Performance of charged aerosol detection with hydrophilic interaction chromatography

The performance of the charged aerosol detector (CAD) was investigated using a diverse set of 29 solutes, including acids, bases and neutrals, over a range of mobile phase compositions, particularly with regard to its suitability for use in hydrophilic interaction chromatography (HILIC). Flow injection analysis was employed as a rapid method to study detector performance. CAD response was 'quasi-universal', strong signals were observed for compounds that have low volatility at typical operating (room) temperature. For relatively involatile solutes, response was reasonably independent of solute chemistry, giving variation of 12-18% RSD from buffered 95% ACN (HILIC) to 10% ACN (RP). Somewhat higher response was obtained for basic compared with neutral solutes. For cationic basic solutes, use of anionic reagents of increasing size in the mobile phase (formic, trifluoroacetic and heptafluorobutyric acid) produced somewhat increased detector response, suggesting that salt formation with these reagents is contributory. However, the increase was not stoichiometric, pointing to a complex mechanism. In general, CAD response increased as the concentration of acetonitrile in the mobile phase was increased from highly aqueous (10% ACN) to values typical in the HILIC range (80-95% ACN), with signal to noise ratios about four times higher than those for the RP range. The response of the CAD is non-linear. Equations describing aerosol formation cannot entirely explain the shape of the plots. Limits of detection (determined with a column for solutes of low k) under HILIC conditions were of the order of 1-3 ng on column, which compares favourably with other universal detectors. CAD response to inorganic anions allows observation of the independent movement through the column of the cationic and anionic constituents of basic drugs, which appear to be accompanied by mobile phase counterions, even at quite high solute concentrations.