Evaluation procedures for reversed-phase HPLC columns in the analysis of strongly basic compounds using principal components analysis for data assessment

The performance of eight high-purity silica-based reversed-phase columns was analysed using chemometric pattern recognition. Tests were performed at pH 3.0 and 7.0, using 10 compounds, three mobile phase modifiers (methanol, tetrahydofuran and acetonitrile) and four parameters {retention factor (k′), column efficiency (N), the Dorsey–Foley efficiency [N(df)] and the asymmetry factor (As])}. Principal components analysis was performed on subsets of the data. Features of both scores and loadings plots, together with the corresponding correlation coefficients between chromatographic tests, are discussed in detail. It is concluded that methanol and acetonitrile exhibit similar properties, the relative performance at each pH differs, the number of compounds used in the test can be reduced to a recommended five, and several of the original tests are redundant. Procedures are recommended for determining a range of tests that provide information on the column performance, which can be applied to other situations.