PSI - Issue 61

Berkehan Tatli et al. / Procedia Structural Integrity 61 (2024) 12–19 B. Tatli et al. / Structural Integrity Procedia 00 (2024) 000–000

13

2

1. Introduction

DP steels stand out as prominent members of the Advanced High Strength Steels (AHSS) family thanks to their exceptional mechanical properties and versatility. These desirable features have positioned DP steels as viable candi dates in the automotive industry, where they play a role in reducing fuel consumption and manufacturing lightweight automotive components. The mechanical advantages of DP steels include high ultimate tensile strength (UTS), low yielding stress, high early-stage hardening, and macroscopically homogeneous plastic deformation without the forma tion of Lu¨ders bands (Tasan et al. (2015)). The exceptional mechanical properties of DP steels are mainly attributed to the presence of brittle martensite islands interspersed within a ductile ferrite matrix. This heterogeneous microstruc tural composition results in complex failure mechanisms due to the distinct hardening mechanisms exhibited by each phase. Throughout extensive experimental and numerical investigations spanning several years on DP steels, three local fracture mechanisms have been identified, namely ferrite (F / F) grain boundary decohesion, ferrite-martensite (F / M) grain boundary decohesion, and martensite cracking (Tang et al. (2021); Avramovic-Cingara et al. (2009)). While those mechanisms do not encompass all the fracture modes of DP steels, they are the most prominent and critical ones of interest. The dependency of the failure modes on the morphology, martensite volume fraction, and distribution was investigated and related to the crack initiation and propagation of DP steels (see, e.g., Aydiner et al. (2024)). Owing to the existence of various fracture modes, it is imperative to analyze each mechanism with a distinct multiscale approach (see, e.g., Yalcinkaya et al. (2019); Yalc¸inkaya et al. (2021); Liu et al. (2022)). Constitutive response of the DP steels is characterized by phenomenological material models in several studies (see, e.g., Hosseini-Toudeshky et al. (2022); Sirinakorn and Uthaisangsuk (2018)). Due to the limited plastic flow within the martensite phase, conventional J 2 plasticity has emerged as a suitable choice for its characterization, as demonstrated in Aydiner et al. (2023). However, employing an isotropic plasticity model for the highly ductile ferrite phase fails to capture its crystallographic nature adequately. To address this, several studies (e.g., Liu et al. (2020); Woo et al. (2012)) have adopted a crystal plasticity formulation, enabling the examination of determinant factors such as crystallographic orientation and slip, which play a significant role in the anisotropic response of the ferrite phase. Several failure frameworks have been proposed to capture crack initiation or propagation within each phase at di ff erent length scales. Ramazani et al. (2016) examined martensite cracking using the extended finite element method (XFEM). The maximum principal stress and the Bao Wierzbicki criterion are also easy-to-implement uncoupled damage models in the literature for modeling martensite cracking (see, e.g., Matsuno et al. (2015)). Void nucleation and growth in the ferrite matrix are modeled using both the modified Mohr-Coulomb (MMC) and Gurson-Tvergaard-Needleman (GTN) models, as demonstrated in studies such as Qin et al. (2020); Ayatollahi et al. (2016). Yalc¸inkaya et al. (2022); Aydiner et al. (2023, 2024) investigated inter granular crack initiation and propagation in both ferrite / ferrite and ferrite / martensite grain boundaries by employing a cohesive zone framework coupled with crystal plasticity formulation. Emdadi and Zaeem (2021) studied transgranular and intergranular crack propagation in polycrystalline brittle materials utilizing a phase field framework. Shanthraj et al. (2016) combined a phenomenological crystal plasticity formulation with a novel obstacle phase field energy model to analyze damage accumulation across various macroscopic and mesoscopic examples. In this study, three-dimensional DP steel RVEs, with random crystallographic orientation and varying martensite volume fractions (15% and 37%), as well as di ff erent morphologies, are generated using Voronoi tessellations. A rate-dependent crystal plasticity framework is utilized to describe the behavior of the ductile ferrite matrix, whereas the brittle martensite phase is characterized using the conventional isotropic J 2 plasticity model. A novel framework combining a ductile phase field fracture formulation with both crystal plasticity and J 2 plasticity is implemented to simulate the accumulation of damage and transgranular crack propagation within each phase. The obtained results are compared and contrasted by observing how the volume fraction, crystallographic orientation, and morphology influence the failure mode of the DP steels.