A micromechanical study of residual stress and its effect on transverse failure in polymer–matrix composites

Abstract

Cure residual stress and its effect on damage in unidirectional fibre-reinforced polymer-matrix composites under transverse loading were studied using a micromechanical unit cell model and the finite element method. The overall residual stress introduced from curing was determined by considering two contributions: volume shrinkage of matrix resin from the crosslink polymerization during isothermal curing and thermal contraction of both resin and fibre as a result of cooling from the curing temperature to room temperature. To examine the effect of residual stress on failure, a model based on the Maximum Principal Stress criterion and stiffness degradation technique was used for damage analysis of the unit cell subjected to mechanical loading after curing. Predicted damage initiation and evolution are clearly influenced by the inclusion of residual stress. Residual stress is always detrimental for transverse compressive loading and pure shear loading. For transverse tensile loading, residual stress is detrimental for relatively low resin strength and beneficial for relatively high resin strength. Failure envelopes were obtained for both biaxial normal loading and combined shear and normal loading and the results show that residual stress results in a shifting and contraction of the failure envelopes.

Life cycle strain monitoring in glass fibre reinforced polymer laminates using embedded fibre Bragg grating sensors from manufacturing to failure

“…High cure temperatures and fast heating/cooling rates can result in highly non-uniform temperature and cure gradients within the part resulting in differential curing and subsequent development of internal stresses. 3,[4][5][6]11 Similar effects in the form of a non-uniform distribution of in-plane shear stresses can arise due to constraints posed by tooling during processing. 10,12 A combination of the above-named residual stress mechanisms and manufacturing defects can cause local premature failure at lower load magnitudes than would otherwise be predicted in a defect-free composite structure.…”

“…For instance, large process-induced residual stresses are known to induce matrix cracking, interfacial debonding and delamination in fibre-reinforced composites. 1 In other works, [4][5][6] demonstrated that for thick-walled composite tubes, residual stresses can be so great that matrix cracking was promoted. At a micromechanic level, Nedele and Wisnom 7 showed that process-induced stresses at the fibre-matrix interface can be as large as 30 MPa, due to the combination of hoop and radial stress components.…”

Multiscale Modeling of Residual Stress Development in Continuous Fiber-Reinforced Unidirectional Thick Thermoset Composites

“…Rosso et al [21] estimated mesolevel cure-induced residual stresses by employing a plane-strain thermoelastic model of a unidirectional carbon fiber composite with hexagonal packing; however, they incorporated cure-shrinkage in an ad hoc manner by modifying the coefficient of thermal expansion of the resin only for a fixed temperature range (thereby not truly accounting for cure-dependent gelation effects). More recently, Zhao and coworkers have reported attempts to capture the cure phenomenology at the RVE level in greater detail with separate accounting for cureinduced shrinkage and thermal expansion effects in unidirectional composites by employing thermoelastic models [22] and thermoviscoelastic models [23]. These RVE simulations involved modeling of isothermal cure and cool-down stages of the thermal cycle.…”

“…Chemical shrinkage was assumed to follow a linear dependence on conversion. Zhao et al [22,23] ignored the interdependence of cure and temperature, which might be a valid assumption for isothermal curing systems with low heats of reaction. However, such an assumption would not be valid for an RVE that is located in the core of a thick composite with a resin that has a very high heat of reaction.…”

“…Cure process models provide a valuable tool to understand the effect of composite microstructure, part geometry, and processing parameters on the development of residual stresses; such models also provide a relatively inexpensive virtual means of optimizing cure processing to minimize the residual stresses. Process models for composite cure are available that address residual stress development at the part level [6,[8][9][10][11][12][13][14][15][16][17][18] or at the "meso" levels involving the microstructure or fiber-matrix interface [19][20][21][22][23]. However, a systematic framework for bridging the information at the two scales has not been demonstrated; such bridging of information is critical for obtaining realistic estimates of residual stresses at the fiber matrix interface and for arriving at an understanding of the efficacy of reinforcement.…”