On the reproducibility of column performance in liquid chromatography and the role of the packing density

The packing behavior of a typical 10 mu m C-18 stationary phase was studied in terms of the resultant column efficiency and capacity factor. The column-to-column reproducibility of these parameters under identical packing procedures is assessed. Correlation of these parameters and the column void volume to the column packing density is reported. Two regimes were studied; that of poor-and well-packed columns. For poorly packed columns, the column-to-column variability is high, but a concomitantly poor same-column reproducibility of measurement suggests that little statistically significant difference exists between different columns packed with the same procedure. There is also no statistically significant correlation between the column parameters and the packing densities, however, the poorest columns skewed a degradation of performance after the drying procedures used to obtain the column masses. Well-packed columns showed much less degradation upon drying. For the well-packed columns, statistically significant column-to-column differences were observable, mainly due to a high same-column precision of measurement. Analysis of the results suggests that even well-packed columns are not optimally packed and that regions of high and low density coexist along the column. The results are compared to those achieved with semi-preparative columns packed with the same slurry procedure and preparative columns packed under dynamic axial compression. Poor day to-day, same-column reproducibility (degradation) under ambient conditions was observed in conjunction with a high column-to-column variability for the semi-preparative columns.