Design of linear ligands for selective separation using a genetic algorithm applied to molecular architecture

Area of Science:

• Computational Chemistry
• Materials Science
• Chemical Engineering

Background:

• Continuous purification via adsorption offers advantages over traditional batch methods for chemical reaction workup.
• Separating structurally similar compounds in pharmaceutical synthesis necessitates highly selective adsorption media, which are difficult to identify.

Purpose of the Study:

• To develop an in silico screening process for designing organic ligands for solid-supported adsorption.
• To achieve selective adsorption of pharmaceutically relevant compounds from complex mixtures.

Main Methods:

• Utilized automated molecular dynamics simulations to assess ligand adsorption energy differences for structurally similar solutes.
• Employed a genetic algorithm for iterative optimization of ligand designs through selection and reproduction.
• Evaluated ligand selectivity based on phenyl-phenyl stacking interactions.

Main Results:

• Identified novel ligand designs beyond those conceived by chemical intuition.
• Achieved selectivity by exploiting sterically hindered phenyl-phenyl stacking.
• Reported selectivity energies ranging from 0.8-1.6 kcal/mol in solvent-free simulations.

Conclusions:

• The developed molecular evolution technique enables efficient exploration of chemical space for molecular design.
• This in silico approach is valuable for creating tailored adsorption media for challenging separations.
• The method facilitates the directed design of ligands for specific chemical properties and applications.