Cu–Cr–Zr alloys are widely used in the field of electrical conduction. In this study, Cu–1Cr–0.15Zr alloy with enhanced strength and electrical conductivity was produced by cyclic channel die compression and aging. After 7 passes of processing at room temperature, the average grain size was reduced from the initial 11.7 to 0.96 μm, and the fraction of low-angle grain boundaries reached 61.9 pct. The texture was significantly intensified, with the maximum pole density increasing from the initial 3.84 to 4.84. The tensile strength, yield strength, and electrical conductivity increased from 249, 74 MPa, and 30.2 pct IACS to 566, 514 MPa, and 36.8 pct IACS, respectively. After further aging at 450 °C for 1 hour, the average grain size increased to 1.15 μm. The texture was further intensified, and the maximum pole density reached 6.76. In addition to Zr-rich nanoparticles, a large number of face-centered cubic structured Cr nanoparticles, which are coherent with the Cu matrix, precipitate from the Cu matrix. The tensile strength, yield strength, and electrical conductivity further increased to 596, 593 MPa, and 80.5 pct IACS, respectively. The dominant strengthening mechanisms are fine-grain strengthening, deformation strengthening, and precipitation strengthening. The precipitation of dissolved Cr and Zr atoms was identified as the primary factor responsible for the substantial improvement in electrical conductivity.
This study investigates the influence of active fluxes (SiO2, TiO2, and their mixture) on arc behavior and weld bead geometry during Gas Tungsten Arc Welding (GTAW) of Inconel 738LC. Arc imaging and quantitative arc width measurements indicate that SiO2 produces a more constricted arc compared to TiO2, while the mixed flux exhibits intermediate behavior with characteristics closer to SiO2. Weld bead analysis reveals pronounced central metal accumulation and crown formation in flux-assisted samples, whereas the flux-free condition results in a relatively flat bead profile. In addition, WDS analysis shows an increase in oxygen content in flux-assisted welds, with the highest values observed in the mixed and TiO2 conditions, suggesting enhanced oxygen transfer into the weld pool. Quantitative bead measurements confirm an increase in bead area from 0.09 mm2 (flux-free) to 0.21 mm2 (SiO2) and 0.24 mm2 (TiO2), while the mixed flux exhibits a dual-hump crown morphology. Overall, the results suggest that flux composition significantly influences arc energy distribution and weld pool behavior through coupled effects of arc constriction and oxygen-related surface tension modifications. These observations suggest that SiO2 is mainly associated with arc constriction effects, whereas TiO2 appears to have a stronger influence on weld pool flow through oxygen-related surface tension modifications, with the mixed flux potentially reflecting a combined contribution of both mechanisms.
Steel grade transition continuous casting is widely applied in steel plants. In this paper, the steel grade transition process inside a two-strand slab tundish is numerically investigated, focusing on the influence of dynamic variation of molten steel weight in the tundish. The simulation uses the VOF multiphase flow and Species transport model, and its reliability is validated against industrial measurement data. The effects of steel throughput ranging from 2.90 to 5.46 tons/min, different filling rates and tundish molten steel weight changes of 4, 12 and 20 tons on mixing behavior were systematically analyzed. The results show that during the steel grade transition process, increasing the steel throughput and the tundish filling rate can shorten the mixing time; meanwhile, the less residual molten steel remains in the tundish, the faster the steel grade transition proceeds. Comprehensively considering the molten steel mixing behavior and cleanliness, the optimal operating parameters are determined as follows: a steel throughput of 5.46 tons/min, a tundish molten steel weight change of 20 tons with subsequent recovery to a steady liquid level, a minimum tundish liquid level of 910 mm, and a corresponding mixing time of 455 seconds. This parameter configuration achieves a reasonable transition slab length while ensuring molten steel quality.
