A methodology for predicting processing induced thermal residual stress in thermoplastic composite at the microscale. (15th February 2022)

• Record Type: Journal Article
• Authors: Parambil, Nithin K.; Chen, Branndon R.; Deitzel, Joseph M.; Gillespie, John W.
• Is Part Of: Composites. Number 231(2022)
• Publication Date: 2022-02-15
• Subjects: Thermoplastic composites -- Residual stress -- Raman spectroscopy -- Micromechanics; Composite materials -- Periodicals; Materials science -- Periodicals; Composite materials; Periodicals; Electronic journals; 620.118
• DOI: 10.1016/j.compositesb.2021.109562
• Languages: English
• ISSNs: 1359-8368

Abstract

Abstract: A computational finite element (FE) model of thermal residual stress was developed for carbon fiber/thermoplastic composites at the microscale and implemented via user material subroutine (UMAT) in ABAQUS. This model accounts for cooling-rate effects on crystallinity and stress-free temperature, temperature-dependent elastic modulus, temperature-dependent coefficient of thermal expansion (CTE) of the matrix, and the temperature-independent transversely isotropic properties of the carbon fiber. Results are generated for a model composite, consisting of a single carbon fiber embedded in a polypropylene thin film. Single filaments are pre-tensioned in the polymer melt to induce different levels of residual axial strain as well as maintain straight fibers during cool-down. Three different preload conditions (1g, 4g, and 8g) were experimentally fabricated and modeled. The residual strain along the length of the fiber is quantified and validated including the shear lag region that develops at the free edge of the sample. Experimentally measured residual strain shows a good correlation with the FE predictions for the applied fiber preloading conditions.