Multiscale Modeling of Thermoplastic Elastomers for Enhanced Mechanical Properties

Zhang, Min

doi:https://doi.org/10.21985/n2-refv-vy49

Work

Multiscale Modeling of Thermoplastic Elastomers for Enhanced Mechanical Properties

Public

Download PDF

Download All Files (.zip)

To relate the mechanical responses of hard-soft copolymer systems with their microstructures, a coarse-grained molecular dynamics (CGMD) approach is employed. With the generic bead-spring polymer model mapped from atomistic simulations, this dissertation first studies the morphology of structures with various hard-soft compositions and interactions between hard beads. Following that, this dissertation investigates the enhancement mechanism of hard domains under tensile and shear loading conditions with pressure. The energy factor that denotes the interaction between hard beads dominates the micro-phase separation and morphology. The numerical experiments also show that pressure is the most crucial factor in shear-under-pressure tests, with larger pressure leading to higher shear resistance of the copolymers. To help understand thermoplastic polyurea and polyurethanes (TPUs) from a macro-mechanical perspective, mechanical properties of both hard and soft domains are calculated using the CGMD model. The viscoelastic behaviors of soft matrix and hard-soft copolymers are computed from the stress autocorrelation function (SACF). The stress relaxation indicates that the soft matrix is in a rubbery state at room temperature while hard domains are “solid-like” and can be viewed as elastic solids in a macroscale model. The dynamic shear modulus is then fitted with Prony series using a two-step optimization method. In addition, local elastic constants of hard domains are computed using the stress-strain fluctuation method with purely local stress and local strain. The results can be used as inputs for macroscale models for copolymers and can provide guidelines for understanding and designing polymeric materials. Once the macroscale mechanical properties of both phases are obtained from the CGMD model, continuum scale simulation of TPUs is needed to understand the reason for their excellent blast mitigation properties. To make clear how hard and soft domains interact with each other in a continuum scale, Finite Element Analysis (FEA) is used to study the linear viscoelastic response of polyurea by introducing a gradient interphase area around hard domains. Niblack-algorithm-based AFM image binarization is employed to provide geometry inputs while the viscoelastic master curves of soft matrix are obtained via Dynamic Mechanical Analysis (DMA) data at low frequencies and Molecular Dynamics (MD) estimations at high frequencies. Interphase property gradually changes when the distance from hard domains increases. Both spatial and property distributions of this interphase region affect the viscoelastic response of the copolymer system. To quantitatively investigate how the structural and property features of the interphase affect the system’s energy storage and dissipation, we represent the interphase region with a 15-element vector. Statistical features of the vector represent physical meanings of the interphase. For example, length of nonzero elements means the interphase thickness. Borrowing the concepts of decision tree and random forest from machine learning, we apply a ranking algorithm to identify feature significance on four mechanical responses. Results show that modulus of hard domains determines the instantaneous storage modulus change while the volume fraction of hard domains dominates the long-term modulus change. Following volume fraction of hard domains, total interphase volume fraction and shifting factor distribution mainly affects tanDelta peak decreases while the mean and maximum value of shifting factors determines the tanDelta peak shifting.

Creator