Optimal experimental designs in RPLC at variable solvent content and pH based on prediction error surfaces.

Area of Science:

• Analytical Chemistry
• Chromatography
• Chemometrics

Background:

• Optimizing reversed-phase liquid chromatography (RPLC) separations using pH requires efficient experimental designs.
• Traditional designs are unsuitable for nonlinear retention behavior of ionizable compounds with pH.
• Existing optimal designs often require iterative application.

Purpose of the Study:

• To investigate an extension of G-optimal designs for RPLC separations involving solvent content and pH.
• To enhance prediction quality for ionizable compounds by optimizing experimental design.
• To determine optimal point distribution and number for various acid-base behaviors and explore common designs.

Main Methods:

• Application of an extended G-optimal design strategy.
• Utilized a stepwise approach for rapid estimation of optimal designs.
• Tested with a set of probe ionizable compounds with known retention behavior.

Main Results:

• The extended G-optimal design effectively addresses nonlinear retention behavior in RPLC.
• A stepwise strategy provided rapid estimation of suitable experimental designs.
• Insights into optimal design point distribution for different compound types were gained.

Conclusions:

• The proposed G-optimal design extension enhances data quality and efficiency in RPLC.
• The method is suitable for optimizing separations of ionizable compounds across varying acid-base properties.
• Derived design construction rules aim to minimize mathematical treatment for practical application.