Nucleation and variant selection during the α–γ–α phase transformation in microalloyed steel

doi:10.1016/j.actamat.2010.11.017

Acta Materialia

Volume 59, Issue 4, February 2011, Pages 1530-1541

https://doi.org/10.1016/j.actamat.2010.11.017 Get rights and content

Abstract

This study addresses variant selection during the α–γ–α phase transformation in recrystallized, low-carbon, microalloyed steel. Since variant selection was found to occur during incipient stages of transformation, the investigation focused mainly on γ-grain nucleation from the α phase. To obtain comprehensive microstructure information, the sample was characterized by high-temperature electron backscatter diffraction in a scanning electron microscope and additional three-dimensional serial sectioning. It is shown that preferred nucleation at a grain boundary near a triple junction requires an orientation relationship close to the Kurdjumov–Sachs (K–S) correspondence to both adjacent α grains (nucleation rule I). Furthermore, a low disorientation of a {1 1 0}_bcc crystallographic plane to the grain boundary plane favors nucleation (nucleation rule II). The results suggest that the nucleation of γ grains is favored at triple junctions by a low surface energy of the nucleus. A corresponding model is proposed. On this basis, variant selection of different ferrite orientations and the product textures of the complete transformation cycle were successfully predicted for a random grain boundary plane distribution.

Introduction

The production of steel sheet involves many processing steps which are accompanied by a change in phase composition, texture and microstructure. These changes strongly influence the material properties. Low-carbon, microalloyed steel at high temperatures is usually hot-rolled in the austenite (γ) regime. This is commonly accompanied by the formation of a pronounced crystallographic texture. During subsequent cooling, the austenite–ferrite phase transformation takes place. This reversible phase transformation causes a change in crystal structure, but is also associated with a dramatic change in microstructure and texture. A review of transformation textures was given by Ray et al. [1]. The crystallography of the phase transformation reveals a specific orientation relationship between austenite (parent phase) and ferrite (product phase). For carbon steels, this crystallographic correspondence typically follows the Kurdjumov–Sachs (K–S) orientation relationship, or close to it [2], [3], [4], [5]. A specific orientation relationship can be used to predict the product texture. The same holds for the back transformation (α–γ). However, experiments show that not all crystallographic variants are generated equally frequently, so that the computed and measured transformation textures do not coincide. This is attributed to variant selection, but the rules and reasons for variant selection in recrystallized materials are not known. As the material properties are strongly influenced by microstructure and texture, a basic understanding of the underlying principles of variant selection is indispensable for texture control during processing.

Owing to experimental difficulties, variant selection has been investigated mostly at room temperature in the ferritic regime after previous deformation of the high-temperature austenite phase. Therefore, respective models had to assume a copper-type rolling or cube-type recrystallization texture of austenite prior to transformation to ferrite. To explain the variant selection of deformed steels, the activity of slip systems [6], [7], the Bain strain [2], [8] or the elastic interaction of transformation events [9] was considered. However, variant selection is also observed for recrystallized austenite (Fig. 1), which is less pronounced than in the deformed state, but distinct and reproducible.

Brückner and Gottstein proposed associating the variant selection in recrystallized steel with the Bain strain and residual stresses in recovered areas [3], but because of limitations in their experimental techniques, these assumptions could not be confirmed. The development of high-temperature orientation imaging, however, opened new avenues for observing phase transformations with spatial resolution and for obtaining information on local orientations. This allowed the mechanisms of phase transformations to be investigated and the underlying physics of variant selection to be revealed.

Most previous investigations of variant selection phenomena pertained to the γ–α phase transformation, whereas little attention was paid to forward transformation. However, since the transformation may occur in both directions during processing, the whole transformation cycle must be investigated [3], [10].

To acquire information on the location and orientation of the product phase with respect to the parent phase, local orientation measurements need to be conducted at the transformation temperature. Therefore, for this study a laser-powered heating stage for a scanning electron microscope was developed which would allow rapid heating and cooling as well as concurrent electron backscatter diffraction (EBSD) measurements. With such a device the whole transformation cycle α → γ and γ → α can be investigated in situ. This study focuses on the α → γ transformation which has been addressed only infrequently so far.

Previous investigations on the same material revealed that the newly developed γ-phase texture did not significantly change with progressing transformation [11]. Accordingly, variant selection must already occur during the nucleation period. Furthermore, it was established that the K–S relationship which was originally developed for the martensitic transformation holds for the α–γ–α phase transformation as well. Typically, a small deviation from the ideal K–S relationship is observed [12].

The goal of this work was to reveal and to analyze all factors that cause a texture change and variant selection during the reversible α–γ–α phase transformation in recrystallized microalloyed steel. The required local information of the parent and product orientation was obtained by high-temperature in situ EBSD and three-dimensional (3D) serial sectioning. These results were analyzed with respect to the underlying mechanisms of variant selection. The relations derived were used to predict the transformation texture. A model is presented to rationalize the observed preference of nucleus orientations.

