Search

International journal of minerals, metallurgy and materials 29(10), 1901 - 1907 (2022). doi:10.1007/s12613-022-2440-5, Steel production causes a third of all industrial CO2 emissions due to the use of carbon-based substances as reductants for iron ores, making it a key driver of global warming. Therefore, research efforts aim to replace these reductants with sustainably produced hydrogen. Hydrogen-based direct reduction (HyDR) is an attractive processing technology, given that direct reduction (DR) furnaces are routinely operated in the steel industry but with CH$_4$ or CO as reductants. Hydrogen diffuses considerably faster through shaft-furnace pellet agglomerates than carbon-based reductants. However, the net reduction kinetics in HyDR remains extremely sluggish for high-quantity steel production, and the hydrogen consumption exceeds the stoichiometrically required amount substantially. Thus, the present study focused on the improved understanding of the influence of spatial gradients, morphology, and internal microstructures of ore pellets on reduction efficiency and metallization during HyDR. For this purpose, commercial DR pellets were investigated using synchrotron high-energy X-ray diffraction and electron microscopy in conjunction with electron backscatter diffraction and chemical probing. Revealing the interplay of different phases with internal interfaces, free surfaces, and associated nucleation and growth mechanisms provides a basis for developing tailored ore pellets that are highly suited for a fast and efficient HyDR., Published by Springer, Bei jing shi

The mechanical properties of metallic materials often significantly deteriorate when used under harsh conditions, particularly at cryogenic temperatures. Yet, safe and lasting low-temperature infrastructures are needed to realize the growing global demand for liquid energy carriers such as hydrogen[1]. Traditional microstructure strategies designed for high strength and ductility at room temperature often fail when exposed to such low temperatures due to the unresolved conflict between strength and damage tolerance under such conditions[2, 3]. Here we introduce a new type of dual-scale atomic ordering nanostructure embedded in a massive metallic solid solution matrix to overcome this challenge. This reconciles advantages of two antagonistic mechanisms in one material concept, namely, a well-balanced enthalpy-driven ordering on the one hand and entropy-driven solid solution on the other. We realize the approach in form of an exceptionally high number density of coexisting sub-nanoscale chemical short-range ordered (SRO, ~2.4×10^26 m-3) and nanoscale long-range ordered (NLRO, ~4.5×10^25 m-3) zones in a CoNiV-based compositionally complex alloy. We observe several types of interactions between the chemical ordering and the dislocations, i.e., the defects that carry both, the material’s inelastic deformation and potentially also its damage initiation. We observe an ordering-induced increase in dislocation shear stress (providing strength) as well as a more rapid dislocation multiplication due to the frequent blocking of the planar dislocation flux by NLRO and the associated generation of new dislocations (providing ductility). The latter effect also releases stress concentrations at NLRO obstacles that otherwise would promote damage initiation and failure (a phenomenon observed in materials with larger ordered zones and precipitates). Consequently, a higher strength-ductility combination (strength-elongation product 76.5 GPa% with 1172 MPa yield strength at 87 K) is achieved in comparison to reference materials devoid of such ordering hierarchy (yield strength 794 MPa), containing only SRO (yield strength 931 MPa) or coherent precipitates of a few tens of nanometers (strength-elongation product 38.8 GPa%). Our results thus reveal the effects of different types (and scales) of coexisting chemical ordering on mechanical properties of complex alloys and provide guidelines to control them, which serves as a new design approach for strong and tough materials particularly for cryogenic applications.

The design of high performance structural materials is always pursuing combinations of excellent yet often mutually exclusive properties such as mechanical strength, ductility and thermal stability. Although crystal-glass composite alloys provide better ductility compared to fully amorphous alloys, their thermal stability is poor, due to heterogeneous nucleation at the crystal-glass interface. Here we present a new strategy to develop thermally stable, ultrastrong and deformable crystal-glass nanocomposites through a thermodynamically guided alloy design approach, which mimics the mutual stabilization principle known from symbiotic ecosystems. We realized this in form of a model Cr-Co-Ni (crystalline)/Ti-Zr-Nb-Hf-Cr-Co-Ni (amorphous) laminate composite alloy. The symbiotic alloy has an ultrahigh compressive yield strength of 3.6 GPa and large homogeneous deformation of ∼15% strain at ambient temperature, values which surpass those of conventional metallic glasses and nanolaminate alloys. Furthermore, the alloy exhibits ∼200 K higher crystallization temperature (TX > 973 K) compared to that of the original TiZrNbHf-based amorphous phase. The elemental partitioning among adjacent amorphous and crystalline phases leads to their mutual thermodynamic and mechanical stabilization, opening up a new symbiotic approach for stable, strong and ductile materials.

