Benchmarking OPLS-AA and OpenFF for HDI–PEG Thermoplastic Polyurethanes with Varying Soft-Segment Length
Year: 2026
Type: article
Abstract: Thermoplastic polyurethane properties are governed by the interplay between soft-segment mobility, hard-segment interactions, and segmented morphology, yet the extent to which atomistic predictions of their thermal and mechanical behavior depend on force-field choice remains insufficiently benchmarked. Here, we combine FTIR, DSC, TGA, and tensile testing with all-atom molecular dynamics simulations to investigate HDI-PEG polyurethane systems across a controlled soft-segment series. Experimentally, films with PEG molecular weights of 400, 1000, and 1500 g/mol were characterized, while simulations were extended to 400-2000 g/mol to compare two complementary force-field frameworks under a consistent protocol: OPLS-AA, a conventional atom-type-based force field, and OpenFF/Sage, a direct-chemical-perception framework augmented here with bespoke torsional refinements. Both force fields reproduce the composition-driven decrease in Tg and density with increasing PEG length, but differ systematically in absolute values, with OPLS-AA predicting higher densities and Tg values than OpenFF. Tensile experiments show the highest elastic modulus for PEG400, a marked decrease at PEG1000, and a partial recovery at PEG1500. Although nanosecond-scale deformation simulations overestimate absolute moduli because they probe high-rate elastic response, they recover composition-dependent stiffness differences, with OpenFF yielding a more pronounced non-monotonic trend than OPLS-AA. Overall, this work provides an experimentally anchored benchmark for assessing which composition-driven trends in HDI-PEG polyurethanes are robust across force-field families, and which observables remain sensitive to model assumptions and simulation scale.
Source: Molecules
Language: English
Subfield: Physical and Theoretical Chemistry
Field: Chemistry
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $3,258
Funder: European Commission
OpenAlex ID: W7153270035
