Procedural Construction of Atomistic Polyurethane Block Copolymer Models for High Throughput Simulations

TL;DR

This paper introduces a method for automatically generating detailed atomistic models of polyurethane copolymers, enabling high-throughput simulations for material design and analysis.

Contribution

The work presents a novel automated routine for constructing atomistic polyurethane models with customizable parameters, integrating chemical structure conversion and structure factor validation.

Findings

Models accurately reproduce experimental scattering data

Parametric effects on structure are systematically analyzed

Routine is suitable for high-throughput material screening

Abstract

In this work, methods are presented to automatically generate a fully atomistic LAMMPS models of arbitrary linear multiblock polyurethane copolymers. The routine detailed here receives as parameters the number of repeat units per hard block, the number of units in a soft block, and the number of soft blocks per chain, as well as chemical formulae of three monomers which will form the hard component, soft component, and chain extender. A routine is detailed for converting the chemical structure of a free monomer to the urethane bonded repeat units in a polymer. The python package RadonPy is leveraged to assemble these units into blocks, and the blocks into copolymers. Care is taken in this work to ensure that plausible atomic charges are assigned to repeat units in different parts of the chain. The static structure factor is calculated for a variety of chemistries, and the results compared with wide angle x-ray scattering data from experiments with corresponding composition. The generated models reproduce the amorphous halo observed in the scattering data as well as some of the finer details. Structure factor calculations are decomposed into the partial structure factors to interrogate the structural properties of the two block types separately. Parametric surveys are carried out of the effects of various parameters, including temperature, soft block length, and block connectivity on the observed structure. The routine detailed here for constructing models is robust enough to be executed automatically in a high throughput workflow for material design and discovery.