Hydrogen embrittlement (HE) is one of the most dangerous yet most elusive embrittlement problems in metallic materials. Advanced high-strength steels (AHSS) are particularly prone to HE, as evidenced by the serious degradation of their load-bearing capacity with the presence of typically only a few parts-per-million H. This strongly impedes their further development and application and could set an abrupt halt for the weight reduction strategies pursued globally in the automotive industry. It is thus important to understand the HE mechanisms in this material class, in order to develop effective H-resistant strategies. Here, we review the related research in this field, with the purpose to highlight the recent progress, and more importantly, the current challenges toward understanding the fundamental HE mechanisms in modern AHSS. The review starts with a brief introduction of current HE models, followed by an overview of the state-of-the-art micromechanical testing techniques dedicated for HE study. Finally, the reported HE phenomena in different types of AHSS are critically reviewed. Focuses are particularly placed on two representative multiphase steels, i.e., ferrite–martensite dual-phase steels and ferrite–austenite medium-Mn steels, with the aim to highlight the multiple dimensions of complexity of HE mechanisms in complex AHSS. Based on this, open scientific questions and the critical challenges in this field are discussed to guide future research efforts.

The combination of different phase constituents to realize a mechanical composite effect for superior strength-ductility synergy has become an important strategy in microstructure design in advanced high-strength steels. Introducing multiple phases in the microstructure essentially produces a large number of phase boundaries. Such hetero-interfaces affect the materials in various aspects such as dislocation activity and damage formation. However, it remains a question whether the characteristics of phase boundaries, such as their chemical decoration states, would also have an impact on the mechanical behavior in multiphase steels. Here we reveal a phase boundary segregation-induced strengthening effect in ultrafine-grained duplex medium-Mn steels. We found that the carbon segregation at ferrite-austenite phase boundaries can be manipulated by adjusting the cooling conditions after intercritical annealing. Such phase boundary segregation in the investigated steels resulted in a yield strength enhancement by 100–120 MPa and simultaneously promoted discontinuous yielding. The sharp carbon segregation at the phase boundaries impeded interfacial dislocation emission, thus increasing the stress required to activate such dislocation nucleation process and initiate plastic deformation. This observation suggests that the enrichment of carbon at the phase boundaries can enhance the energy barrier for dislocation emission, which provides a favorable condition for plastic flow avalanches and thus discontinuous yielding. These findings extend the current understanding of the yielding behavior in medium-Mn steels, and more importantly, shed light on utilizing and manipulating phase boundary segregation to improve the mechanical performance of multiphase metallic materials.

High-strength Al-Zn-Mg-Cu alloys are highly susceptible to stress corrosion cracking (SCC) which severely limits their lifetime. Reversion and re-aging (RRA) temper provides a higher SCC resistance at no loss in strength, yet the microstructural origins of these enhanced properties remain elusive. In an Al-Zn-Mg-Cu alloy, we show that the fine precipitate dispersion in the grain interiors is similar in the peak aged and RRA tempers. However, upon RRA, precipitates inside the grains are enriched in Cu, lowering the Cu matrix content, and reducing the relative difference in the Cu precipitate composition between bulk and grain boundaries. This study enriches the current understanding on the critical role of Cu related to SCC resistance in Al-Zn-Mg-Cu alloys.

