The effect of chemical constitution on polyisoprene dynamics.

Polyisoprene (PI) melts have been studied, with most reports focusing on systems with high 1,4-cis content. In contrast, 1,4-trans PI homopolymers or random copolymers have seldom been examined, despite a handful of investigations suggesting a distinct dynamic behavior. Herein, we employ all-atom simulations to investigate the effect of chemical architecture on the dynamics of cis and trans-PI homopolymers, as well as copolymers. We examine the thermodynamic, conformational, and structural properties of the polymers and validate the performance of the models. We probe chain dynamics, revealing that cis-PI presents accelerated translation and reorientation modes relative to trans as recorded by the mean square displacement of the chain center-of-mass as well as by the characteristic times of the lower modes in a Rouse analysis. Interestingly, progressing to higher modes, we observe a reversal with trans units exhibiting faster dynamics. This was further confirmed by calculations of local carbon-hydrogen vector reorientation dynamics, which offer a microscopic view of segmental mobility. To obtain insight into the simulation trajectories, we evaluate the intermediate incoherent scattering function that supports a temperature-dependent crossover in relative mobility that extends over separations beyond the Kuhn-length level. Finally, we analyzed the role of non-Gaussian displacements, which demonstrate that cis-PI exhibits increased heterogeneity in dynamics over short-timescales in contrast to trans-PI, where deviations persist over times extending to terminal dynamics. Our all-atom simulations provide a fundamental understanding of PI dynamics and the impact of microstructure while providing important data for the design and optimization of PI-based materials.

Coupling between Polymer Conformations and Dynamics Near Amorphous Silica Surfaces: A Direct Insight from Atomistic Simulations.

The dynamics of polymer chains in the polymer/solid interphase region have been a point of debate in recent years. Its understanding is the first step towards the description and the prediction of the properties of a wide family of commercially used polymeric-based nanostructured materials. Here, we present a detailed investigation of the conformational and dynamical features of unentangled and mildly entangled cis-1,4-polybutadiene melts in the vicinity of amorphous silica surface via atomistic simulations. Accounting for the roughness of the surface, we analyze the properties of the polymer chains as a function of their distance from the silica slab, their conformations and the chain molecular weight. Unlike the case of perfectly flat and homogeneous surfaces, the monomeric translational motion parallel to the surface was affected by the presence of the silica slab up to distances comparable with the extension of the density fluctuations. In addition, the intramolecular dynamical heterogeneities in adsorbed chains were revealed by linking the conformations and the structure of the adsorbed chains with their dynamical properties. Strong dynamical heterogeneities within the adsorbed layer are found, with the chains possessing longer sequences of adsorbed segments ("trains") exhibiting slower dynamics than the adsorbed chains with short ones. Our results suggest that, apart from the density-dynamics correlation, the configurational entropy plays an important role in the dynamical response of the polymers confined between the silica slabs.

Backmapping coarse-grained macromolecules: An efficient and versatile machine learning approach.

Multiscale modeling of polymers exchanges information between coarse and fine representations of molecules to capture material properties over a wide range of spatial and temporal scales. Restoring details at a finer scale requires us to generate information following embedded physics and statistics of the models at two different levels of description. Techniques designed to address this persistent challenge balance among accuracy, efficiency, and general applicability. In this work, we present an image-based approach for structural backmapping from coarse-grained to atomistic models with cis-1,4 polyisoprene melts as an illustrative example. Through machine learning, we train conditional generative adversarial networks on the correspondence between configurations at the levels considered. The trained model is subsequently applied to provide predictions of atomistic structures from the input coarse-grained configurations. The effect of different data representation schemes on training and prediction quality is examined. Our proposed backmapping approach shows remarkable efficiency and transferability over different molecular weights in the melt based on training sets constructed from oligomeric compounds. We anticipate that this versatile backmapping approach can be readily extended to other complex systems to provide high-fidelity initial configurations with minimal human intervention.