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.
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.
Metallic materials have been widely investigated for use in aluminum electrolysis cells due to their electrical conductivity, mechanical robustness, and ease of fabrication. Under Hall–Héroult conditions, however, their long-term stability is limited by chemical corrosion driven by interactions between alloying elements, oxide scales, and fluoride-based electrolytes. This review considers the chemical corrosion behavior of metallic systems exposed to molten fluorides relevant to aluminum electrolysis, with particular emphasis on Cu–Ni-based, Ni–Fe-based, high-entropy alloys, and Ni-based superalloys. Evidence from cryolite and cryolite-analog electrolytes is analyzed to identify dominant degradation mechanisms, including oxide dissolution, fluoridation reactions, selective elemental leaching, outward metal diffusion, and interfacial delamination. The roles of key alloying elements (Ni, Fe, Cu, Cr, Co, Mo, and dopants like Y) are discussed with respect to oxide-scale stability, interfacial chemistry, and contamination of the aluminum product. Rather than treating corrosion resistance as a bulk alloy property, the reviewed studies indicate that long-term stability is governed by the dynamic evolution of metal-oxide–electrolyte interfaces and by composition-dependent trade-offs between protective spinel formation and fluoride reactivity. On this basis, recurring limitations of conventional metallic systems are identified, along with unresolved gaps in early-stage oxide evolution, time-dependent degradation, and the transferability of results from alternative molten-salt systems.
Precise regulation of flow and heat transfer in the mold is critical to improving the subsurface segregation of high-carbon cord steel billets. This study aims to clarify the regulation mechanism of electromagnetic swirling flow in the nozzle (EMSFN) on the metallurgical behavior of high-carbon cord steel continuous casting mold and identify the optimal magnetic flux density parameters for quality improvement. A three-dimensional transient multi-field coupling mathematical model integrating the electromagnetic field, flow, heat transfer, solidification, and solute transport was established for a 150 × 150 mm high-carbon cord steel billet mold. The evolution characteristics of the molten steel flow, temperature distribution, and carbon solute transport under different EMSFN magnetic flux densities ranging from 25 to 150 mT were systematically investigated. The results show that EMSFN effectively induces circumferential rotational flow in the submerged entry nozzle (SEN), and this rotational flow is transmitted to the mold to homogenize the internal flow. At 150 mT, the molten steel impact depth is reduced by 29.22 pct compared with the non-magnetic field condition, and the flow symmetry index at different positions remains above 0.85 when the magnetic flux density exceeds 125 mT. The rotational flow also reshapes the mold temperature field, leading to a 42.75 pct upward shift and radial expansion of the high-temperature zone. For solute transport, the optimal carbon distribution uniformity is achieved at 125 mT, which is characterized by the disappearance of the high-carbon zone below the meniscus and the stabilization of carbon content at 0.759 wt pct at both the edge and corner of the mold outlet. This study reveals the core mechanism by which EMSFN regulates the mold’s flow-heat-mass transfer behavior through rotational diffusion and kinetic energy dissipation and clarifies the optimal application parameter range of EMSFN in high-carbon cord steel continuous casting. It provides a novel technical strategy and theoretical basis for the quality control of high-carbon steel billets and further expands the engineering application of electromagnetic metallurgy technology in the high-quality production of special steel.
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.
A multiphase computational fluid dynamics (CFD) model is developed to simulate the fluid flow and oxygen jet behaviour during top blowing in an industrial steelmaking converter. The study investigates a seven-hole lance that combines a central supersonic jet with six inclined peripheral nozzles to reveal its impact in basic oxygen furnace (BOF) steelmaking. The turbulent multiphase flow of the molten steel and oxygen is simulated by the combined realizable k-ω model and volume of fluid (VOF) method. Parametric studies are undertaken to examine the effects of number of nozzles, flow rates, nozzle inclinations, jet interactions on overall flow field, jet penetration, and droplet generation. Particular attention is paid to assess the flow structures and performance of a central supersonic, multi-angle seven-hole lance as compared to a conventional six-hole lance in BOF converter. This is accomplished by coupling the compressible flow physics with multiphase kinetics to ascertain the difference in hydrodynamic characteristics and impact behaviour of high intensity jets on the melt bath. The results demonstrate that the variation in nozzle inclination angle directly influences the rate of velocity attenuation. Supersonic central hole configuration leads to higher droplet generation signifying enhanced mass transfer at the jet impact site. With the supersonic central jet, the cavity exhibited a characteristic shape with a concentrated central depression and a well-defined rim. The maximum cavity area generated by the supersonic jet increased by approximately 12 pct compared to the subsonic jet configuration. Results also reveal distinct jet interaction regimes for different lance configuration. The velocity field for seven-hole lance indicates minimal boundary layer separation and sustained pressure enabling organized flow development.