The formation of κ-carbides in austenite Fe-30Mn-9Al-1.2C (wt. %) lightweight steels is tuned via alloying of Si (0, 1, 2 wt. %), an element that can remarkably raise the activities of Al and C based on thermodynamic calculations. Ordered L12 nano-domains (with a size

Strain partitioning and localization were investigated in a high-Mn steel (17.1 wt.% Mn) during tensile testing by a correlative probing approach including in-situ synchrotron X-ray diffraction, micro- digital image correlation (μ-DIC) and electron microscopy. By combining Warren's theory with the μ-DIC analysis, we monitored the formation of planar faults (stacking faults and mechanical twins) and correlated them with the local strain partitioning behavior within the microstructure. Starting with an initial microstructure of austenite (γ) and athermally formed e- and α’-martensite, strain accumulates preferentially near the γ/e interfaces during tensile straining. The local microscopic von Mises strain (evM) maps obtained from μ-DIC probing show that these local strain gradients produce local strain peaks approximately twice as high as the imposed macroscopic engineering strain (e), thus locally triggering formation of e-martensite already at early yielding. The interior of the remaining austenite, without such interfacial strain peaks, remained nearly devoid of planar faults. The local strain-driven growth of the e-domains occurs concomitantly with the α’-martensite formation. At intermediate macroscopic applied strains, austenite grain size is considerably reduced to a few nanometers and the associated γ/e interfacial microscopic strain peaks increase in magnitude. This scenario favors twinning to emerge as a competing strain hardening mechanism at engineering strain levels from e = 0.075 onwards. At large tensile strains, the γ → e → α’ transformation rates tend to cease making both twinning and SFs formation to operate as the main strain hardening mechanisms. The findings shed light on the transformation micro-mechanisms in multiphase Mn-TRIP steels by revealing how strain localization among the constituents can directly influence the kinetics of the competing strain hardening mechanisms.

Superplasticity proceeds from fine-grained structures and requires high intrinsic resistance to grain growth at the deformation temperature. Here, we show that a mixture of two kinds of brittle ordered phases enables superplastic behavior through dynamic recrystallization in an equiatomic medium-entropy VCoNi alloy as a model material. The alloy annealed at 900 °C exhibits a face-centered-cubic single phase. However, in-depth characterization at various length scales reveals that the alloy, when annealed at 800 °C, comprises two ordered phases: κ (Co1.2Ni1.3V) and σ (Co1.5NiV2.5). As a result of the conventional cold-rolling/annealing process and with the aid of an underlying eutectoid reaction, the alloy exhibits a duplex structure with an average grain size of less than 1 μm, i.e. microduplex structure. The size, morphology, and crystallographic orientation do not substantially change during static isothermal holding at 800 °C, which implies a very high resistance to grain growth. With tensile deformation at 800 °C, however, both phases develop into an equiaxed microstructure with low dislocation density and a dramatic change occurs in the crystallographic texture of the κ phase. These variations result from dynamic recrystallization (DRX), which leads to superplastic elongations of 330–450% at 700–800 °C and at strain rates ranging from 10 to 4 to 10−2 s−1. Notably, the superplastic behavior is favorable at the high strain rate due to the enhanced DRX activity, leading to the larger elongation with increasing strain rates. However, deformation-enhanced grain growth occurs concomitantly with dynamic recrystallization; these competitive processes are investigated to elucidate the mechanism of superplasticity in this model material.

We applied two types of hot-rolling direct quenching and partitioning (HDQ&P) schemes to a low-C low-Si Al-added steel and obtained two ferrite-containing TRIP-assisted steels with different hard matrix structures, viz, martensite or bainite. Using quasi in-situ tensile tests combined with high-resolution electron back-scattered diffraction (EBSD) and microscopic digital image correlation (µ-DIC) analysis, we quantitatively investigated the TRIP effect and strain partitioning in the two steels and explored the influence of the strain partitioning between the soft and hard matrix structures on the TRIP effect. We also performed an atomic-scale analysis of the carbon partitioning among the different phases using atom probe tomography (APT). The results show that the strain mainly localizes in the ferrite in both types of materials. For the steel with a martensitic hard-matrix, a strong strain contrast exists between ferrite and martensite, with the local strain difference reaching up to about 75% at a global strain of 12.5%. Strain localization bands initiated in the ferrite rarely cross the ferrite/martensite interfaces. The low local strain (2%–10%) in the martensite regions leads to a slight TRIP effect with a transformation ratio of the retained austenite of about 7.5%. However, for the steel with bainitic matrix, the ferrite and bainite undergo more homogeneous strain partitioning, with an average local strain in ferrite and bainite of 15% and 8%, respectively, at a global strain of 12.5%. The strain localization bands originating in the ferrite can cross the ferrite/bainite (F/B) interfaces and increase the local strain in the bainite regions, resulting in an efficient TRIP effect. In that case the transformation ratio of the retained austenite is about 41%. The lower hardness difference between the ferrite and bainite of about 178 HV, compared with that between the ferrite and martensite of about 256 HV, leads to a lower strain contrast at the ferrite/bainite interfaces, thus retarding interfacial fracture. Further microstructure design for TRIP effect optimization should particularly focus on adjusting the strength contrast among the matrix structures and tuning strain partitioning to enhance the local strain partitioning into the retained austenite.

