HSLA steels with different polygonal ferrite and granular bainite contents resulting from two different cooling rates were investigated. Micropillar compression tests, electron channeling contrast imaging (ECCI) and electron backscatter diffraction (EBSD) experiments were performed to reveal microscopic strength differences and their origin. The obtained results indicate that a higher cooling rate caused a smaller granular bainite substructure size and a higher dislocation density for both ferrite and bainite. In addition, the critical resolved shear stress (CRSS) values for both phases were found to be higher for the faster cooling process. This is ascribed to the increased dislocation density for faster cooling rather than the grain size as will be discussed in the manuscript. Interestingly, the macroscopic yield strength can be closely estimated by the CRSS obtained from micropillar compression considering the corresponding phase fractions. The achieved results can be used in future as input variables for crystal plasticity models.
In this work, a computational framework for discrete dislocation dynamics simulations of bcc metals, which naturally accounts for non-Schmid effects on a/2 screw dislocation mobility based on atomistic simulation results, is developed. The model describes a physical basis to explain many experimental observations on bcc metals. The role of dislocation interactions towards plastic flow are studied. A new mechanism responsible for anomalous slip in micrometer sized W pillars is identified.
Abstract Differentiation of granular bainite and polygonal ferrite in high-strength low-alloy (HSLA) steels possesses a significant challenge, where both nanoindentation and chemical analyses do not achieve an adequate phase classification due to the similar mechanical and chemical properties of both constituents. Here, the kernel average misorientation from electron backscatter diffraction (EBSD) was implemented into a Matlab code to differentiate and quantify the microstructural constituents. Correlative electron channeling contrast imaging (ECCI) validated the automated phase classification results and was further employed to investigate the effect of the grain tolerance angle on classification. Moreover, ECCI investigations highlighted that the grain structure of HSLA steels can be subdivided into four grain categories. Each category contained a different nanohardness or substructure size that precluded a nanoindentation-based phase classification. Consequently, the automated EBSD classification approach based on local misorientation achieved a reliable result using a grain tolerance angle of 5°. Graphical abstract
Differentiation of granular bainite and polygonal ferrite in high strength low alloy (HSLA) steels poses a significant challenge, where both nanoindentation and chemical analyses do not achieve an adequate phase classification due to the similar mechanical and chemical properties of both constituents. Here, the kernel average misorientation from electron backscattered diffraction was implemented into an automated code to differentiate and quantify both microstructural constituents. Correlative electron channelling contrast imaging (ECCI) validated the automated phase classification results, and was further employed to investigate the effect of the grain tolerance angle on classification. Moreover, ECCI investigations highlighted that the grain structure of HSLA steels can be subdivided into four distinct grain categories of different granular bainite substructure sizes or orientation gradients within polygonal ferrite grains. Each category contained grains with a different nanohardness, but a continuous nanohardness transition between categories precluded a nanoindentation-based phase classification. Consequently, the automated classification approach based on local misorientation achieved a reliable result by using a grain tolerance angle of 5°, despite the similar mechanical properties of both constituents.
The elbow method and K-means clustering were tested on a dual phase (DP) and high strength low alloy (HSLA) steel to identify the optimal number of clusters and to achieve a phase classification by using hardness and indentation modulus from nanoindentation tests as input variables. The elbow method indicated the optimal number of clusters only for the DP steel. Electron backscatter diffraction (EBSD) measurements were used to correlate the clustering result with the microstructure. The correlation between the calculated clusters and the microstructure showed a good separation of polygonal ferrite, martensite and indents located on grain boundaries or next to martensite islands. On the other hand, K means did not reveal a phase separation of the HSLA constituents after comparing the clustering results with selected grains. After comparing the hardness level of the selected HSLA grains and the minimum distinguishable hardness level at the DP steel, it is assumed that the K-means input variable have to differ from each other by at least 10%. A further sample was heat treated to produce large polygonal ferrite grains. Four selected grain pairs were tested perpendicular to the high angle grain boundaries (HAGB). The goal was to investigate the impact of indentation near grain boundaries between two adjacent grains with similar mechanical properties on the cluster determination and K means clustering. We could show that the optimal number of clusters was not identical for all grain pairs to achieve the best separation. Consequently, we determined for the HSLA steel and its complex microstructure (i) the influence on the measured hardness in the vicinity of grain boundaries, (ii) the similar mechanical properties of both phases and (iii) the inhomogeneous substructure of granular bainite as the limiting factors to obtain a phase classification by using K means clustering based on nanoindentation input variables.
Automated clustering of input nanoindentation data using a K-means algorithm was performed on dual phase (DP) and high strength low alloy (HSLA) steels with the aim of accurate microstructural classification of the phases present. Electron microscopy and diffraction techniques were used to correlate the clustering result with the microstructure, and for the DP steel an accurate separation of polygonal ferrite, martensite and anomalous indents (i.e. located on grain boundaries or adjacent to martensite islands) was achieved. In direct contrast, K-means did not reveal an accurate separation of the HSLA constituent phases. After comparing the hardness level of selected HSLA grains and the minimum distinguishable hardness level of the DP steel, it is found that K-means input variables should differ from each other by at least 10% in order to achieve reasonable separation of clustering results. Further targeted nanoindentation experiments on polygonal ferrite grain pairs were then performed to investigate the impact of indentation near grain boundaries between two adjacent grains with similar mechanical properties on the K-means clustering analysis. It is shown that the optimal number of clusters could not converge, and consequently, for the HSLA steel and its complex microstructure the limiting factors to obtain a phase classification by using K-means clustering based on nanoindentation input variables are: (i) the influence on the measured hardness in the vicinity of grain boundaries; (ii) the similar mechanical properties of both phases; (iii) the inhomogeneous substructure of granular bainite.
The application of an inverse method for determining the parameters of a crystal plasticity constitutive law of a body-centered-cubic (BCC) single phase material is presented. Nanoindentation is used as the primary experimental input. An objective function, based on the deviation between the experimentally measured imprint and the simulated one, is minimized by a differential evolution algorithm to obtain the best fitting crystal plasticity parameters. To aid the identification procedure additional experimental data is used: the upper bounds and the ratios of the critical resolved shear stresses of the three slip plane families in BCC are estimated from micropillar compression experiments and used as a constraint in the optimization. The effect of the imposed constraints and the chosen strategy for mapping experimental to simulated displacements is presented and discussed. The validation of the method is done in a macroscopic regime by comparing an experimental tensile test with a simulated one using the obtained crystal plasticity parameters. Accurate results are achieved from two different indents. Therefore, the method is a promising path for determining crystal plasticity parameters in the case where a direct fitting from a macroscopic stress–strain curve is not possible, i.e. in the case of multi-phase materials.
