Grain Boundary Nucleation of Displacive Phase Transformations in Steels

다국어 초록 (Multilingual Abstract)

Austenite Grain Boundaries (AGBs) are known to be the most preferable nucleation site for reconstructive ferrite transformation due to the presence of excess free-volume which provides a higher probability for satisfying the local atomic configuration appropriate for the nucleation by random phase fluctuations. On the other hand, it has also been reported that nucleation of displacive phase transformations, i.e. both the isothermal bainitic transformation and the athermal martensitic transformation, preferentially occurred at prior AGBs. However, no clear mechanism for the preferred nucleation of displacive phase transformations at AGBs has been successfully postulated. This is primarily due to the difficulties in experimental demonstration. For example, crystallographic features of a nucleation of bainitic ferrite at AGBs cannot be directly investigated in low-carbon, lean-alloyed steels due to the presence of martensitic transformation products which hinder the precise determination of the crystallographic orientation of the prior austenite. On the other hand, it is well known that both the bainite and martensite hold specific orientation relationships (OR) with respect to austenite, and that the transformations are confined to occur within a single austenite grain. These characteristic features of displacive transformation make it possible to determine the orientation of prior austenite grain and the structure of prior AGBs from the orientation distribution of transformed products measured by the Electron Back Scattered Diffraction (EBSD) technique.

In the present study, the nucleation behavior of upper bainite in low-carbon, lean-alloyed steel was investigated by Scanning Electron Microscopy (SEM)-EBSD. It is shown that the bainitic ferrite nucleated at AGBs having the crystallographic orientation of a variant with a Kurdjumov-Sachs (K-S) OR with respect to one of the neighboring austenite grain. This variant selection rule indicates that the bainitic ferrite nucleation is enhanced by interfacial energy reduction as the initial AGBs are converted to low energy inter-phase boundaries (IPBs) in the process.

As the nucleation of upper bainite is controlled by an interfacial energy reduction mechanism, the transformation behavior of upper bainite is strongly affected by B addition. It was observed that the addition of B clearly affected both the transformation kinetics of isothermal bainitic transformation and the morphology of the bainite. The addition of B increased the incubation time for the formation of a detectable amount of transformation product, and reduced the overall transformation kinetics. The bainite microstructure consisted of a bainitic ferrite matrix and the Martensite/Austenite (M/A) constituent. The M/A constituent had an elongated morphology in B-free steel, whereas the M/A constituent in B-added steel had a granular structure. In B-free steel, bainitic ferrite nucleated homogeneously within the austenite grains, and the overall transformation was controlled by the increase of transformed area within the initial austenite grain. In contrast, in B-added steel, the transformation was confined to the interior of the austenite grains, and the overall transformation kinetics was controlled by an increase in the number of the locally transformed grains. The variant of bainitic ferrite in B-added steel was selected within a group of K-S variants related to the same Bain variant. The characteristic bainite microstructure in B-added steel is therefore due to the inhibition of the bainitic ferrite nucleation at ABGs.

The microstructure and crystallography of BCC α’-martensite formed in a sensitized AISI 304 stainless steel was also studied in detail by means of Transmission Electron Microscopy convergent beam Kikuchi line pattern analysis. It was suggested that GBs containing Intrinsic Grain Boundary Dislocation (IGBD) might act as preferential nucleation site according to a faulting mechanism of martensitic nucleation proposed by Olson and Cohen.
Different from the nucleation behavior of an upper bainite in low carbon, lean-alloyed steels, the martensite variants do not appear to be selected to achieve interfacial energy reduction, or the easy accommodation of transformation strain by slip in the neighboring austenite grain. This implies that both the martensite and the upper bainite preferentially nucleate at prior AGBs, but the operating nucleation mechanisms are totally different.

From the observations that interfacial energy reduction mechanism plays a pivotal role in the nucleation of upper bainite, specific alloy designs were made to produce ultra-high strength (UHS) martensitic grades in conventional continuous galvanizing lines. The thermal treatments related to the hot dip galvanizing and galvannealing processes needed to obtain a Zn or Zn-alloy coated martensitic steel did not degrade the mechanical properties significantly when the coating process was applied to the un-transformed austenite phase. Therefore, it is shown that martensitic steel designed according to the critical physical metallurgy principles can be successfully processed in galvanizing lines not originally intended to produce UHS grades.

