suggests)? If the system is unused for more than a day or so, we purge the pumps and run several blanks and standards. We change HPLC solvents weekly, as directed in our protocols. Because of this and other problems, we wondered about the reliability of the 140B pumps used in our system, but we don't know if they are the sole problem with the shifting RT's. What have been your experiences with the 140B pumps? Have you had problems with maintenance/performance? Have you had ABI PTC column problems? If you have a 421/172 system, what suggestions do you have for consistent optimal performance? We seem to have had a plethora of problems since it was installed, some operator-related but most not. Another 421/172 instrument in our area has not had so many problems, but only recently has it been running as frequently as ours. Both are operated with identical protocols, so we'll have to wait and see how the other one performs over time. By the way, ABI has been working hard with us to try to solve all of our problems. Their sales and service performance has been excellent for us, even when the problems have been very hard to pinpoint.

Thank you for your help!
Nadine Ritter
Analytical Validation R&D
Abbott Laboratories
RITTEN.ADD@notes.abbott.com

The opinions expressed herein are entirely my own and do not necessarily reflect those of Abbott Labs.

We have two ABI 420/172 units in nearly continuous operation and have not observed any retention time problems like the ones you describe. Our experience suggests that there is not an inherent problem with the the 140B pumps.

Shifting retention in only one portion of a gradient chromatogram suggests a pump problem, however. Since the hydrophobics are eluting with a high percentage of the B solvent, one might at first suspect the B side of the pump. Actually, our experience with other gradient pumping systems has generally been that unexpected faster elution of the later components is generally due to a lower than expected flow of A rather than to a higher flow of B. This may point to a problem on the A side, in which the A pump cannot deliver the required amount at the higher percentages of B.

The biggest problem we have experienced with this chromatographic system is mold growth in the solvent A bottle, lines, and pump head. This growth is particularly noticeable if the instrument has not been used for several days. To forestall it, we rinse the A bottle with acetone or methanol, then water, before installing a new batch of mobile phase. We also add 200 ul of 50 mg/ml EDTA to each liter of mobile phase A, but we're not sure whether this really helps or not. Unchecked, this growth becomes severely detrimental to column performance.

Relative to not using the instrument for a while and then expecting retention times to drift: given that your mobile phases are mixed well and you purge the pumps, there is no reason to believe that runs should not be reproducible. Retention characteristics may be different from the previous week, due to partial evaporation of volatile components, but at least the injections should be consistent from one to the next.

We have had some trouble with poor resolution on certain batches of PTC columns, but those problems were addressed promptly by ABI's technical service people.

John G. Hoogerheide
Analytical Methods and Services
The Upjohn Company

We state:
In order to prevent the molt growth, we add 5 % acetonitrile to the Solv.A. You have of course to change the gradient then, but the 'animals' are not there anymore.
Additionally we also saw shifting sometimes in the chromatogram. These were mainly correlated with pumps/seal problems or eventually a precolumn that after replacement gave again good separation.
Houthaeve Tony
Protein & Peptide Group
EMBL
Meyerhofstrasse 1
69012 Heidelberg
Germany

tel.: (49)-6221-387 533
fax.: (49)-6221-387 306
HOUTHAEVE@EMBL-HEIDELBERG.DE