we were having problems with buffer reproducibility, so for our HILIC methods we have resorted to gravimetric preparation. This means that once established the right amount of the eluent components, we weigh them all (salts and liquids). So far this has proven extremely reproducible with no K' prime shift in over 1000 runs.

This will avoid: 1) measuring the pH in organic (which is not accurate) and 2)contaminating the eluent with the pH probe

Extremely small deviations in pH will still be present, but given that a good method chooses a buffer pH far from the analyte pKam this should not be a problem. At the same time you will be gaining extra reproducibility from constant organic/water mix