The present study contends with the room temperature microstructural response of a non-equiatomic metastable high entropy alloy to the elastic cyclic deformation. The stress and strain-induced martensite formation and reversion are recognized as the main microstructural evolutions which are directly correlated with the reversibility of dislocation movement. Two different patterns of reversion for deformation driven epsilon martensite are identified. Full reversion of stress-induced epsilon martensite results in development of nano-twined matrix, the various aspects of which have been described through a dislocation-based model. The strain-induced martensite also goes through partial reversion leading to lath fragmentation which in-turn significantly influences the martensite stability. Interestingly, the presence of a well-developed dislocation substructure is characterized within the martensite bands, which seems to be phenomenal owing to the low imposed strain, low temperature and low stacking fault energy of the experimented material. The development of vein and wall/channel structures is justified through proposing a conceptual based model regarding the interaction of the stacking faults and subsequent generation of the perfect dislocations.

The risk of hydrogen embrittlement (HE) is currently one important factor impeding the use of medium Mn steels. However, knowledge about HE in these materials is sparse. Their multiphase microstructure with highly variable phase conditions (e.g. fraction, percolation and dislocation density) and the feature of deformation-driven phase transformation render systematic studies of HE mechanisms challenging. Here we investigate two austenite-ferrite medium Mn steel samples with very different phase characteristics. The first one has a ferritic matrix (~74 vol.% ferrite) with embedded austenite and a high dislocation density (~1014 m−2) in ferrite. The second one has a well recrystallized microstructure consisting of an austenitic matrix (~59 vol.% austenite) and embedded ferrite. We observe that the two types of microstructures show very different response to HE, due to fundamental differences between the HE micromechanisms acting in them. The influence of H in the first type of microstructure is explained by the H-enhanced local plastic flow in ferrite and the resulting increased strain incompatibility between ferrite and the adjacent phase mixture of austenite and strain-induced α'-martensite. In the second type of microstructure, the dominant role of H lies in its decohesion effect on phase and grain boundaries, due to the initially trapped H at the interfaces and subsequent H migration driven by deformation-induced austenite-to-martensite transformation. The fundamental change in the prevalent HE mechanisms between these two microstructures is related to the spatial distribution of H within them. This observation provides significant insights for future microstructural design towards higher HE resistance of high-strength steels.

High-entropy alloys are solid solutions of multiple principal elements, capable of reaching composition and property regimes inaccessible for dilute materials. Discovering those with valuable properties, however, too often relies on serendipity, as thermodynamic alloy design rules alone often fail in high-dimensional composition spaces. We propose an active-learning strategy to accelerate the design of high-entropy Invar alloys in a practically infinite compositional space, based on very sparse data. Our approach works as a closed-loop, integrating machine learning with density-functional theory, thermodynamic calculations, and experiments. After processing and characterizing 17 new alloys out of millions of possible compositions, we identified 2 high-entropy Invar alloys with extremely low thermal expansion coefficients around 2×10-6 K-1 at 300 K. We believe this to be a different pathway for the fast and automated discovery of high-entropy alloys with attractive optimal thermal, magnetic and electrical properties.High-entropy alloys are solid solutions of multiple principal elements, capable of reaching composition and property regimes inaccessible for dilute materials. Discovering those with valuable properties, however, too often relies on serendipity, as thermodynamic alloy design rules alone often fail in high-dimensional composition spaces. We propose an active-learning strategy to accelerate the design of high-entropy Invar alloys in a practically infinite compositional space, based on very sparse data. Our approach works as a closed-loop, integrating machine learning with density-functional theory, thermodynamic calculations, and experiments. After processing and characterizing 17 new alloys out of millions of possible compositions, we identified 2 high-entropy Invar alloys with extremely low thermal expansion coefficients around 2×10-6 K-1 at 300 K. We believe this to be a different pathway for the fast and automated discovery of high-entropy alloys with attractive optimal thermal, magnetic and electrical properties.

