Background: A frequent impediment to accurate quantitation in bioanalytical LC-MS arises from carryover. For many new chemical entities in drug discovery carryover is not limited to the autosampler, but instead arises from several different sources.
Method: We tested several different columns, injector wash sequences and gradient compositions to understand and eliminate these sources. In many instances carryover was dictated by the elution gradient and column as much as the autosampler hardware and wash protocol.
Conclusion: Several trends were observed. First, different columns resulted in significantly different amounts of carryover (even for nominally the same column chemistry). Second, a continuous high organic wash of the column was not as effective at removing carryover as cycling between high and low organic mobile phases during the column wash. Combining our observations (column, gradient and autosampler configuration) we devised a short 3-min method that is appropriate for a diverse set of new chemical entities and minimizes carryover while still being sufficiently robust to use in a drug-discovery setting.