This study investigated the interpass microstructural evolution of Fe–0.2C–3.5Mn low-Mn medium-manganese steel during multi-pass hot working by conducting double-pass isothermal hot compression tests. The effects of deformation temperature, strain rate, and time interval on the flow stress behavior, static softening fraction, microstructural evolution, and static recrystallization kinetics were systematically studied. The results showed that, during double-pass hot deformation, the yield stress in the second pass was markedly higher than the initial yield stress in the first pass but lower than the peak stress in the first pass, indicating that only incomplete static softening occurred after the first-pass deformation. With increasing deformation temperature and prolonged time interval, the static softening fraction increased significantly, and the dominant recovery mechanism gradually shifted from static recovery to static recrystallization, accompanied by a tendency for static grain growth under high-temperature and long-time conditions. The strain rate regulated the interpass softening behavior by altering the stored energy after the first pass, the substructural morphology, and the resistance to grain-boundary migration. Within the processing window investigated in this study, a higher strain rate corresponded to more effective static softening. Electron backscatter diffraction results revealed that static recrystallization after double-pass hot deformation in this steel mainly proceeded via two mechanisms: one was grain-boundary bulging nucleation along prior austenite grain boundaries (PAGBs), characterized by a typical necklace-type recrystallized structure; the other was subgrain-coalescence nucleation governed by intragranular dislocation rearrangement, subgrain rotation, and coalescence, exhibiting features of continuous static recrystallization. Based on the static softening fraction, an Avrami-type kinetic model for static recrystallization was established. These results provide a theoretical basis for the design of hot rolling and multi-pass hot forming processes for this low-Mn medium-manganese steel.
Although Ce treatment has been shown to modify non-metallic inclusions in H13 hot-work die steel, the transformation sequence and precipitation behavior of Ce-bearing inclusions under coupled composition–temperature conditions remain insufficiently clarified. In this study, FactSage 8.2 thermodynamic calculations, melting experiments, and SEM–EDS characterization were used to investigate the effects of O, S, Al, and Ca contents, the S/O ratio, and temperature on Ce-modified inclusion evolution. Increasing the O content stabilized Ce2O3 and CeAlO3, whereas low O content and a high S/O ratio favored CexSy formation. Increasing the O content or decreasing the S/O ratio shifted CexSy formation to higher Ce contents and increased the critical Ce content required for its formation. The Al content mainly promoted CeAlO3 and Al2O3 and had a limited effect on the overall transformation pathway. Increasing the Ca content from 0.0001 to 0.0025 wt pct decreased CeAlO3 and Ce2O2S, whereas Ce2O3 increased when the Ca content reached 0.0025 wt pct. At 0.01 wt pct Ca, CeAlO3 and Ce2O2S no longer appeared, and CaO and CaS became the main inclusions. Lower temperatures generally favored Ce2O2S, CeAlO3, and CexSy relative to Ce2O3. Experimental observations were qualitatively consistent with the calculated trends. These results establish a thermodynamic basis for controlling Ce-bearing inclusion evolution in H13 steel through composition and temperature adjustment.
SUS430/Cu interfacial diffusion was investigated by in situ HT-CLSM, cross-sectional SEM/EDS, CALPHAD, and MD. After 2400 seconds, the HT-CLSM apparent affected-zone thickness increased from 14.65 ± 0.56 μm at 950 °C to 18.54 ± 0.64 μm at 1000 °C. Boltzmann–Matano/Sauer–Freise analysis of normalized, smoothed, monotonic Fe EDS profiles over Y = 0.2 to 0.8 yielded pseudo-binary effective interdiffusion coefficients D_mid= (9.68 ± 0.56) × 10⁻15 m2 s⁻1 at 950 °C and D_mid= (13.52 ± 0.65) × 10⁻15 m2 s⁻1 at 1000 °C, with corresponding w_10 to 90 = 14.27 ± 0.16 μm and w_10 to 90 = 17.92 ± 0.82 μm , respectively. CALPHAD calculations provided phase-field constraints for the Cu–Fe–Cr system and indicated no stable Fe–Cu intermetallic layer. In constrained nanoscale Cu|(Fe–Cr) cells, Cu mobility was resolved for 0Cr, 8Cr, and 16Cr, whereas Fe migration in 0Cr and Fe/Cr migration in 8Cr were not reliably resolved (N.R.) over 100 to 500 ps; Fe and Cr mobilities were resolved in the investigated 16Cr cells. In this context, the MD-derived diffusion coefficients represent apparent interfacial mobilities rather than general bulk diffusivities, and the current data do not support establishing a monotonic Cr-content law.