Deformation-induced martensitic transformation (DIMT) has been used for designing high-performance alloys to prevent structural failure under static loads. Its effectiveness against fatigue, however, is unclear. This limits the application of DIMT for parts that are exposed to variable loads, although such scenarios are the rule and not the exception for structural failure. Here we reveal the dual role of DIMT in fatigue crack growth through in situ observations. Two antagonistic fatigue mechanisms mediated by DIMT are identified, namely, transformation-mediated crack arresting, which prevents crack growth, and transformation-mediated crack coalescence, which promotes crack growth. Both mechanisms are due to the hardness and brittleness of martensite as a transformation product, rather than to the actual transformation process itself. In fatigue crack growth, the prevalence of one mechanism over the other critically depends on the crack size and the mechanical stability of the parent austenite phase. Elucidating the two mechanisms and their interplay allows for the microstructure design and safe use of metastable alloys that experience fatigue loads. The findings also generally reveal how metastable alloy microstructures must be designed for materials to be fatigue-resistant.

The third-generation advanced high strength medium manganese (3–12 wt.%) steels typically consist of ultrafine-grained dual-phase (austenite-ferrite) microstructure, obtained through the intercritical annealing of martensite at temperatures typically ≤ 0.5Tmelt, where the bulk diffusion of Mn is extremely slow. Yet, the manganese partitioning plays a prominent role in the austenite growth from the martensitic matrix during this annealing step. Therefore, the ‘short circuit’ diffusion paths provided by grain boundaries (GBs) and dislocations must be crucial to the austenite growth. However, this influence is not well understood across the literature. In the present work, we study the mechanisms of austenite growth in a cold-rolled intercritically annealed medium manganese steel of composition Fe-10Mn-0.05C–1.5Al (wt.%). We provide evidence of manganese transport to austenite through GB diffusion, GB migration and dislocation pipe diffusion. Furthermore, the influence of GB misorientation on austenite growth is also reported.

Ever more stringent regulations on greenhouse gas emissions from transportation motivate efforts to revisit materials used for vehicles1. High-strength aluminium alloys often used in aircrafts could help reduce the weight of automobiles, but are susceptible to environmental degradation2,3. Hydrogen ‘embrittlement’ is often indicated as the main culprit4; however, the exact mechanisms underpinning failure are not precisely known: atomic-scale analysis of H inside an alloy remains a challenge, and this prevents deploying alloy design strategies to enhance the durability of the materials. Here we performed near-atomic-scale analysis of H trapped in second-phase particles and at grain boundaries in a high-strength 7xxx Al alloy. We used these observations to guide atomistic ab initio calculations, which show that the co-segregation of alloying elements and H favours grain boundary decohesion, and the strong partitioning of H into the second-phase particles removes solute H from the matrix, hence preventing H embrittlement. Our insights further advance the mechanistic understanding of H-assisted embrittlement in Al alloys, emphasizing the role of H traps in minimizing cracking and guiding new alloy design.

Fossil-free ironmaking is indispensable for reducing massive anthropogenic CO2 emissions in the steel industry. Hydrogen-based direct reduction (HyDR) is among the most attractive solutions for green ironmaking, with high technology readiness. The underlying mechanisms governing this process are characterized by a complex interaction of several chemical (phase transformations), physical (transport), and mechanical (stresses) phenomena. Their interplay leads to rich microstructures, characterized by a hierarchy of defects ranging across several orders of magnitude in length, including vacancies, dislocations, internal interfaces, and free surfaces in the form of cracks and pores. These defects can all act as reaction, nucleation, and diffusion sites, shaping the overall reduction kinetics. A clear understanding of the roles and interactions of these dynamically-evolving nano-/microstructure features is missing. Gaining better insights into these effects could enable improved access to the microstructure-based design of more efficient HyDR methods, with potentially high impact on the urgently needed decarbonization in the steel industry., Scripta Materialia, 213, ISSN:1359-6462, ISSN:1872-8456

