Study on the plastic anisotropy of advanced high strength steel sheet: Experiments and microstructure-based crystal plasticity modeling

Highlights

• 3D DIC and microstructure based CP modeling were used to study DP980’s anisotropy.

• r − values exhibit significant directionality and deformation dependence.

• The dual-phase heterogeneous microstructure reduce the overall r − values.

• Lattice rotation and plastic heterogeneity dominate the evolution of anisotropy.

Abstract

For their versatile microstructure controlling techniques, advanced high strength steels (AHSSs) show great potential to be tailored for various application scenarios. The unique microstructure of AHSSs renders their complex plasticity behaviors and mechanical properties. In this work, uniaxial tensile tests combined with a 3D digital image correlation system were carried out to capture the mechanical anisotropy evolution of a cold-rolled dual-phase AHSS (DP980) sheet. In contrast to the conventional body centered cubic (BCC) steel sheets, the studied DP980 sheet shows much smaller Lankford coefficients and a mild strength anisotropy. The Lankford coefficients depend significantly on tensile directions and vary obviously with deformation; they increase with deformation first and then decrease after macroscopic strain localization takes place. Postmortem experimental characterizations as well as microstructure based full-field crystal plasticity simulations were employed to uncover the underneath mechanisms. Plasticity heterogeneities and micromechanical interactions between the soft ferrite matrix and hard martensite islands primarily account for the mechanical anisotropy of the studied steel sheet.

Introduction

The stringent environmental challenge and fuel-efficiency requirement increase the demand of incorporating lightweight metal alloys into automobile industry. Advanced high strength steels (AHSSs), for their superior strength-to-weight ratio, constitute the major lightweight alloys involved in the structural components of automotive products [1,2]. In particular, dual-phase (DP) steels, which consist of a ferrite matrix with hard martensite islands, are widely used to manufacture various critical components such as B-pillar, floor panel tunnel, engine cradle, front sub-frame package tray, etc. [3] The soft ferrite matrix provides excellent ductility, while the hard martensite islands create high work-hardening rate. Such unique binary microstructure renders DP steels large ultimate tensile strength, small yielding-strength ratio and good formability, but also complex micromechanical behavior and plastic anisotropy [4].

Due to the preferential crystallographic texture resulting from prior thermo-mechanical processes, polycrystalline metal sheets generally exhibit mechanical anisotropies in terms of the dependence of elastic modulus, plastic flow, strength, springback, stretch formability, and damage on loading direction [5], [6], [7], [8]. The typical texture components of BCC metals includes the γ ({111}||ND) fiber, i.e., the {111} crystal axis aligned with the normal direction (ND) of metal sheets, and the α (〈110〉||RD) fiber, i.e., the 〈100〉 crystal axis aligned with the rolling direction (RD). The γ fiber, a typical recrystallization texture component, is beneficial to the deep drawing ability of metal sheets while the α fiber generally not [9]. A critical material parameter for characterizing plastic anisotropy is the r − value (Lankford coefficient), i.e., the plastic strain ratio of width to thickness obtained through uniaxial tensile tests. The estimate of r − values is of interest to metallurgists and designers, as the variation of r − values with the uniaxial tensile direction is a first-order parameter in scenarios like sheet metal forming [10,11]. A higher r − value means a smaller reduction of thickness, thus a better normal formability [12].

As an intrinsic and critical feature of metal sheets, plastic anisotropy has been being extensively studied for single-phase steels, such as interstitial free (IF) steel [13], low carbon steel [14], ferrite stainless steel [15,16], etc. These single-phase BCC metals generally have large r − values, e.g., 1.64∼2.43 for IF steel [13], 1.4∼1.8 for low carbon steel [14], and 1.2∼1.8 for ferrite stainless steel [15,16]. For AHSSs, by contrast, their r − values were commonly reported about 0.85∼1.1 [17,18]. Such small r − values mean that it is more challenging to form an AHSS sheet part compared with the conventional mild steels. Yu et al. [7] stated that the edge stretchability degradation of a DP steel sheet is sensitive to the r − value when the mean r − value is below 1.0 but not when it is above 1.0, and cracks are prone to occur along the direction with the smallest r − value. The r − value thus is a critical formability indicator for rolled metal sheets since the deep drawing ability of metal sheet generally correlates with its r − value. Moreover, for they are easily accessible via experiments, r − values obtained from different tensile directions are frequently used to identify the material parameters of various yield functions, such as Hill48 [19,20], Yld2000 [21], Yld2004-18 [22], and HAH [23] criteria, etc. Understanding the plastic anisotropy and calibrating the anisotropy indicator are crucial for optimizing the forming process of AHSS sheets.