In this study, we investigate the mechanical behavior of single-crystal vanadium and introduce an enhanced dislocation-based crystal plasticity framework to model its response under various loading conditions. Our experiments, conducted over a wide range of temperatures and strain rates on single-crystal vanadium samples, reveal progressive thermal hardening at low strain rates and elevated temperatures. This behavior is indicative of dynamic strain aging (DSA), a phenomenon previously observed in polycrystalline vanadium. Because of DSA, conventional single-crystal plasticity models—effective for other body-centered cubic (BCC) metals such as tantalum or molybdenum—cannot accurately capture the mechanical response of vanadium within the DSA regime. To overcome this limitation, we refined the saturation dislocation density formulation to incorporate DSA effects while maintaining predictive capability for other BCC crystals. Our approach remains relatively simple yet robust, providing satisfactory predictions of the material’s strength behavior both inside and outside of the DSA regime. It successfully predicts the reported temperature dependence of polycrystalline vanadium flow stress. In addition, it also predicts stress adjustments associated with negative strain-rate sensitivity in rate-jump tests.
In this study, an Ag-free, rare-earth-free Sn–Ti–Ni solder was designed, fabricated, and evaluated for ultrasound-enabled flux-free joining of Al alloys in air at 400 °C. Ti was incorporated through Ni–Ti pre-alloying to mitigate the intrinsic immiscibility of the binary Sn–Ti system, producing a more homogeneous solder microstructure with refined Ti-bearing precipitates and suppressing the coarse Ti-rich domains observed in binary Sn–Ti alloys. However, under the present low-temperature joining conditions, no evidence of classical Ti-mediated interfacial reaction-layer formation with Al was observed. This is likely due to the thermodynamic stability of Al_2O_3 and the kinetic immobilisation of Ti within stable Sn–Ti intermetallics. Instead, joint formation was governed by ultrasonic oxide disruption, local substrate dissolution, and molten-solder infiltration. Joints produced on commercially pure Al showed limited adhesion and low shear strength ( ∼ 20 MPa), whereas the use of Al-5754 promoted a favourable substrate-assisted joining pathway: Mg appears to dissolve into the molten solder and is associated with the formation of Mg_2Sn -like phases during cooling, together with deeper grain-boundary infiltration and reinforcement of the solder seam. Ultrasonic activation promoted oxide rupture and interfacial disruption and appeared to have enhanced solute redistribution within the seam, enabling stronger joints with more pronounced mechanical interlocking and seam microstructures consistent with chemical reinforcement. Under these conditions, shear strengths exceeding 50 MPa were obtained for Al-5754. These results show that strong joints in Al alloys can be achieved at low temperature in air without Ag or rare-earth additions, and that ultrasound-enabled flux-free joining with a Sn–Ti–Ni solder depends on the combined effects of solder design, oxide disruption, and substrate-assisted chemistry.
To minimize impurity introduction during the leaching of pyrolusite and mitigate the adverse effects of impurities on downstream processing, sodium thiosulfate (Na2S2O3, ST), sodium sulfite (Na2SO3, SS), and sodium disulfite (Na2S2O4, SD) were systematically evaluated as reductive leaching agents. A comparative analysis revealed distinct leaching performances under optimized conditions: for ST, a molar dosage of 1.4 × relative to MnO2, a reaction time of 80 minutes, an H2SO4 concentration of 3 mol L⁻1, and a temperature of 363 K yielded a manganese extraction efficiency of 90.57 pct; for SS, under identical dosage and time but at 2 mol L⁻1 H2SO4 and 323 K, extraction reached 95.42 pct; and for SD, a lower dosage (0.8 × MnO2 molar ratio), a shorter duration (20 minutes), milder acidity (2 mol L⁻1 H2SO4), and ambient temperature (293 K) achieved 88.03 pct extraction. Kinetic analysis indicated apparent activation energies of 30.84 and 24.83 kJ·mol⁻1 for ST- and SS-mediated leaching, respectively, confirming their higher thermal dependence compared to SD. XRD and SEM–EDS mapping analyses of raw pyrolusite and post-leaching residues confirmed structural transformation and compositional evolution; the residue exhibited a porous, irregular morphology with SiO2 identified as the predominant phase. Notably, SD enabled rapid, low-temperature leaching—attributed to its participation in multiple concurrent redox pathways—whereas ST and SS required elevated temperatures to overcome kinetic barriers and attain high extraction yields. This study offers mechanistic insights and practical guidance for selecting sulfur-based reductants in pyrolusite hydrometallurgy.