Titanium distribution ratio ( L_Ti ) between Ti-bearing ferritic stainless steel (Ti-FSS) and CaO–Al2O3-based VOD slag was measured at 1823 K (1550 °C). Titanium oxide in the CaO–Al2O3–MgO–TiO2 slag exists as ‘ TiO_3^2- ’ complex and that the titanate capacity increases with slag basicity. However, L_Ti decreases with basicity due to a significant decrease in oxygen potential at slag-metal interface although titanate capacity of the slag increases. Hence, the highly basic slag is favorable to improve Ti yield.
This study presents a thermodynamic analysis of the behavior of oxygen (O) during the early stage of solidification in titanium–rare-earth (Ti–RE) alloys in the presence of rare-earth fluorides (REF3) and rare-earth oxyfluorides ( RE_xO_yF_z ). The partitioning of rare-earth elements (RE) and O between the melt and -Ti that precipitates at 1943 K (1670 ^∘ C) during cooling of Ti-rich Ti–RE alloys was estimated using thermodynamic data. The present analysis provides thermodynamic trends rather than exact quantitative predictions, owing to uncertainties in high-temperature thermodynamic data near 1943 K (1670 ^∘ C). For example, when molten Ti–Y alloy with Y concentration of 1 mass pct (10,000 mass ppm) is cooled, -Ti containing 0.2 mass pct Y (2000 mass ppm Y) is obtained. In addition, when solidification proceeds in equilibrium with liquid YF3 and solid Y5O4F7, the resulting -Ti contains approximately 0.053 mass pct O (530 mass ppm O). These results indicate that combining RE–REF3 flux deoxidation with solidification refining offers a potential route to producing high-purity solid Ti with both low RE and low O concentrations directly from molten Ti. Consequently, this technique provides a promising approach for advancing upgrade recycling (upcycling) technologies for Ti scrap.
To achieve effective control of macrosegregation in large steel ingots and maximize material utilization, an integrated casting–forging modeling framework for the formation and back-tracking of metallurgical defects was developed. Based on this approach, the formation and evolution of macrosegregation under different process conditions were systematically investigated. The results show that increasing the hot top height from 250 to 950 mm reduces the segregation exceedance radius at the hot top line from 0.604 to 0.399 m, indicating that macrosegregation defects become increasingly concentrated within the hot top region. Increasing the hot top height significantly reduces solute enrichment in the outer usable region of the ingot, while the segregation level in the central region exhibits a non-monotonic variation with a peak at intermediate hot top heights. For hollow forgings where the center region is removed during piercing, a larger hot top height effectively reduces segregation along the inner wall of the forging. The influence of hot top insulation is strongly dependent on the hot top height and mainly leads to an upward redistribution of macrosegregation within the ingot. Based on the proposed model and defect back-tracking analysis, an extreme hot top height of 300 mm was determined for the 225 t ingot used to produce an ultra-large hollow cylindrical forging. Compared with the conventional 650 mm hot top, the proposed design saves approximately 19 t of molten steel per casting cycle and approximately 13.0 MWh of electricity per cycle in electric-furnace melting and refining. Industrial production trials confirmed the quality of the resulting forging. The proposed modeling and process design methodology provides an effective tool for optimizing casting–forging processes of large steel forgings via extreme hot top design.
To reduce energy consumption in the steel industry, increasing scrap utilization rate is imperative. However, residual elements such as Cu and Sn tend to accumulate during the recycling of scrap steel, which impairs the hot ductility of steel, alters its stress–strain response, and ultimately affects the quality of continuous casting slabs. In this paper, DP780 automotive steel was taken as the research object, and the influences of residual elements (Cu and Sn) on its high-temperature mechanical behavior were systematically investigated via high-temperature tensile tests under different strain rates (0.001 to 0.1 s−1) and temperatures (700 °C to 900 °C). The modified Johnson-Cook (JC) model and the strain-compensated Arrhenius model were adopted to establish the constitutive equations for the stable plastic‑flow stage of the steel. The effects of residual elements on the parameters of different constitutive models were revealed, and the applicable ranges of the two models were also clarified. The results show that the peak stress increases progressively with increasing residual element content. In the strain-compensated Arrhenius model, positive even-order coefficients of α first decrease then increase with Sn content, while negative odd-order coefficients show the opposite trend; even-order coefficients of N, Q, and ln(A) decrease with Sn content, and odd-order ones increase. In the modified JC model, strain hardening coefficient B2 decreases with Sn content (B1 increases); for strain-rate hardening parameter C, even-order coefficients first rise then fall, with odd-order coefficients changing oppositely. In terms of model applicability, the strain‑compensated Arrhenius model is optimal for compositions from Sn-0 to Sn-2 at strain rates of 0.001 to 0.01 s−1. Within the strain‑rate interval of 0.01 to 0.1 s−1, it is suitable for Sn‑0 to Sn‑1, while the modified JC model is more appropriate for Sn-1 to Sn-2.
