United atom and coarse grained models for crosslinked polydimethylsiloxane with applications to the rheology of silicone fluids

• Published in: Physical chemistry chemical physics : PCCP Vol. 25; no. 13; pp. 9669 - 9684
• Main Authors: Khot, Aditi, Lindsey, Rebecca K, Lewicki, James P, Maiti, Amitesh, Goldman, Nir, Kroonblawd, Matthew P
• Format: Journal Article
• Language: English
• Published: England Royal Society of Chemistry 29.03.2023; Royal Society of Chemistry (RSC)
• Subjects: CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND SUPERFLUIDITY; Crosslinking; INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Iterative methods; Kinematics; Material properties; MATERIALS SCIENCE; Molecular dynamics; Polydimethylsiloxane; Rheological properties; Rheology; Shear viscosity; Silicones; Simulation models; Viscosity; Workflow
• ISSN: 1463-9076; 1463-9084
• DOI: 10.1039/d2cp04920a

Siloxane systems consisting primarily of polydimethylsiloxane (PDMS) are versatile, multifaceted materials that play a key role in diverse applications. However, open questions exist regarding the correlation between their varied atomic-level properties and observed macroscale features. To this effect, we have created a systematic workflow to determine coarse-grained simulation models for crosslinked PDMS in order to further elucidate the effects of network changes on the system's rheological properties below the gel point. Our approach leverages a fine-grained united atom model for linear PDMS, which we extend to include crosslinking terms, and applies iterative Boltzmann inversion to obtain a coarse-grain "bead-spring-type" model. We then perform extensive molecular dynamics simulations to explore the effect of crosslinking on the rheology of silicone fluids, where we compute systematic increases in both density and shear viscosity that compare favorably to experiments that we conduct here. The kinematic viscosity of partially crosslinked fluids follows an empirical linear relationship that is surprisingly consistent with Rouse theory, which was originally derived for systems comprised of a uniform distribution of linear chains. The models developed here serve to enable quantitative bottom-up predictions for curing- and age-induced effects on macroscale rheological properties, allowing for accurate prediction of material properties based on fundamental chemical data. Particle-based models of crosslinked PDMS were developed at multiple resolutions for direct numerical simulations of silicone rheological properties and behavior.