Section snippets

Material

The material investigated was a commercial hot band of a microalloyed low-carbon steel. The chemical composition is shown in Table 1. Initially, the samples of this material were cold rolled in a reversing manner to 50% thickness reduction. The cold-rolled samples were annealed at 700 °C for 5 h in a vacuum furnace, in order to obtain a stable microstructure for transformation annealing. Grain growth was suppressed during transformation annealing by the presence of stable micro carbo-nitrides.

Transformation texture development

The texture development in the course of the α–γ phase transformation was measured to determine the essential period where variant selection occurred.

To obtain good statistics for texture calculation, especially for the low transformed volume fractions, many samples were investigated by high-temperature in situ EBSD. The γ-phase orientations were binned with respect to their volume fraction (1–10% = 5%, 15–25% = 20%, 35–45% = 40%), and the texture was calculated (Fig. 4).

The results demonstrate that

Nucleation

From previous investigations, it is known [11] that variant selection takes place already during nucleation. The product texture is therefore essentially determined by the nucleation texture. As detailed in this study, the γ grains typically comprised a K–S relationship (with some scatter) to both α grains at a grain boundary next to a triple junction. The allowed deviation from the ideal K–S relationship was commonly in the range 10–15°. Furthermore, in one of the two α grains, a close-packed

Summary

Variant selection during $α \to γ \to α$ phase transformation was investigated in recrystallized low-carbon microalloyed steel. High-temperature orientation imaging by EBSD was applied to determine the location of transformation nuclei and their orientation relationship to the environment. The following results were obtained.

1.
The preferred selection of variants occurs during the nucleation stage of the phase transformation.
2.
Nucleation of the new phase occurs predominantly at triple junctions.
3.
The new phase

Acknowledgements

The authors wish to thank Professors Lasar S. Shvindlerman and John J. Jonas for stimulating discussions. Financial support by the Deutsche Forschungsgemeinschaft (DFG) through grant Go 335/33 is gratefully acknowledged.

References (35)

M. Humbert et al.
Acta Mater
(2002)
P. Bate et al.
Acta Mater
(2000)
S. Kundu et al.
Scripta Mater
(2006)
S. Kundu et al.
Scripta Mater
(2007)
T. Furuhara et al.
Acta Metall Mater
(1991)
J.F. Nie et al.
Scripta Mater
(1998)
T. Furuhara et al.
Mater Sci Eng
(2001)
M. Tong et al.
Acta Mater
(2004)
A.T.W. Kempen et al.
Acta Mater
(2002)
B. Zhao et al.
Acta Mater
(2010)

J.C. Bokros et al.

Acta Metall

(1963)

Y. Higo et al.

Acta Metall

(1974)

R.K. Ray et al.

ISIJ Int

(1994)

G. Kurdjumov et al.

Z Phys

(1930)

G. Brückner et al.

ISIJ Int

(2001)

G. Nolze

Z Metallkunde

(2004)

G. Nolze

Cryst Res Technol

(2006)

Cited by (74)

