A mixed-mode I/II fracture criterion for predicting the fracture response of composite materials is proposed. This criterion is derived based on a comprehensive study and the consideration of the physics of fracture onset. The fracture phenomenon that causes the various damage mechanisms at the vicinity of the crack tip is examined. It is elucidated that the stress distribution at the crack tip ought to be defined using the reinforcement isotropic solid (RIS) stress state. This new criterion, which is called Improved Strain Energy Density with Mid-point (ISEDM), includes the effect of the fracture process zone (FPZ) and T-stress, which remarkably affect the mixed-mode fracture process, particularly, when mode II is dominant. Substituting the strain energy density in a pure mode I with the strain energy density in the midpoint of mixed-mode I/II is the creative idea employed in proposing this criterion. Because in pure mode I the effects of FPZ are minimal, change of fitting point on fracture limit curve (FLC) from K_Ic to midpoint critical stress intensity factors (CSIF's) makes it possible to consider the effects of FPZ more accurately. In the ISEDM criterion, fracture behavior depends on mechanical properties and CSIF's of the midpoint. Crack initiation angle is considered along minimum strain energy density and this angle is derived in the midpoint of mixed-mode I/II. RIS theory is used as an applicable theory for modeling orthotropic materials in this paper and causes valid and reliable fracture behavior to be extracted. In addition, changing fracture point from pure mode I to the mid-point of experimental data causes the effects of FPZ to be considered without estimating the toughening mechanisms in this zone, and the fracture behavior is extracted with higher accuracy. FLC's in comparison with available experimental data prove that the ISEDM criterion anticipates the fracture behavior of orthotropic materials well. Finding K_IIc based on the analytical method is a valuable achievement. In this article, K_IIc can be predicted with appropriate accuracy, only by CSIF's of mid-point and using ISEDM criterion.

This paper demonstrates how the critical strain energy density in the delamination tip vicinity may be used to explain the physics of delamination growth under mixed mode I/II. A theory previously proposed to physically relate mode I and mode II delamination growth is further extended towards describing the onset of mixed mode I/II delamination. Subsequently, data from the literature is used to demonstrate that this new concept of the critical strain energy density approach indeed explains, based on the physics of the problem, the strain energy release rate level at which crack onset occurs. This critical strain energy density for the onset of delamination appears to be independent of the opening mode. This means that, in order to characterize the fracture behaviour of a laminate, fracture tests at only one loading mode are necessary. Because the load level at which the physical delamination onset occurs at the microscopic level is much lower than the traditional engineering definition of macroscopic onset, further work must reveal the relationship between the macroscopically visible delamination onset, and the microscopic onset.

In this paper, by considering the absorbed energy in the fracture process zone and extension of the minimum strain energy density theory for orthotropic materials, a new mixed mode I/II failure criterion was proposed. The applicability of the new criterion, to predict the crack growth in both laminated composites and wood species, was investigated. By defining a suitable damage factor and using the mixed mode I/II micromechanical bridging model, the absorbed energy in the fracture process zone was considered. It caused the new criterion to be more compatible with the nature of the failure phenomena in orthotropic materials, unlike available ones that were conservative. A good agreement was obtained between the fracture limit curves extracted by the present criterion and the available experimental data. The theoretical results were also compared with those of the minimum strain energy density criterion to show the superiority of the newly proposed criterion.