The Mechanical Anisotropy and Failure Mechanism of Poly(p-phenylene terephthalamide) Using Machine Learning Molecular Dynamics Simulations

As a peculiarly polymeric crystalline material, poly(p-phenylene terephthalamide) (PPTA) presents anisotropy according to the different intermolecular interactions, such as van der Waals (vdW) bonding along its a-axis, hydrogen bonding along its b-axis, and covalent bonding along its c-axis. The high anisotropy leads to the difficulties and challenges for experimental and theoretical studies on the complex mechanical behavior. In this paper, a machine learning potential for PPTA is established and the anisotropic mechanical behavior and failure mechanism under tensile conditions are investigated with molecular dynamics simulations. The results show that the PPTA undergoes a multilevel growth behavior of stress when stretched along the van der Waals interaction direction, which originates from the multiple changes in the molecular structure of PPTA under tensile stress. Furthermore, the hydrogen-bonding network goes through a reconfiguration and increases the elongation of the material upon stretching along the hydrogen-bonding interaction direction. Additionally, it is found that the Cbenzene–C starts to break first and leads to material failure upon stretching along the chain direction, which is different from previous findings. This work shows the applicability of machine learning potentials in complex crystalline polymer materials and provides atomic-scale insights into the design of high-performance polymer crystalline materials.