Optimisation of gradient elution with serially-coupled columns. Part I: single linear gradients

Area of Science:

• Analytical Chemistry
• Chromatography

Background:

• Single chromatographic columns often fail to resolve complex mixtures.
• Previous work established an isocratic approach for optimizing mobile phase, stationary phase, and column length.
• Serial column combinations offer enhanced resolution for challenging separations.

Purpose of the Study:

• To implement optimal linear gradients for serially combined columns.
• To decrease analysis time for compounds across a wide polarity range.
• To resolve a mixture of 15 sulfonamides using a novel gradient approach.

Main Methods:

• Combined five ACE columns with different selectivities (C18, cyano, phenyl).
• Developed a gradient predictive system using numerical integration to simulate chromatograms.
• Compared two optimization approaches: pre-selecting gradient vs. pre-selecting columns.
• Incorporated refinements for gradient delays and peak half-width modeling.
• Utilized genetic algorithms to reduce computation time for gradient optimization.

Main Results:

• Successfully simulated chromatograms under linear gradient profiles.
• Identified optimal separation conditions using Pareto plots (analysis time vs. resolution).
• Reduced gradient optimization computation time to approximately 3 minutes using genetic algorithms.
• Demonstrated the effectiveness of the developed gradient optimization method.

Conclusions:

• Optimal linear gradients for serial columns can significantly decrease analysis time.
• The developed gradient predictive system improves the realism of separation predictions.
• Genetic algorithms provide an efficient solution for computationally intensive gradient optimization.
• This approach enhances the capability of serial column chromatography for complex mixture analysis.