Brittle Fracture Arrest - The State of Art in a Multiscale Universe

Snartland, Brage D. (Norwegian Univ. of Science and Technology) | Østby, Erling (DNV-GL) | Hauge, Mons (Statoil) | Ren, Xiaobo (SINTEF Materials & Chemistry) | Thaulow, Christian (Norwegian Univ. of Science and Technology)

OnePetro 

Abstract

This state-of-the-art paper is devoted to testing and evaluation of microstructural crack arrest. Testing and analysis of crack arrest have developed in the last decades, enhancing our understanding of the mechanisms behind crack arrest in a continuum mechanics perspective. Understanding crack arrest is important when operations are moving towards Arctic regions as low temperatures are detrimental to most steel’s fracture toughness. Large-scale testing is expensive and unpractical, and current methods fail to reflect the microstructural and micromechanical features of the fracture process. In order to increase the effectiveness of characterizing crack arrest properties, small-scale tests, as well as numerical methods, have been developed. The mechanical basis and mechanisms behind crack arrest are presented. Global and micro-arrest is considered. Key methods for understanding, evaluating and obtaining arrest parameters are presented: (i) statistical treatment of experimental results, (ii) barrier models for separating fracture and arrest sequences, and (iii) numerical tools for determining arrest behaviour. Brief presentations of the main mechanisms of crack arrest are presented with focus on the micromechanisms of arrest. The effect of grain boundaries, lattice orientation and second-phase particles upon propagation controlled cleavage are discussed, as well as their role in the arrest mechanism. Developments in arrest testing and evaluation are presented. Experimentally and numerically obtained results are linked to relevant mechanisms and theory, exhibiting the predictability and importance of crack arrest properties, and the understanding of the governing mechanisms behind crack arrest. The potential for increased understanding of the brittle fracture arrest phenomenon associated with new methods for nanomechanical testing of the material properties inside individual grains, and over grain boundaries, as well as the rapidly improving capabilities of atomistic modelling of deformation and fracture, is presented to pave the way for the future research within this field. Areas where further research could enhance our knowledge of crack arrest are listed.

Introduction

Crack arrest is considering running cracks that are halted due to increasing resistance to crack propagation and/or reduced crack driving force. The former may be due to microstructural barriers or thermal gradients in the material. The latter may occur under partly displacement controlled loading, where the crack extension may increase the compliance of the structure and reduce the local crack driving force, or as a result of dynamic effects caused by impact loading or stress oscillations in the structure. This paper is mainly concerned with aspects related to the material’s resistance to crack propagation, i.e. the arrest toughness. Further, crack propagation is assumed to be dominated by cleavage fracture, i.e. ductile fracture and fatigue are not considered. The relative importance of these factors depends on the scale of which the arrest is considered. Further, the arrest can also be considered for different scenarios ranging from arrest of single grain sized microcracks up to arrest of macroscopic cracks on in the centimeter to meter range. In the first group the arrest happens locally, probably highly influenced by local microstructural features like grain boundary orientation, and would rather be categorized as avoidance of cleavage initiation on the macroscopic scale. In the latter group the problem is more of a conventional engineering fracture mechanics issue, ideally assessed through knowledge or measurements of the macroscopic arrest toughness, Kia. Ultimately, the two groups are part of the same problem, and there is a research aim to establish quantitative relations at different scales in orderto arrive at a general treatment of the problem.