Linear and quadratic relationships between retention and organic modifier content in eluent in reversed phase high-performance liquid chromatography:A systematic comparative statistical study

Statistical evaluation of linear and quadratic models of chromatographic re tention was performed. The relationships describing retention were derived for a set of 23 carefully selected test analytes chromatographed on 18 HPLC columns using methanol-water or acetonitrile-water solutions as mobile pha se. It was ascertained whether the square term in the quadratic model impro ves the description of chromatographic retention in a statistically signifi cant manner. It was also checked whether the retention data extrapolated to a hypothetical neat water eluent (log k(w)) obtained with the two models a nd the two organic modifiers are equivalent or should be considered differe nt. The research proved that both models yield similar results and the extr apolated log k(w) values do not differ statistically significantly in the c ase of methanol-containing mobile phases. In the case of acetonitrile-water systems the log k(w) values obtained with linear and quadratic models are normally statistically different, Correlation analysis for log k(w) vs. log P parameters was performed for data determined on six columns. The best co rrelations were achieved employing log k(w) data derived with the linear mo del for methanol/water systems.