There are several facets of aluminum when it comes to sustainability. While it helps to save fuel due to its low density, producing it from ores is very energy-intensive. Recycling it shifts the balance towards higher sustainability, because the energy needed to melt aluminum from scrap is only about 5% of that consumed in ore reduction. The amount of aluminum available for recycling is estimated to double by 2050. This offers an opportunity to bring the metallurgical sector closer to a circular economy. A challenge is that large amounts of scrap are post-consumer scrap, containing high levels of elemental contamination. This has to be taken into account in more sustainable alloy design strategies. A "green aluminum " trend has already triggered a new trading platform for low-carbon aluminum at the London Metal Exchange (2020). The trend may lead to limits on the use of less-sustainable materials in future products. The shift from primary synthesis (ore reduction) to secondary synthesis (scrap melting) requires to gain better understanding of how multiple scrap-related contaminant elements act on aluminum alloys and how future alloys can be designed upfront to become scrap-compatible and composition-tolerant. The paper therefore discusses the influence of scrap-related impurities on the thermodynamics and kinetics of precipitation reactions and their mechanical and electrochemical effects; impurity effects on precipitation-free zones around grain boundaries; their effects on casting microstructures; and the possibilities presented by adjusting processing parameters and the associated mechanical, functional and chemical properties. The objective is to foster the design and production of aluminum alloys with the highest possible scrap fractions, using even low-quality scrap and scrap types which match only a few target alloys when recycled.

This study investigates the high strain-rate tensile properties of a cold-rolled medium-Mn steel (Fe–12Mn–3Al-0.05C % in mass fraction) designed to have a multi-phase microstructure and positive strain-rate sensitivity. At the intercritical annealing temperature of 585 °C, increasing the annealing time from 0.5 h to 8 h increased the phase volume fraction of ultrafine-grained (UFG) austenite from 2% to 35% by reversion. The remainder of the microstructure was composed of UFG ferrite and recovered α′-martensite (the latter resembles the cold-rolled state). Servo hydraulic tension testing and Kolsky-bar tension testing were used to measure the tensile properties from quasi-static strain rates to dynamic strain rates ( e ˙ = 10-4 s-1 to e ˙ = 103 s-1). The strain-rate sensitivities of the yield strength (YS) and ultimate tensile strength (UTS) were positive for both annealing times. Tensile properties and all non-contact imaging modalities (infrared imaging and digital image correlation) indicated an advantageous suppression of Luders bands and Portevin Le Chatelier (PLC) bands (a critical challenge in multi-phase medium-Mn steel design) due to the unique combination of microstructural constituents and overall composition. Fracture surfaces of specimens annealed for 0.5 h showed some instances of localized cleavage fracture (approximately 30 μm wide areas and lath-like ridges). Specimens annealed for 8 h maintained a greater product of strength and elongation by at least 2.5 GPa % (on average for each strain rate). The relevant processing-structure-property relationships are discussed in the context of recommendations for design strategies concerning multi-phase steels such that homogeneous deformation behavior and positive strain-rate sensitivities can be achieved.

Ultrafine austenite-ferrite duplex medium Mn steels often show a discontinuous yielding phenomenon, which is not commonly observed in other composite-like multiphase materials. The underlying dislocation-based mechanisms are not understood. Here we show that medium Mn steels with an austenite matrix (austenite fraction ∼65 vol%) can exhibit pronounced discontinuous yielding. A combination of multiple in situ characterization techniques from macroscopic (a few millimeters) down to nanoscopic scale (below 100 nm) is utilized to investigate this phenomenon. We observe that both austenite and ferrite are plastically deformed before the macroscopic yield point. In this microplastic regime, plastic deformation starts in the austenite phase before ferrite yields. The austenite-ferrite interfaces act as preferable nucleation sites for new partial dislocations in austenite and for full dislocations in ferrite. The large total interface area, caused by the submicron grain size, can provide a high density of dislocation sources and lead to a rapid increase of mobile dislocations, which is believed to be the major reason accounting for discontinuous yielding in such steels. We simultaneously study the Luders banding behavior and the local deformation-induced martensite forming inside the Luders bands. We find that grain size and the austenite stability against deformation-driven martensite formation are two important microstructural factors controlling the Luders band characteristics in terms of the number of band nucleation sites and their propagation velocity. These factors thus govern the early yielding stages of medium Mn steels, due to their crucial influence on mobile dislocation generations and local work hardening.