Mold electromagnetic stirring (M-EMS) is widely used in the continuous casting (CC) of IF steel to improve the slab subsurface cleanness with proper installation position and stirring parameters. In this study, numerical simulation combined with industrial experiment was employed to systematically investigate the effects of M-EMS parameters on Al2O3 inclusion transport and slag entrainment. The results indicate that the area fraction within the optimal velocity range of 0.2 to 0.4 m/s at the solidification front gradually increases with increasing stirring intensity. Meanwhile, the capture rate of Al2O3 inclusions within the 10 mm subsurface region gradually decreases, which is consistent with the industrial experimental results. With the distance between the upper edge of the stirring device and the slag–steel interface increasing from 20 to 100 mm, the inclusion capture rate within the 10 mm subsurface region correspondingly increases, resulting in the deterioration of surface quality. When the stirring current is below 500 A, no abnormal fluctuation is detected at the steel–slag interface, and the number of exceedances of the Kelvin–Helmholtz critical threshold remains limited, indicating a relatively low risk of slag entrainment. However, when the current reaches 600 A, the proportion of abnormal fluctuations in the meniscus region increases to 50 pct, with multiple threshold exceedances observed at all monitoring points. Based on a comprehensive evaluation of Al2O3 transport behavior and the risk of slag entrainment, it is recommended that the M-EMS device be positioned 20 mm below the steel–slag interface, with the current intensity controlled below 500 A.
The efficient extraction of niobium and tantalum is closely associated with the mineral phase restructuring that occurs during oxidative roasting pretreatment. However, the microscopic mechanisms by which Fe/Mn oxidation drives both the lattice disruption of (Fe,Mn)(Nb,Ta)2O6 and oxygen migration remain unclear. In this study, X-ray photoelectron spectroscopy (XPS) and Raman spectroscopy were employed, combined with thermodynamic calculations and semi-quantitative X-ray diffraction (XRD) analysis, to systematically elucidate the synergistic mechanism between Fe/Mn oxidation and oxygen migration. The results indicate that Fe/Mn oxidation is the primary driving force behind lattice activation. At 1000 °C, the Fe3+ content rises to 69.61 pct and the Mn2+ content to 40.94 pct; the full width at half maximum of the (131) plane increased by 64.2 pct, the oxygen vacancy content peaked at 23.93 pct and the maximum red shift of the Nb/Ta–O bond reached 17.7 cm−1 (occurring at 1100 °C). In the 900 °C to 1000 °C range, Raman blue shift coexists with an increase in oxygen vacancy content, revealing the ‘short-range order, long-range disorder’ structural characteristics of intermediate phases such as MnTa2O6 and MnNb2O6, and demonstrating that mineral phase reconstruction is a process in which destruction and reconstruction occur simultaneously. Thermodynamic calculations and XRD analysis jointly indicate that Fe2+ is preferentially oxidized to Fe2O3 rather than directly combining with Nb/Ta to form Fe-based reconstructed phases; Mn2+ is more readily able to combine with Nb/Ta to form reconstructed phases than Fe2+. Acid leaching experiments indicate that the leaching rates of niobium and tantalum both reach their maximum values at 1000 °C, at 91.9 and 95.9 pct, respectively, which is consistent with the fact that lattice distortion and the oxygen vacancy content peak at this temperature. Based on these findings, this paper proposes a five-stage migration pathway for oxygen atoms: detachment → entry → rearrangement → encapsulation → stabilization, this pathway describes the process by which oxygen, originating from the (Fe,Mn)(Nb,Ta)2O6 lattice, sequentially enters Fe2O3/Mn3O4 and MnTa2O6/MnNb2O6, undergoes local rearrangement at 1000 °C and is encapsulated by MnSiO3 before entering Nb2O5/Ta2O5 at temperatures above 1100 °C. The oxidation of Fe and Mn, together with oxygen migration, synergistically drives the complete mineralogical reconstruction process from lattice activation to structural stabilization.
