Designing porous molecularly imprinted polymers via simulation of pre-polymerisation mixtures: a case study with trinitrotoluene

Abstract

Selective adsorption of hazardous micropollutants from water remains a critical challenge in sustainable materials design. Herein, we demonstrate a combined computational–experimental approach to rationally engineer molecularly imprinted polymers for targeted porosity, using 2,4,6-trinitrotoluene as a model template. By simulating pre-polymerisation mixtures of monomers, crosslinkers, and solvent using molecular dynamics, we capture key template–monomer interactions and predict the resulting porosity of the final polymer network. Surface area and free volume predictions from simulations show excellent agreement with experimental nitrogen sorption data across varying solvent compositions. Our findings highlight a fundamental trade-off between imprinting efficiency (favoured in acetonitrile-rich environments) and porous structure (promoted by dimethyl sulfoxide). We validate that pre-polymerisation simulations alone can accurately guide formulations toward high-performance materials, opening new pathways for computationally-driven design of porous polymeric adsorbents.