Problems involving the determination of the column-only band broadening in columns producing narrow and tailed peaks

Peak deconvolution to correctly assess the band broadening of chromatographic columns

“…Since this method explicitly assumes the peak is perfectly Gaussian shaped, large errors may arise if applied to asymmetrical peak. Nevertheless, it is still the most commonly used method 30,53,54]. In the present study, the following peak width methods are used: w = 2.35 t at 50% (i.e., the half width method), w = 4 t at 13.5%, w = 5 t at 4.4% and w = 6 t at 1.1% of the peak maximum.…”

“…This also puts a serious strain on the ability to measure the "columnonly" efficiency of these columns [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31], especially when using weakly retained components. This is most regrettable, for it can be inferred that it are especially these peaks that are most sensitive to the hydrodynamic quality of the packing, which is precisely one of the main parameters one is after when trying to determine the column-only efficiency [9][10][11][12][13][14][15][16]32,33].…”

“…subtraction method, owing its popularity to the fact that it is based on a simple subtraction [14][15][16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] …”

“…In the specific case of chromatography, the variance subtraction method furthermore also suffers from the following inherent problem: because the method can only be correctly applied when the variance is accurately calculated via the 2nd order moment integral, and since the value of this integral can still grow very strongly when going deep into the tail of an asymmetric peak [15,20,30], the resulting variances can become irrelevantly large compared to what is relevant for the practical reality [15,20,29,[34][35][36]. Consider for example a peak whose main body (e.g., upper 95%) is rather narrow while its lower part is severely tailed because of some dead zones in the detector or elsewhere in the extra-column volumes.…”

“…Consider for example a peak whose main body (e.g., upper 95%) is rather narrow while its lower part is severely tailed because of some dead zones in the detector or elsewhere in the extra-column volumes. Since the 2nd order moment integral can only be correctly determined by integrating the peak until deep into the tail [29,30,[34][35][36], the corresponding variance-based plate height can become so strongly dominated by the tail of the peak that its value no longer bears any relation to the aim of the separation, i.e., achieving a given resolution or peak purity at half height or the 4-, 5-or 6-width. The emphasis on the peak's tail in the variance subtraction method in fact leads to the following, highly unfortunate situation: the higher the sensitivity with which one tries to measure the band broadening, the more of the peak tail will be observed, and hence the higher and the more practically irrelevant the resulting plate height value.…”