PSI - Issue 13

2

Gustavo Henrique Bolognesi Donato et al. / Procedia Structural Integrity 13 (2018) 1879–1887 Leonardo G. F. Andrade and Gustavo H. B. Donato / Structural Integrity Procedia 00 (2018) 000–000

1880

Nomenclature Latin a :

Crack depth ( mm ) a/W : Relative crack depth ( mm/mm ) b: Remaining ligament B : Thickness ( mm ) C(T) : Compact under Tension specimen E : Elastic modulus ( GPa ) H : SE(T) daylight length ( mm ) J :

J integral. Non-linear energy release rate ( J/mm 2 )

M:

Deformation limit Hardening expoent

n :

W :

Width ( mm )

Greek σ 0 :

Reference stress

σ ys : σ uts :

Yield stress

Ultimate tensile stress

υ:

Poisson ratio

1. Introduction The increasing demand for safety, performance, reliability and efficiency of mechanical structures during the last decades has led to optimized components characterized by complex geometries and reduced thicknesses operating under severe loadings. Consequently, high stress and strain fields can be observed, representing potential risks, in special for applications containing crack-like defects. In this context, the use of high-resistance high-toughness materials is of great interest, since they can withstand large stresses combined to large plastic deformations before failure. From a fracture mechanics perspective, this usually implies the need to employ the theoretical background of the Elastic-Plastic Fracture Mechanics – EPFM – represented by the J -integral (Anderson, 2017), whose validity limits should be addressed to guarantee accurate and safe structural integrity assessments and failure predictions. To guarantee the validity of the single-parameter EPFM, the stress fields quantified by the J -integral (employing the well-known HRR field, Fig. 1) must representative ones found in small-scale laboratory specimens and, at the same time, in structures made of the same material and loaded under Small Scale Yielding – SSY – conditions. As illustrated by Figs. 1(a-c), this is the similitude concept and validates the use of single-parameter fracture mechanics to describe crack driving forces and thus stresses in the near-crack region of a structure or laboratory specimen (Anderson, 2005). This work focuses on the use of the EPFM, quantified by the J -integral (Rice, 1968), to characterize crack-driving forces and to support design activities and structural integrity assessments. Landes and Begley (1972) where the first researchers to propose a failure criterion based on elastic-plastic parameters. The technique included a structural evaluation based on the stress states, which, in its turn, were described by J-integral; this quantity univocally quantifies the stresses in the crack region for a monotonically loaded structure or component (Rice and Rosengren, 1968; Hutchinson, 1968). However, for J -integral to be valid as a stress intensity quantity and as a material property, it is mandatory that some premises are respected, including: monotonic loading and small-scale yielding. The loss of stress triaxiality is the most common reason for invalidating J , since makes mechanical properties dependent not only to the material being tested, but also to the geometry and loading mode being tested. In this context, this work is motivated by the need for robust and pragmatic criteria for the use and validation of J integral as a stress intensity quantity capable of describing fields ahead of a loaded crack. In order to achieve this goal, refined numerical models based on the Finite Element Method (FEM) allow the detailed study of fracture mechanics laboratory specimens and, as a reference of high stress triaxiality, the study of Modified Boundary Layer (MBL) models. The central objective is to establish the valisity limites of the EPFM based on J -integral, guaranteeing the similitude concept and, thus, the transferability of mechanical properties from laboratory to real applications.