Grain boundary precipitation strengthening is an effective strategy for improving the high-temperature mechanical properties of heat-resistant alloys. However, the grain boundary coverage fraction of intergranular precipitates cannot be controlled simply by thermodynamic factors, such as solute supersaturation, because precipitation behavior is also governed by grain boundary character, including both the crystallographic relationship between adjacent grains and the grain boundary plane orientation. In this study, the effect of grain boundary character on α-Cr precipitation along γ/γ grain boundaries was investigated in Ni–Cr binary alloys using electron backscatter diffraction and focused ion beam cross-sectional observations. The α phase precipitated preferentially along grain boundaries at low Cr supersaturation, whereas intragranular and discontinuous precipitation occurred with increasing supersaturation. Even under conditions where the average grain boundary coverage fraction became nearly saturated, a large boundary-to-boundary scatter was observed, indicating the importance of local grain boundary character. Notably, among grain boundaries with the same crystallographic relationship between adjacent γ grains, the precipitation morphology and coverage fraction varied systematically with grain boundary plane orientation. Based on the local crystallographic orientation analysis, grain boundary α precipitates were classified into three types: Type I, satisfying orientation relationships with both adjacent γ grains; Type II, satisfying an orientation relationship with only one adjacent γ grain and consisting of a single variant; and Type III, satisfying an orientation relationship with only one adjacent γ grain but consisting of multiple variants. The coverage fraction decreased in the order Type I > Type II > Type III. For Types II and III, the coverage fraction increased with decreasing angle ϕ between the grain boundary plane and the 111γ habit plane. This trend was interpreted in terms of the reduced nucleation barrier for faceted γ/α interfaces. These results suggest that controlling both the crystallographic relationship between adjacent γ grains and the local grain boundary plane is important for enhancing grain boundary coverage in Ni–Cr alloys.
This work presents a combined experimental and numerical investigation of local overheating in NiAl during spark plasma sintering (SPS). Experimental studies were conducted on samples sintered under different temperature and pressure conditions to capture both the early and final stages of densification. Both perpendicular and parallel loading directions were analysed for microstructural evidence of local melting. Experimental research comprised post-process analysis using SEM and EBSD, while supplementary XRD analysis was used to assess phase stability and crystallographic characteristics. Microstructural analysis focused on regions most susceptible to overheating (contact areas, small particles trapped between larger ones). Sintered samples showed progressive densification, neck growth and grain coarsening with increasing temperature. EBSD maps of sintered compacts reveal distinct orientations across contacts, with no evidence of epitaxial or equiaxed grain growth. Experimentally, no evidence of melting due to local overheating was detected. The experimental observations were complemented by particle-scale thermo-electric FEM simulations. Geometries of equal-sized particles, arranged in a straight line and in a zig-zag chain, connected by the necks with increased resistance, were considered to analyse the temperature rise in the necks. Geometry consisting of a small particle between two large particles was analysed to assess temperature nonuniformity in heterogeneous microstructures. The results show temperature increase in contacts or in small particles; however, the resulting temperature differences are less than 1 ^∘ C in all cases studied. Numerical results support the experimental findings that NiAl is not prone to local overheating in SPS under the studied conditions.
The microstructure characteristics exert a decisive influence on the ratcheting behavior and fatigue strength of low-carbon microalloyed steels under asymmetric cyclic loading. In this study, systematic stress‑controlled asymmetric fatigue tests (stress ratio R(σ_min/σ_max)=0 ) were conducted at room temperature on an ultra-low-carbon microalloyed steel with three distinct microstructural states: as‑hot‑rolled (HR), as‑quenched (Q), and tempered (T). Combined with electron backscatter diffraction (EBSD) analysis, the cyclic deformation behavior and failure mechanisms were thoroughly investigated. The results indicate that although the HR specimen exhibits the highest initial yield strength (621 MPa), the 850 °C quenched (Q-850°C) specimen demonstrates the best fatigue resistance, with a fatigue strength limit of approximately 370 MPa. Under a high load ratio (LR(σ_max/σ_0.2))=1.1 for the Q-850°C and T-600°C specimens, or 1.07 for the HR specimen), the three microstructures exhibit distinct ratcheting behaviors: the HR specimen shows ratcheting divergence and macroscopic necking; the 600 °C tempered (T-600°C) specimen exhibits steady‑state ratcheting accumulation; and the Q-850°C specimen, characterized by a heterogeneous structure consisting of a soft matrix with approximately 9.0 pct martensite as the hard phase, displays a unique ratcheting shakedown state, i.e., ratcheting strain ceases to accumulate after a certain number of cycles. Meanwhile, the random lattice rotation of ferrite effectively accommodates the strain mismatch with martensite, delaying crack propagation. A comparison of fatigue life predictions using the conventional Basquin model and the Smith–Walker model indicates that the Basquin model tends to be conservative.