A hot-rolled medium Mn (0.2C5Mn) steel is annealed at 650 °C to produce an ultrafine-grained duplex microstructure with different austenite volume fractions by austenite reverted transformation (ART) annealing, and the orientation relationship strictly obeys K–S orientation relationship before deformation. Tensile tests are carried out in a temperature range from − 196 to 400 °C to examine the effects of the austenite volume fraction and the deformation temperature on the tensile properties and the austenite stability. Microstructural observations reveal that the metastable austenite gradually transformed into α-martensite, which is controlled by the deformation strain, the temperature and the austenite volume fraction. Both strain hardening behavior and ductility of the studied steel are dependent on austenite volume fraction and deformation temperature significantly. The stress–strain curves of ART-annealed 0.2C5Mn steel assume an S shape and a very large work hardening rate of about 10 GPa is obtained at liquid nitrogen deformation temperature. Based on the experimental data, a quantitative relation is proposed to describe the ductility dependence on both the austenite volume fraction and its mechanical stability.

Grain boundary segregation, embrittlement and phase nucleation are interconnected phenomena that are often treated separately, which is partly due to limitations of the current models to predict grain boundary segregation in non-ideal solid solutions. Here, a simple model is introduced to predict grain boundary segregation in solid solutions by coupling available bulk thermodynamic data with a mean-field description of the grain boundary character. The model is confronted with experimental results obtained in Fe-Mn alloys analysed by atom probe tomography. This model successfully predicts a first order transition or a discontinuous jump in the composition of the grain boundary which kinetically implies the formation of spinodal Mn fluctuations that tend to grow further with time within the segregated region. The increase in solute concentration at the grain boundary leads to an increase of the enthalpy of the boundary and to its embrittlement at lower temperatures. Once austenite is formed, the amount of segregated solute Mn on the grain boundaries is drastically reduced and the toughness of the grain boundary is increased.

This article reviews the concept of metastability in alloy design. While most materials are thermodynamically metastable at some stage during synthesis and service, we discuss here cases where metastable phases are not coincidentally inherited from processing, but rather are engineered. Specifically, we aim at compositional (partitioning), thermal (kinetics), and microstructure (size effects and confinement) tuning of metastable phases so that they can trigger athermal transformation effects when mechanically, thermally, or electromagnetically loaded. Such a concept works both at the bulk scale and also at a spatially confined microstructure scale, such as at lattice defects. In the latter case, local stability tuning works primarily through elemental partitioning to dislocation cores, stacking faults, interfaces, and precipitates. Depending on stability, spatial confinement, misfit, and dispersion, both bulk and local load-driven athermal transformations can equip alloys with substantial gain in strength, ductility, and damage tolerance. Examples include self-organized metastable nanolaminates, austenite reversion steels, metastable medium- and high-entropy alloys, as well as steels and titanium alloys with martensitic phase transformation and twinning-induced plasticity effects.

The formation of submicron structural defects within austenite (γ), e- and α′-martensite during cold rolling was followed in a 17.6 wt.% Mn steel. Several probes, including XRD, EBSD, and ECCI-imaging, were used to reveal the complex superposition of the strain hardening mechanisms of these phases. The maximum amount of e-martensite is observed at a strain of e = 0.11. At larger strains, the amount of e decreases suggesting that it precedes the α′-formation (γ → e → α′). Stacking faults and twins are the main planar defects noticed in e-martensite. The remaining γ is finely subdivided by stacking faults and twins up to e = 0.22. From e = 0.51 on, twinning and multiplication of dislocations are the principal strain hardening mechanisms in austenite. Deformation is accommodated in α′ by the rearrangement of dislocation tangles into dislocation cells plus shear banding at e = 1.56. During cold rolling, austenite develops a Brass-type texture component, which can be associated to mechanical twinning. e-martensite presents its basal planes tilted ∼24° from the normal direction towards the rolling direction. The α′-martensite develops and strengthens both, the bcc α- and γ-texture fibers during cold rolling.

