Improving retention-time prediction in supercritical-fluid chromatography by multivariate modelling

Area of Science:

• Analytical Chemistry
• Chromatography

Background:

• Supercritical-fluid chromatography (SFC) retention is influenced by multiple factors.
• Existing liquid chromatography retention models often overlook pressure and temperature effects crucial for SFC.

Purpose of the Study:

• To investigate and improve the prediction of chromatographic retention under SFC conditions.
• To assess the efficacy of established and novel theoretical models, including multivariate approaches.

Main Methods:

• Studied retention models considering modifier content, pressure, and temperature.
• Developed and evaluated multivariate retention surfaces combining modifier fraction and pressure.
• Assessed models based on mass fraction, volume fraction, pressure, and density.

Main Results:

• Multivariate models significantly improved retention-time prediction compared to univariate models.
• A "mixed-mode" model predicted retention times within 5% (mostly within 2%).
• Using mass fraction and density enhanced modeling accuracy over volume fraction and pressure.

Conclusions:

• Multivariate modeling, particularly using mass fraction and density, offers superior retention prediction in SFC.
• Advanced models improve the accuracy of predicting retention times for isocratic SFC separations.
• These findings enhance the development and optimization of SFC methods.