Thermochemical hot isostatic pressing (TC-HIP) in a nitrogen (N) atmosphere using a HIP system with uniform rapid quenching (URQ®) integrates high pressure, elevated temperature, and rapid cooling to introduce significant amounts of N into stainless steels. Interstitially dissolved N increases hardness and corrosion resistance while stabilizing the austenitic phase, making TC-HIP suitable for microstructural design and the improvement of near-surface properties. In this feasibility study, ferritic stainless steel X6Cr17 (DIN EN 1.4016, AISI 430) was TC-HIP heat-treated in an N2 atmosphere followed by URQ®. Microstructural analysis by scanning electron microscopy (SEM), electron backscatter diffraction (EBSD), X-ray diffraction (XRD), and glow discharge optical emission spectrometry (GDOES) demonstrated the formation of nitrides, N-enriched austenite, and martensite. GDOES, CALPHAD calculations, and DICTRA simulations were used to investigate N uptake during the nitriding process. Microhardness and corrosion testing revealed that CrN precipitation increases hardness (880 HV0.05) but impairs corrosion resistance, while the N-enriched austenite combines elevated hardness (365 HV0.05) with excellent corrosion performance. Martensitic regions exhibit high hardness (640 HV0.05) together with a corrosion resistance comparable to conventional austenitic grades. The present work thus lays the foundation for TC-HIP as a tool for microstructural design and targeted enhancement of near-surface properties in stainless steels.
Low-temperature toughness affects the service performance of high-strength materials in extreme environments. To investigate the mechanisms affecting low-temperature impact toughness, this study examines the effects of tempering temperature and tempering time. The effects of microstructural evolution, retained austenite (RA) morphology, distribution, and thermal stability on low-temperature impact toughness during the tempering process of bainite–martensite dual-phase steel were studied using metallographic characterization, electron backscatter diffraction (EBSD), and X-ray diffractometer (XRD) analyses. The low-temperature Charpy impact energy first increased and then decreased with increasing tempering temperature, reaching the highest value at 220 °C, while prolonged tempering further reduced toughness at relatively high tempering temperatures. Results indicate that the initial increase in toughness during tempering is primarily due to the decomposition of blocky retained austenite (BRA) and the enhanced stability of film-like retained austenite (FRA). However, further increases in tempering temperature and duration lead to carbide coarsening within bainite, which subsequently reduces toughness.
This study investigates the influence of aging treatment on the microstructural evolution and mechanical response of precipitation-hardenable XH55MБЮ (55Ni–17Cr–12Fe–9Mo–2Nb–1.5Al) superalloy, with properties evaluated from 77 K to 900 K. Although conventional double aging (730 °C/15 hours + 650 °C/10 hours) is routinely employed, the individual roles of the two aging stages remain unclear. To decouple the precipitation pathway, the standard treatment was separated into single-step aging at 730 °C (A-series: 10 to 30 hours) and 650 °C (B-series: 5 to 20 hours), and compared with the solution-treated as well as standard double-aged (STA) conditions. Microstructural evolution was characterized using SEM, TEM, and APT, while tensile and impact properties were evaluated at 77 K, room temperature, and 900 K. The 650 °C treatment promoted incipient γ′/γ″ clustering, whereas 730 °C aging produced a higher number density of mature spherical γ′ and disc-shaped γ″ precipitates. Sequential aging in the STA condition generated a refined bimodal γ′/γ″ distribution with pronounced Nb and Al partitioning, resulting in the highest hardness (∼332 HV) and superior tensile strength across all temperatures. Notably, the alloy retained ductile dimpled fracture even at 77 K, indicating suppression of the ductile-to-brittle transition. The enhanced response of the STA condition arises from non-additive strengthening through sequential precipitation refinement, enabling a favorable combination of strength and cryogenic toughness.
