Strategies for automated chromatographic method development and data interpretation

Chapter 1 provides an overview of the work done in the recent years in so-called chemometrics for chromatography, specifically for the optimization of one-dimensional (1D) and two-dimensional (2D) chromatography. One of the most common techniques to optimize chromatographic separations is by retention modeling. During retention modeling, relationships between conditions (e.g. the mobile-phase strength, temperature or pressure) and amount of time that an analytes remains in the column (i.e. is retained) are established. Some chromatographic modes, such as supercritical-fluid chromatography, multiple of those relationships are abundantly present. Therefore, in Chapter 2 strategies for retention modeling using multiple variables is discussed, more specifically the change in mobile-phase strength and the pressure in the system. In Chapter 3 the closed-loop communication between peak-tracking algorithms, retention modeling, design of candidate gradient profiles and the LC and MS systems are discussed. This chemometric strategy was able to optimize and LC-MS separation using only a few iterations. Further development of the closed-loop system to be capable of optimizing LC×LC-MS separations, required peak-tracking algorithms that are able to take both dimensions into consideration. Therefore, Chapter 4 focuses on the development of such an algorithm. Chapter 5 discusses a strategy to expand peak-tracking algorithms to multiple chromatograms. Chapter 6 is focused on the closed-loop method development in LC×LC-MS using so-called shifting gradients. In Chapter 7, the development of an automated feature-mining (i.e. detection of distributions) algorithm is discussed. In Chapter 8, a chemometric approach is discussed that improves the quantitative information obtained by Py-GC and allows for more accurate determination of the average sequence length in copolymers.