Peak tailing of olanzapine and doxylamine on LC-QTOF method

We have a method for several hundred drugs using an Agilent Infnity II LC and 6545 QTOF. The separation is using a Waters BEH C18 50 * 3.0 1.8um column with 0.1% formic acid in water / acetonitrile as the mobile phase. 

We see severe peak tailing and poor shape for several analytes but most notable are olanzapine and doxylamine. We have tried many things to troubleshoot the problem including flushing of the system and column with a range of solvents, passivating the system and column with phosphoric acid and/or medronic acid, moving the method to a different instrument, replacing the entire flow path of the LC, different brands of moble phase solvents. We have also tried a number of different lot numbers of the same column, and different C18 columns and while the chromatography can be slightly better or worse from column to column, the tailing generally persists.

The image attached shows the chromatogrtaphy of a typical unaffected compound (ketamine) and the two problem compounds. 

Has anybody else experienced this problem or have any suggestions on other things to try?

Was this helpful?