Objective functions in experimental and simulated chromatographic optimization

A chromatographic objective function to characterise chromatograms with unknown compounds or without standards available

“…Elementary resolution criteria describe the separation between two solutes exhibiting consecutive peaks, or each particular solute from all other solutes in the sample [1][2][3][4][5][6][7][8][16][17][18][19][20]. For this purpose, several COFs of diverse complexity have been proposed.…”

“…Computer-based methodologies are currently the most efficient tools to search the optimal separation conditions in LC [1][2][3][4]. A practical way for determining such conditions is the use of global measurements to appraise the separation quality, using chromatographic objective functions (COFs) [5][6][7][8].…”

Evolution from univariate to multivariate optimization methods in liquid chromatography

“…The mobile phases of 90 and 100% methanol as well as 85 and 100% acetonitrile were selected to optimize both solvent strength and selectivity (4 total experiments). By applying the method outlined in Equations (13) to (16) above, all possible permutations of these solvents were examined. The optimum solvent sequence was determined to be 100% methanol and 90% methanol in zones of 116 and 1810 cm length, respectively.…”

Chromatographic response functions in 1D and 2D chromatography as tools for assessing chemical complexity

“…Such information is crucial for planning the optimization of the chromatographic separation, the objective of which is to find the best conditions that lead to the best chromatographic results according to the degree of separation, the number of peaks, the analysis time, and other chromatographic parameters. The conventional ''trial and error'' approach is still frequently used for the experimental optimization [1], which means that the best chromatographic separation depends on the expertise and the intuition of the chromatographer [2]. However, this type of optimization does not foresee the possibility of interaction between the experimental variables, which makes it very slow and limited, especially when dealing with analytical problems containing several variables.…”