Pearlitic steel has been widely used for engineering and industrial applications such as bridge cable, tire cord and rail steel because of its excellent strength and ductility. Pearlitic steel is composed of alternating body-centered cubic ferrite (α-Fe) and orthorhombic cementite (Fe3C) layer with specific crystallographic orientation relationships. Many researchers have studied on the lamellar structure of pearlitic steel to explain the origin of its excellent mechanical properties. Most researchers just assumed that the ferrite/cementite interface is merely an obstacle to lattice dislocation and they explained the reason of the high strength of pearlitic steel is lattice dislocation pile-up at the ferrite/cementite interface. Unfortunately, the studies underestimated the roles of the interface during plastic deformation in pearlitic steel. Therefore, in this study, the various roles of the ferrite/cementite interface were investigated to understand the relationship between the mechanical properties of pearlitic steel and interface structure using atomistic simulation.
Pearlitic steel is formed by the diffusive eutectoid reaction from parent austenite into resultant ferrite and cementite phase. The ferrite/cementite interface has a specific crystallographic orientation relationship and its habit plane to minimize the energy barrier to phase transformation. In previous experimental observations, IS (Isaichev), BA (Bagaryatsky) and PP (Pitsch-Petch) orientation relationships in pearlitic steel have been consistently reported. According to a recent report, however, IS, Near BA and Near PP orientation relationships are true orientation relationships in pearlitic steel because the angular resolution of experimental equipment was 3° to 5° at that time. Nevertheless, theoretical and experimental studies using BA and PP orientation relationships are still being carried out to explain the phase transformation and mechanical behavior of pearlitic steel. Therefore, in this study, the atomic structure of the ferrite/cementite bilayer for each orientation relationship was modeled and interface energy was computed to verify the existence of each orientation relationship. From the interface energies, we concluded that Near BA and Near PP orientation relationships are thermodynamically more favorable than BA and PP orientation relationships. This supports experimental results which exclude the existence of BA and PP orientation relationships based on accurate crystallographic measurements in the literature.
The heterophase boundary, which has a specific crystallographic orientation relationship, generally forms a semi-coherent interface. The semi-coherent interface consists of misfit dislocations and a coherent region. The characteristics of misfit dislocations developed on the semi-coherent interface are changed by the crystallographic orientation relationship of the heterophase boundary. Likewise, the characteristics of misfit dislocations developed on the ferrite/cementite interface will greatly vary depending on a given crystallographic orientation relationship and its habit plane. It is expected that the characteristics of misfit dislocation at the FCI may have a great influence on the mechanical properties of pearlitic steel composed of a fine lamellar structure because the interface fraction becomes large. In this study, we performed xAIFB (extended atomically informed Frank-Bilby) and disregistry analysis to characterize the characteristics of misfit dislocation for each orientation relationship, such as Burgers vector, core-width, line orientation and line spacing.
According to the literature, an extremely large number of dislocation is observed near the ferrite/cementite interface, but only a small number of dislocation is observed in the ferrite layer. This is because the lattice dislocation trapping mechanism is activated when interface shear occurs by the stress field induced by the lattice dislocation. It is an important unit process to determine the strength of multi-layered metallic composite for a few nanometers of lamellar thickness. So, in-plane shear response of the ferrite/cementite interface was investigated to estimate the effect of misfit dislocations on the interfacial shear strength of the ferrite/cementite. For this purpose, we carried out the in-plane shear deformation using ferrite/cementite bilayer model for IS, Near BA and Near PP orientation relationships. The simulation results reveal that the ferrite/cementite interface for IS and Near BA orientation relationship show dislocation-mediated plasticity for all in-plane shear directions except two directions and the ferrite/cementite interface for Near PP orientation relationship shows mode II (in-plane shear) fracture for all in-plane shear directions. From the disregistry analysis results, we concluded that the in-plane shear behavior of the ferrite/cementite interface is governed by bond strength between interface atoms, Burgers vector and core-width of misfit dislocation.
Based on the results of the in-plane shear deformation of the ferrite/cementite interface, the lattice dislocation trapping mechanism was investigated using the ferrite/cementite bilayer model for IS orientation relationship which has the lowest interfacial shear strength. For this purpose, a straight lattice dislocation paralleled to interface was implemented in the ferrite layer. After that, an equilibration process was carried out to observe the motion of the lattice dislocation. From the equilibration process, we observed that the lattice dislocation moves to the interface when the lattice dislocation is located at a vertical distance of 1.3 nm or less under the misfit dislocation core, and the lattice dislocation moves to the interface when the lattice dislocation is located at a vertical distance of 4.1 nm or less under the middle of two misfit dislocations. In order to analyze the motion of the lattice dislocation, the spatial stress distribution induced by the arrays of misfit dislocations and the lattice dislocation was computed, and the P-K (Peach-Kohler) force applied to the lattice dislocation for each position was calculated using the spatial stress distribution. Moreover, we performed the 2D disregistry analysis to analyze the interface shear caused by the stress field induced by the lattice dislocation. From the analysis, we concluded that the motion of the lattice dislocation located farther than 1.3 nm from the interface is governed by the stress field induced by the misfit dislocations and lattice dislocation, while the motion of the lattice dislocation located closer than 1.3 nm from the interface is governed by the image force induced by interface shear.
According to the previous studies, the ferrite/cementite interfaces act as nucleation sites for lattice dislocation in pearlitic steel. However, there is no prior study on the effect of the characteristics of misfit dislocation on the nucleation mechanism of the lattice dislocation at the ferrite/cementite interface. So, we carried out the atomistic simulation to investigate the effect of the misfit dislocation on the lattice dislocation nucleation at the ferrite/cementite interface for each orientation relationship. In order to study the nucleation mechanism, we performed tensile simulation using ferrite/cementite sandwich structure for IS and Near BA orientation relationship to induce the nucleation of the lattice dislocation at the ferrite/cementite interface. The simulation results show that the nucleation of the lattice dislocation occurs at the core region of the misfit dislocation when the applied stress reaches the critical stress. In order to analyze the lattice dislocation nucleation at the ferrite/cementite interface, we utilized the nucleation criteria proposed by several authors. The nucleation criteria was computed using resolved shear stress and geometric compatibility between slip system and misfit dislocation. From the nucleation criteria, we found that it does not predict the slip system and nucleation site. It is because the core-width of the misfit dislocation for IS and Near BA orientation relationship is extremely wide. Therefore, we concluded that the measure of geometric compatibility between slip plane traces and misfit dislocation line is inappropriate, and it is necessary to refine the geometric compatibility based on the core-width of misfit dislocation.