Rate-dependent elastic and elasto-plastic cohesive zone models for dynamic crack propagation

Abstract

To overcome deficiencies with existing approaches a new cohesive zone model is introduced and trialled in this paper. The focus is on rate-dependent cohesive zone models which have appeared in the recent literature but can be shown to suffer unrealistic behaviour. Different combinations of material response are examined with rate effects appearing either in the bulk material or localised to the cohesive zone or in both. A benefit of using a cohesive-zone approach is the ability to capture plasticity and rate effects locally. Introduced is a categorisation of bulk-material responses and cohesive zone models with particular prominence to the role of rate and plasticity. The shape of the traction separation curve is shown to have an effect and captured in this paper with application of a trapezoidal cohesive zone model. Rate dependency for the cohesive zone model is introduced in terms of a ratedependent dashpot models applied either in parallel and/or in series. Traditionally, two possible methods are adopted to incorporate rate dependency, which are either via a temporal critical stress or a temporal critical separation. Applied singularly, tests reveal unrealistic crack behaviour at high loading rates. The new rate-dependent cohesive model introduced here couples the temporal responses of critical stress and critical displacement and is shown to provide for a stable realistic solution to dynamic fracture. Dynamic trials are performed on a cracked specimen to demonstrate the capability of the new approach.

A computationally efficient cohesive zone model for fatigue

“…From the results shown in Figures 11 and 12, it can be concluded that any value higher than 6 for the updates parameter N u provides satisfactory results with a significant reduction in computational cost. It is clear from Figure 11 that the predicted results are reasonably close to each other for values of N u set to (56,28,14,7), which correspond to ΔN ≈ (100, 200, 400, 800), respectively. Differences are more noticeable however for values of N u lower than 6, where N u = 3 corresponds to ΔN≈2000 and N u = 1 corresponding to ΔN≈5600, as shown in Figure 12.…”

“…The effect of the tractionseparation law shape is discussed in Tvergaard and Hutchinson, 27 where it is concluded that the shape of the law does not significantly affect the analysis results. However, it is shown in previous studies 28,[30][31][32] that the TSL has an effect on fracture behaviour but the significance of this effect depends on geometry and material properties.…”

Computational Methods for Ductile Fracture Modeling at the Microscale

“…It also allows to introduce interesting physical mechanisms into the fracture model through the use of different kind of traction-separation laws and the energy dissipation rate can be controlled very accurately. These traction-separation laws can include rate effects (Salih et al (2016)), account for stress triaxiality ratio (Banerjee and Manivasagam (2009)) and fatigue effects (Nguyen et al (2001)), among others.…”

Rate-Dependent Cohesive Models for Dynamic Mode I Interfacial Propagation and Failure of Unidirectional Composite Laminates

“…A few studies were conducted to understand the feasibility of existing cohesive models in delamination studies of composite interfaces [29][30][31]. May et al [29] identified the necessary of considering simultaneously the rate effect of interface strength and fracture toughness, through the comparison study of four different cohesive models for the impact simulation of composites.…”