[Study on relationship between three-dimensional quantitative structure and retention parameters of chromatography]

Area of Science:

• Computational Chemistry
• Analytical Chemistry
• Cheminformatics

Context:

• Predicting retention parameters in liquid chromatography is crucial for compound separation and analysis.
• Existing methods may lack accuracy or applicability across diverse chemical structures.
• Understanding structure-retention relationships aids in drug discovery and chemical process optimization.

Purpose:

• To develop a novel computational method for predicting liquid chromatography retention parameters.
• To establish three-dimensional quantitative structure-retention relationship (3D-QSRR) models using molecular field analysis.
• To assess the predictive performance of the developed models for various arene compounds.

Summary:

• A new method determines molecular conformations, defines molecular fields using electrostatic and steric interactions, and applies partial least square (PLS) for analysis.
• Comparative molecular field analysis (CoMFA) models were built to study 3D-QSRR for arenes in methanol/water on a C18 column.
• The models demonstrated considerable predictive ability for alkylation-substituted benzenes, multi-ring arenes, and chlorobenzenes, with cross-validated q2 > 0.5 and non-cross-validated r2 > 0.995 for some forms.

Impact:

• The developed method can predict retention parameters for unknown compounds with similar structures.
• Provides insights into the class of compounds based on their molecular field characteristics.
• Facilitates further research into the relationship between molecular structure and liquid chromatography retention behavior.