A medium-Mn steel (Fe-12Mn-3Al-0.05C wt%) was designed using Thermo-Calc® simulations to balance the fraction and stacking fault energy of reverted austenite. Intercritical annealing for 0.5, 8 and 48 h was carried out at 585 °C to investigate the microstructural evolution. X-ray diffraction (XRD), electron backscatter diffraction (EBSD), 3-dimensional EBSD, energy-dispersive spectroscopy via scanning-transmission electron microscopy (STEM-EDS) and atom probe tomography (APT) enable characterization of phase fraction, grain area, grain morphology and alloy partitioning. An increase in annealing time from 0.5 h to 48 h increases the amount of ultrafine-grained (UFG) reverted austenite from 3 to 40 vol %. EBSD and TEM reveal multiple morphologies of UFG austenite (equiaxed, rod-like and plate-like). In addition, most of the remaining microstructure consists of recovered α′-martensite that resembles the cold-rolled state, as well as a relatively small fraction of UFG ferrite (i.e., only a small amount of martensite recrystallization occurs). Multi-scale characterization results show that the location within the cold-rolled microstructure has a strong influence on boundary mobility and grain morphology during austenite reversion. Results from APT reveal Mn-decoration of dislocation networks and low-angle lath boundaries in the recovered α′-martensite, but an absence of Mn-decoration of defects in the vicinity of austenite grains, thereby promoting recovery. STEM-EDS and APT reveal Mn depletion zones in the ferrite/recovered α′-martensite near austenite boundaries, whereas gradients of C and Mn co-partitioning are visible within some of the austenite grains after annealing for 0.5 h. Relatively flat C-enriched austenite boundaries are present even after 8 h of annealing and indicate certain boundaries possess low mobility. At later stages the growth of austenite followed the local equilibrium (LE) model such that the driving force between two equilibrium phases moves the mobile interface, as confirmed by DICTRA simulations (a Thermo-Calc® diffusion module). The sequence of austenite reversion is: (i) formation of Mn- and C-enriched face-centered-cubic nuclei from decorated dislocations and/or particles; (ii) co-partitioning of Mn and C and (iii) growth of austenite controlled by the LE mode.

We propose a novel “Mn preservation-cold deformation-intercritical annealing” treatment for medium Mn steels. The process produces a large fraction of retained austenite with an enhanced stability, which increases the ductility by ~40% compared to the conventional single-step intercritical annealing in a model 10Mn steel, without sacrificing the ultrahigh strength level.

Medium Mn steels possess a composite like microstructure containing multiple phase constituents like metastable austenite, ferrite, δ-ferrite and α′-martensite with a wide range of fractions for each constituent. The high mechanical contrast among them and the deformation-driven evolution of the microstructure lead to complex fracture mechanisms. Here we investigate tensile fracture mechanisms of medium Mn steels with two typical types of microstructures. One group consists of ferrite (α) plus austenite (γ) and the other one of a layered structure with an austenite-ferrite constituent and δ-ferrite. Samples with the first type of microstructure show a dimple-type fracture due to void formation primarily at the ferrite/strain-induced α′-martensite (α′) interfaces. In contrast, the fracture surface of δ-ferrite containing steels shows a combination of cleavage in δ-ferrite and dimple/quasi-cleavage zones in the γ-α (or γ/α′-α) constituent. The embrittlement of δ-ferrite is due to the formation of B2 ordered phase. Failure of these samples is govern by crack initiation related to δ-ferrite and crack-arresting ability of the γ-α layers. Austenite stability is critical for the alloys' fracture resistance, in terms of influencing void growth and coalescence for the first type of microstructure and crack initiation and termination for the microstructure containing δ-ferrite. This effect is here utilized to increase ductility and toughness. By tailoring austenite stability towards higher fracture resistance, the total elongation of δ-ferrite containing steels increases from ∼13% to ∼33%. This approach opens a new pathway towards an austenite-stability-controlled microstructural design for substantially enhanced damage tolerance in steels containing metastable austenite and δ-ferrite.