Retention time prediction in temperature-programmed, comprehensive two-dimensional gas chromatography: modeling and...

Area of Science:

• Analytical Chemistry
• Chromatography
• Computational Modeling

Background:

• Temperature-programmed comprehensive two-dimensional gas chromatography (GC × GC) lacks analytical solutions for retention time prediction.
• Numerical integration is required, but models are prone to errors and overfitting during parameter fitting.

Purpose of the Study:

• To develop a computational physical model for accurately predicting retention times in GC × GC.
• To implement robust error assessment techniques for model predictions.

Main Methods:

• Developed a computational physical model for GC × GC retention time prediction.
• Applied K-fold cross-validation to detect and mitigate overfitting.
• Utilized error propagation with Jacobians to estimate prediction accuracy based on parameter derivatives (entropy and enthalpy).

Main Results:

• The model accurately predicts retention times in both dimensions of GC × GC.
• Error assessment techniques provide a probability distribution of retention times rather than single values.
• Treating predictions as intervals increases the robustness of optimization algorithms.

Conclusions:

• The developed computational model offers accurate retention time predictions for GC × GC.
• Integrated error assessment methods improve model reliability and robustness.
• This approach enhances the utility of GC × GC data analysis and optimization.