Overcoming the strength-ductility trade-off in low-alloy advanced high-strength steels (AHSS) remains a pivotal challenge. This work establishes a fully martensitic start microstructure, which subjected to intercritical annealing and overaging to unlock excellent strength-ductility synergy in transformation-induced plasticity (TRIP)-aided steels. The key finding is the significant but non-equilibrium redistribution of carbon and substitutional elements during intercritical annealing, which creates a pronounced chemical heterogeneity. Subsequent overaging further partitions carbon into austenite, while substitutional atoms indirectly govern the final carbon content and retention of austenite. Austenite reversion exhibits pronounced variant selection, dominated by parent-oriented and twin-related variants. Their subsequent decomposition generates low-angle boundaries and disrupts twin relationships. The retained austenite (RA) fraction is accurately predicted by models incorporating synergistic martensitic/bainitic carbon partitioning, with precision enhanced by considering the non-equilibrium substitutional distribution. The comparable hardness of soft phases renders mechanical properties primarily dependent on the hard martensite/austenite (M/A) islands, alongside with the contribution of TRIP effect. Ultimately, this microstructural strategy achieves a product of strength and elongation approaching or exceeding 30 GPa pct. The superior performance stems from a synergistic microstructure comprising a continuous soft phase matrix, dispersed hard M/A islands, and RA with a stability spectrum deriving from tailored chemical heterogeneity, wherein RA enhances yield strength and undergoes progressive martensitic transformation to delay necking.
Austenitic lightweight steels can be crucial in achieving net-zero emissions and energy efficiency due to their superior mechanical properties and lower density. The mechanical response of these steels is strongly influenced by the precipitation of nanoscale kappa (κ)-carbides, and understanding their evolution and impact on microstructure and properties is crucial for tailoring the strength–ductility balance. This study investigates the precipitation of κ-carbides and their effect on the microstructure, texture, and mechanical behavior of a Fe–18.52Mn–6.78Al–4.17Cr–0.61Si–0.91C steel after solution treatment and subsequent aging at 550 °C and 600 °C, in conjunction with thermo-kinetic simulations of precipitation. The solution-treated condition exhibited a fully austenitic microstructure with high ductility ( 90 pct). Aging at 550 °C led to the formation of fine ( 5 to 6 nm) intragranular κ′-carbides maintaining coherent-to-semi-coherent interfaces and enhancing strength. In contrast, aging at 600 °C promoted the nucleation and growth of coarser intergranular κ*-carbides, leading to loss of coherency and grain coarsening. Magnetic measurements confirmed the ferromagnetic nature of κ′-carbides. The results further indicated that S-type texture (for austenite phase) was predominant in the solution-treated (ST) and aged at 550 °C samples, whereas Brass (B) and E{111}⟨ 110 ⟩ texture components developed after aging at 600 °C. Optimal properties were achieved after aging at 550 °C for 60 minutes (tensile strength 935 MPa, elongation 62 pct). The results establish a direct correlation between the mechanisms of κ-carbide precipitation, coherency evolution, and mechanical response.
Continuous casting, as a key process in steel production, relies primarily on macrostructure examination for bloom quality assessment. However, this method is characterized by time lag and subjective human judgment, which impedes timely feedback and guidance for optimizing continuous casting process parameters. This study addresses this issue using 20# steel as an example, proposing a method based on IoT multi-source process parameters and a generative model to predict the macrostructure morphology of blooms, enabling early prediction of shrinkage cavities. For macrostructure images collected on-site, U-Net and YOLOv8 networks are employed for preprocessing to eliminate environmental interference and identify shrinkage cavities, achieving a recognition accuracy of 92.1 pct. Box plots and random forest algorithm analysis identified specific water flow rate and casting speed as the primary influencing factors. A generative model integrating autoencoder, parameter encoder, and latent code matching was constructed. Peak signal-to-noise ratio (PSNR), integrated similarity, and structural similarity index measure (SSIM) were used to evaluate different autoencoder models. The enhanced autoencoder with CBAM attention mechanism outperformed others. Parameter optimization determined the optimal latent code layer size as 256. Validation with 15 datasets confirmed the model’s reliability and stability: PSNR exceeded 33 dB, and integrated similarity and SSIM were greater than 0.73 and 0.7, respectively, with a maximum mean squared error (MSE) of 0.018. The generated images meet industrial accuracy requirements, enabling second-level prediction of bloom macrostructure from process parameters and significantly improving continuous casting quality control efficiency.