In-situ heating-stage EBSD validation of algorithms for prior-austenite grain reconstruction in steel
2024, Scripta Materialia
Prior-austenite grain (PAG) structures play a critical role in determining steel microstructures and properties. Over recent years, the use of algorithms that reconstruct PAGs from electron backscatter diffraction (EBSD) measurements of martensitic or bainitic microstructures have become commonplace. These appear to work well, but there has yet to be a comparison made of the reconstructed PAG microstructure to the true PAG microstructure of a steel measured above Ae3 before transformation. Here, in-situ EBSD was used to record the austenite grain structure (at 1000 °C) before transformation, as well as the progress of subsequent martensitic and bainitic transformations. The resulting dataset, which we have made available to the community here: https://doi.org/10.5281/zenodo.8348372, was used to assess of two popular algorithms for PAG reconstruction. They were found to be accurate enough for PAG size measurement and general texture analysis, although the exact morphologies of some PAG and twin boundaries are sometimes lost.
Segregation mechanism of alloying elements at the fcc-Fe/bcc-Fe interface and its effects on carbon diffusion across the boundary
2023, Journal of Physics and Chemistry of Solids
Understanding the structure and energetics of the fcc/bcc phase interface is essential for lightweight design using advanced high strength steels (AHSSs). The present work investigated the solute segregation and its effects on the interfacial properties of fcc-Fe/bcc-Fe interface utilizing the first-principles calculations. The convergence of fcc-Fe (111) and bcc-Fe (011) slabs was examined, and a 7-layer slab was selected to construct the fcc/bcc interface based on K–S relationship. Then, the segregation behavior of various alloying elements at the interface and the consequent impact on the work of adhesion and interfacial energy were explored. The diffusion behavior of carbon atoms across the interface under the effect of segregation was also investigated. The findings provide theoretical guidance for regulating carbon partitioning via interface engineering and optimizing the design of AHSSs.
Interplay between hierarchical microstructure and graded residual stress in a stainless steel fabricated by laser powder bed fusion
2023, Materials Characterization
The intrinsic heat treatment during laser powder bed fusion (LPBF) fabrication process restricts the precise control of three-dimensional size and stabilization of properties, thereby hindering the wide application of LPBF parts. This work unveils the formation mechanism of graded microstructure and graded properties along building direction and interplay between them for the AISI 420 stainless steel containing solid phase transformation. The graded microstructure is mainly characterized by increasing volume fraction of martensitic α' phase and decreasing thickness of martensitic α' lath from bottom substrate to top surface. Martensitic {112} 〈111〉_α′ compound twin and interfacial ω phase exist in the area close to the top surface, while θ-Fe₃C cementite was additionally indexed in the area close to the bottom substrate. The hierarchical microstructural featured by the complementary austenitic γ phase and multi martensitic α' variants were revealed. The martensitic variants selection indeed occurs and 5–12 types of martensitic variants form during LPBF. Graded and hierarchical microstructure is responsible for the evolution of graded micro-hardness and graded residual stress. The reduction in thickness of martensitic laths and increase in volume fraction of martensite lead to the increase of micro-hardness and strengthening effect with the gradient of 3.54 ± 0.16 HV/mm. The coarsened martensitic α' lath and reversed phase transformation from martensitic α' to austenitic γ for area close to bottom substrate counteract the volume shrinkage and trigger partial stress relief.
Effect of crystallographic features on low-temperature fatigue ductile-to-brittle transition for simulated coarse-grained heat-affected zone of bainite steel weld
2023, International Journal of Fatigue
The coarse-grained heat-affected zone (CGHAZ) of bainite steel welds is prone to lose toughness due to welding thermal history. This study aims to reveal the correlation between crystallographic characteristics and fatigue ductile-to-brittle transition (FDBT) for typical microstructures (lath bainite and granular bainite) in the CGHAZ at low temperatures. Compared to the granular bainite, the lath bainite presents retarded growth of fatigue cracking and significantly lowered FDBT temperature (-20℃ vs −40℃). The superior low-temperature fatigue performances of lath bainite are ascribed to the high-fraction high-angle grain boundary formed by its unique variant selection.
Novel reconstruction approaches of austenitic annealing twin boundaries and grain boundaries of ultrafine grained prior austenite
2023, Materials and Design
Gaining access to the parent microstructural information is of great significance to explore transformation mechanism and mechanical properties. As a powerful route, existing automatic reconstruction techniques still have two unsolved issues for lath martensitic steels. The first one is to reconstruct annealing twin boundaries in prior austenite accurately, avoiding the distorted twin boundaries caused by orientation-shared variants on both sides. As for the second one, i.e., reconstruction of ultrafine grained prior austenite, no attempt has been succeeded, simply because the number of martensitic variants inside is insufficient for all current algorithms by now. In this paper, two novel approaches are proposed to overcome these difficulties respectively. For the first one, a three-step algorithm combined with parallel computing is developed to implement the basic reconstruction with much higher efficiency, then a physics-based method using phase field simulation is established to resolve the fake twinning morphology. The second one named as Bi-Adjacent-Block method aims at exploring morphologies (i.e., size and shape) of the ultrafine prior austenite grains (PAGs) but not intending to determine their orientations. To identify the PAG boundaries with higher accuracy, an approximate analytic determination of orientation relationship by using only two adjacent variants is developed for the first time.
Variant selection of α precipitates formed at β triple junctions in titanium alloy
2022, Materials Characterization
Understanding the orientation of the product phase during β to α phase transformation, i.e., the transformation texture is important for optimizing the microstructure and mechanical properties of titanium alloys. As the α phase prefers to nucleate from the triple junction of β grain boundaries (GBs), it is essential to understand the orientation preference, i.e., the variant selection of the α formed at the triple junction (denoted as the α_TJ). Recent work indicated that in theory the α_TJ can keep strict Burgers orientation relationship (BOR) with all the adjacent β grains for some special types of triple junctions. However, the experimental studies are very limited so far. In this work, a number of α precipitates formed at the triple junctions were analyzed. It was shown that at the triple junctions close to the special types, the observed α_TJ obeyed the triple BOR criterion, i.e., tended to keep the BOR with all the adjacent β grains. However, at the triple junctions greatly deviated from the special types, the selection of the α_TJ variants obeyed the double BOR criterion of one GB. Moreover, it was revealed that the special low angle GB, at which a double BOR α can in theory be possible, had no advantage for the selection of α_TJ variants compared to the general high angle GBs forming the triple junction. This work sheds light on understanding the general rule governing the transformation texture in Ti alloys.

View all citing articles on Scopus

View full text

Nucleation and variant selection during the α–γ–α phase transformation in microalloyed steel

Abstract

Introduction

Section snippets

Material

Transformation texture development

Nucleation

Summary

Acknowledgements

Acta Mater

Acta Mater

Scripta Mater

Scripta Mater

Acta Metall Mater

Scripta Mater

Mater Sci Eng

Acta Mater

Acta Mater

Acta Mater

Acta Metall

Acta Metall

ISIJ Int

Z Phys

ISIJ Int

Z Metallkunde

Cryst Res Technol