Year: 2026
Type: preprint
Abstract: A systematic exploration of combining micro-to meso-to macro-scale models of crosslinking in three-dimensional polymer networks of Hexanediol Diacrylate (HDDA) is presented. Multi-scale modelling connecting realistic reaction kinetics from population balance (PB) models and network topology from molecular dynamics (MD) form the elements of the new modelling pipeline. Comparison of PB and MD predictions identify the initial radical concentration and model of initiation reactions as the principal source of discrepancy in species concentration profiles. They also largely account for the persisted differences in gel point and gel fraction predictions between PB and MD. Surprisingly, despite unavoidable widely different model conditions MD and PB find common ground in placing HDDA-gelation in a living polymerization regime. Two Random Graph methods are used to consolidate the inputs; the standard Configuration Model (CM) and the more recently popular Clustered Configuration Model (CCM) that incorporates self-cyclisation, double bond formation and small-cycles. It is found that CCM gives more accurate MDderived mesostructure than CM due to the better representation of neighbourhoods. Predictors for macroscale material properties consisting of degree of cyclisation, polymer-size distribution, and spectral graph measures are computed, with improved accuracy for the new CCM.
Source: ChemRxiv
Authors: T. Schilperoort, P.D. Iedema
Institution: University of Amsterdam
Language: English
Topic: Epoxy Resin Curing Processes
Subfield: Polymers and Plastics
Field: Materials Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221789253