Unlike that of single-phase rolled metal sheets, the study on the anisotropic properties of multi-phase steels is still rather insufficient. Previous research demonstrates that the formability and strength of AHSS sheets are distinctly sensitive to loading direction. For instance, Shih et al. [24] observed a higher shear tensile strength in the transverse direction (TD) than that in the RD for four different AHSSs. Mendiguren et al. [25] reported a DP steel with moderate anisotropy; its r − value is about 0.8, which is much smaller than that of traditional deep drawing steel sheets. Li et al. [26] investigated the anisotropic fracture behavior of a DP980 sheet and theorized that the out-of-plane shear fracture strength is 15% lower than the in-plane one. They attributed this diversity to the non-uniformly distributed micro-voids; in the middle layer of the sheet, there are more martensite particles congregating so that more interfaces between the hard martensite islands and the matrix as well as more primary nucleated micro-voids. Thus, the sheet strength is lower in the perpendicular direction [27]. Joo et al. [28] studied the effect of crystallographic texture on the mechanical anisotropy of a hot-rolled multi-phase steel, and found that the steel is inclined to brittle fracture when subjected to tensile deformation. It was attributed to an exaggerated volume of grains with the {100} plane normal to 45°. Recently, Ha et al. [29] reported a V-shape evolution of r − value with respect to deformation for a DP780 sheet under non-proportional loading. The authors ascribed it to the non-uniform deformation resulting from the hard martensitic structures.

Concerning the numerical simulations of AHSS sheets, crystal plasticity (CP) models, which use the crystallographic texture as input, naturally describe the anisotropic behavior of polycrystalline materials and successfully consider the stress/strain partitions among different phases and grains [30]. Furthermore, the state-of-the-art full-field CP modeling, in which each grain is spatially resolved into a large number of material points, is computationally feasible for studying the micro-mechanical behaviors of multi-phase polycrystalline metals. Woo et al. [31] used neutron diffraction and CPFEM to determine the micromechanical hardening parameters of both ferrite and martensite phases in a commercial DP980 steel. Their simulations demonstrated that during uniaxial tension the crystallographic orientation of the ferrite phase significantly affects the strain localization and void initiation in the ferrite matrix adjacent to the martensite islands. Diehl et al. [32] presented a full-field CP simulation based on three dimensional DP microstructure to investigate the strain and stress partitioning among the ferrite and martensite phases in a DP steel. Pagenkopf et al. [33] employed a full-field CPFEM scheme to assess the effect of martensite morphology on the flow stress and the r − value of a DP600 steel under uniaxial tension in different directions. Their study demonstrated that the predictive capability of the modeling scheme strongly depends on the level of detail with which microstructural features are represented in simulations.

In summary, the multi-phase microstructure and the complex micromechanical interaction among phases render the plastic anisotropy and mechanical properties of AHSSs more complex than those of the conventional structural steels. Calibrating the plastic anisotropy and probing the underneath micromechanical mechanisms requires state-of-the-art experimental and simulation approaches, such as full-field strain measurement based on the digital image correlation (DIC) technique and high-resolution CP modeling incorporating the authentic microstructure of materials. In this work, a series of uniaxial tensile tests aligned with different material directions were performed on a cold-rolled DP980 sheet with a commercial 3D high-resolved DIC system; the directionality of mechanical properties as well as the full-field strain distribution in a large region were successfully captured. Microstructure characterizations and full-field CP simulations were jointly employed to uncover the mechanism of the plastic anisotropy of the AHSS.