목차 (Table of Contents)

• Abstract I
• Contents IV
• Chapter I. Introduction 1
• 1.1 General Introduction 2
• 1.1.1 Ultra-high Strength Steels for automotive application 2
• 1.1.2 Scope and objective of present study 4
• 1.2 Martensite crystallography 6
• 1.3 Grain boundaries and Grain boundary dislocations 14
• References 21
• Chapter II. Experimental Procedures 24
• 2.1 Steel chemical compositions 25
• 2.2 Dilatometry 28
• 2.3 Mechanical testing 28
• 2.4 Microstructure Characterization 29
• 2.4.1 Scanning Electron Microscopy and Electron Back Scattered Diffraction technique 29
• 2.4.2 Transmission Electron Microscopy 29
• 2.4.3 Secondary Ion Mass Spectrometry 30
• 2.5 X-ray Diffraction 30
• 2.6 Crystal orientation Determination 31
• 2.6.1 Determination of crystal orientation of the prior austenite by EBSD 31
• 2.6.2 Orientation determination by Convergent Beam Kikuchi Pattern analysis 34
• References 39
• Chapter III. Processing of Martensitic steels in Continuous Galvanizing Lines
• Part 1: Tempered martensite 41
• 3.1 Introduction 42
• 3.2 Experimental procedures 43
• 3.3 Results 46
• 3.3.1 Tensile testing 46
• 3.3.2 Fracture surface 49
• 3.3.3 Microstructural analysis 50
• 3.3.3.1 SEM micrographs 50
• 3.3.3.2 XRD Results 52
• 3.3.3.3 TEM micrographs 54
• 3.3.4 Internal Friction 61
• 3.4 Discussion 66
• 3.4.1 Microstructure evolution 66
• 3.4.2 Mechanical properties 69
• 3.5 Conclusions 76
• References 78
• Chapter IV. Processing of Martensitic steels in Continuous Galvanizing Lines
• Part 2: Bainite-Martensite steel 82
• 4.1 Introduction 83
• 4.2 Experimental procedures 85
• 4.3 Results 88
• 4.3.1 Continuous Cooling Transformation Behavior 88
• 4.3.2 Isothermal Transformation Behavior 91
• 4.3.3 GI/GA cycle simulation 94
• 4.3.3.1 Dilatometry and SEM 94
• 4.3.3.2 EBSD and TEM Analysis of Direct cooled, GI and GA specimens 97
• 4.3.3.3 Tensile Properties 107
• 4.4 Discussion 109
• 4.4.1 Structure and property of bainite formed during GI and GA 109
• 4.4.2 Effects of bainite transformation on the strength of mixed microstructure 111
• 4.5 Conclusions 117
• References 120
• Chapter V. Effect of Boron on the isothermal Bainitic transformation 123
• 5.1 Introduction 124
• 5.2 Experimental procedures 126
• 5.3 Results 128
• 5.3.1 Elemental B Distribution 128
• 5.3.2 Isothermal transformation kinetics 129
• 5.3.3 Microstructure of the 420ºC and 500ºC transformation product 131
• 5.3.4 EBSD analysis of microstructures developed in the initial stage of the 500ºC transformation 140
• 5.4 Discussion 148
• 5.4.1 Grain boundary nucleation and bainitic ferrite variant selection 148
• 5.4.2 Effect of B on the of bainitic transformation kinetics 151
• 5.4.3 Effect of B on the morphologies of bainitic 155
• 5.5 Conclusions 163
• References 165
• Chapter VI. Martensite Nucleation at Grain Boundaries containing Intrinsic Grain Boundary Dislocations 169
• 6.1 Introduction 170
• 6.2 Experimental procedures 174
• 6.3 Results and discussion 176
• 6.3.1 EBSD Study and applicability for Brandon’s criterion 176
• 6.3.2 TEM Analysis of Martensite nucleated at grain boundary 179
• 6.4 Conclusions 194
• References 195
• Chapter VII. General Conclusion 197
• Appendix. Orientation of Crystallite 203
• A.1 Matrix representation of vector 204
• A.2 Coordinate transformation 205
• A.3 Misorientation and rotation axis 207
• A.4 Symmetry operation 208
• References 211
• Acknowledgements 212
• Publications 214
• Curriculum Vitae 215