Influence of the acid-base ionization of drugs in their retention in reversed-phase liquid chromatography.

The effect of the ionization in the RP-HPLC retention of 66 acid-base compounds, most of them drugs of pharmaceutical interest, is studied. The retention time of the compounds can be related to the pH measured in the mobile phase (pwsH) through the sigmoidal equations derived from distribution of the neutral and ionic forms of the drug into the stationary and mobile phases. Fitting of the obtained retention vs. pH profiles provides the retention times of the ionic and neutral forms and the pK values of the drugs in the mobile phase (pwsK). The obtained pwsK values are linearly correlated to the pK values in water (pwwK) with two different correlations, one for neutral acids and another for neutral bases that reflect the different influence of the dielectric constant of the medium in ionization of acids and bases. The retention of the neutral species is well correlated to the octanol-water partition coefficient of the drugs as measure of the lipophilicity of the drug, which affects chromatographic retention. Also, the retention time of the ionized forms is related to the retention time of the neutral forms by two different linear correlations, one for anions and the other for cations. These last correlations point out the different retention behaviour of anions and cations: anions are less retained than cations of the same lipophilicity, as measured by the octanol-water partition coefficient of the neutral form. The different retention behaviour of anionic, cationic and neutral forms is confirmed by the hold-up times obtained from different approaches: pycnometry and retention times of anionic (KBr and KI) and neutral (DMSO) markers. Hold-up times obtained by pycnometric measurements agree with those obtained by retention of neutral markers (0.83-0.85 min), whereas hold-up time for anions is mobile phase pH dependent. At acidic pH it is similar to the hold-up time for neutral markers (0.83 min), but then it decreases with the increase of mobile phase pH to 0.65 min at pH 11. The decrease can be explained by the ionization of the silanols of the column and exclusion of anions by charge repulsion. Although not directly measured, the obtained retention data and correlations indicate hold-up time for cations are similar or slightly lower than hold-up time for neutral compounds (0.77-0.83 min). The model proposed and the correlations obtained can be very useful for its implementation in retention prediction algorithms for optimization of separation purposes.