Abstract
It is still a challenge to numerically achieve the interactive competition between ductile damage and brittle fracture in ductile-to-brittle transition (DBT) region. In addition, since two types of fracture occur at two independent material length scales, it is difficult to process them with the same mesh size by using finite element method. In this study, a framework of modelling DBT of a thermal mechanical controlled-rolling (TMCR) steel is explored by using the cellular automata finite element (CAFE) method. The statistical feature of material's microstructure is incorporated in the modelling. It is found that DBT curve cannot be reproduced with only one temperature dependent flow property, for which another temperature dependent variable must be considered. A temperature dependent effective surface energy based on typical cleavage fracture stage is proposed and obtained through a continuum approach in present work. The DBT of TMCR steel is simulated by using CAFE method implemented with a temperature dependent effective surface energy. It is found that numerical simulation is able to produce a full transition curve, especially with scattered absorbed energies in the transition region represented. It is also observed that simulation results can reproduce a comparable DBT curve contrasting to the experimental results.
Highlights
Ductile-to-brittle transition (DBT) is normally found in the body centered cubic (BCC) materials, e.g., steel, due to temperature decreasing and loading rate elevation
Unstable cleavage fracture is commonly initiated by second-phase particle cracking due to dislocation pile-up, which refers to the sequence of three steps: particle breakage, transgranular fracture within a single grain and overcoming of the grain boundary [6]
We firstly present the predicted results of DBT by using a constant effective surface energy
Summary
Ductile-to-brittle transition (DBT) is normally found in the BCC materials, e.g., steel, due to temperature decreasing and loading rate elevation. Ductile damage models (e.g., GTN, Rousselier) combined with RKR criterion model or local approach (e.g. Beremen model) has been widely applied to model the DBT of steel under quasi-static load [11,12] or dynamic load [13,14,15,16,17,18] It is basically a post-processing solution to evaluate the occurrence of cleavage after stress field ahead of crack tip obtained from the constitutive equation of ductile model. A temperature dependent misorientation of grain boundary proposed by Shterenlikht et al [20] has been implemented into the CAFE method to model the DBT of Charpy test of TMCR steel.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
