- 百种中国杰出学术期刊
- 中国精品科技期刊
- “中国科技期刊卓越行动计划二期”领军期刊
- 入选“中国精品科技期刊顶尖学术论文(F5000)”
- 入选《科技期刊世界影响力指数(WJCI) 报告》
- EI、CA、JST收录期刊
- 中国金属学会会刊
- 中文核心期刊
- 中国科技核心期刊
- 中国科学引文数据库(CSCD)来源期刊
High-nitrogen austenitic stainless steel has excellent comprehensive performance, which exhibits wide potential applications in fields of marine engineering, petrochemical and biomedicine. With the development of technology, powder metallurgy technology and metal additive manufacturing have demonstrated unique advantages in realizing the structural and functional integration of high-nitrogen stainless steel and efficient nitrogen-controlled preparation. However, significant challenges remain in achieving synergistic optimization between powder quality and forming processes. This article systematically reviews current research progress in high-nitrogen stainless steel powder preparation, comparatively analyzes nitrogen control mechanisms and process characteristics of solid powder nitriding, gas atomization, plasma rotating electrode atomization methods. Furthermore, the influence of processes such as powder injection molding, hot isostatic pressing and additive manufacturing on the properties of materials is deeply discussed. Finally, for current technological bottlenecks, the future research directions of high-quality powders in fields of raw material preparation and molding are also prospected, which can provide reference for advancing their application in high-end manufacturing.
With the development of the electronic information industry and the intelligent revolution, China′s electronics sector has imposed increasingly stringent demands on the quality and supply of semiconductor raw materials. In the semiconductor industry, copper—particularly high-purity copper—demonstrates significant market potential due to its excellent physical and chemical properties, such as high electromigration resistance, electrical conductivity, thermal conductivity, ductility, low dielectric constant, and corrosion resistance. However, the quality and quantity of its production have consistently fallen short of industry requirements. Hydrometallurgy is the main process for preparing electrolytic copper, which is used for the production of high-purity copper. It enjoys advantages such as simple preparation method, mature process, low production cost, and low energy consumption, and thus dominates the high-purity copper production industry. However, during the process of preparing high-purity electrolytic copper, the differences in the preparation techniques and parameter control result in significant fluctuations in the quality of high-purity electrolytic copper. This paper discusses different electrolytic copper production routes, provides a detailed analysis of the advantages, disadvantages, and implementation cases of various production processes, and summarizes the strengths and challenges of ultra-high-purity metal preparation methods. Finally, based on the current status of production and demand for high-purity electrolytic copper in China, a forward-looking perspective is presented. Accelerating the industrial production and construction of high-purity electrolytic copper will be Accelerating the industrial-scale production and construction of high-purity electrolytic copper will be further conducive to accelerating the localization process of high-purity electrolytic copper needed for the development of China′s semiconductor industry and solving the problem of China′s reliance on imported raw materials in high-end, precision and cutting-edge fields.
Chalcopyrite is regarded as one of the most abundant copper minerals, yet its low leaching efficiency and slow kinetics under conventional metallurgical conditions severely restrict efficient resource utilization. Microbial leaching technology, recognized as an environmentally friendly and efficient metal extraction method, has attracted widespread attention in recent years for chalcopyrite processing. Primary reaction mechanisms of chalcopyrite bioleaching are reviewed, covering direct leaching, indirect leaching, and synergistic effects induced by microbial metabolites. The essential role of microorganisms in mineral crystal structure destruction and metal release is highlighted. Commonly employed strains, including Acidithiobacillus ferrooxidans, Leptospirillum ferrooxidans, and sulfur-oxidizing bacteria, are summarized, and their adaptability and synergistic performance in different leaching systems are analyzed. Key factors that influence bioleaching efficiency, such as ore mineralogy, chemical and environmental parameters, and catalytic leaching systems, are systematically discussed. Recent advances in microbe-mineral interfacial reactions, multi-species cooperative metabolism, and electrochemical regulation are integrated. Construction of high-performance composite microbial consortia, optimization of pretreatment processes, and development of industrial-scale reactors are identified as critical directions for enhancing chalcopyrite bioleaching efficiency.
Inclusions are a key factor affecting the quality of continuous casting slabs. However, the behavior and removal of inclusions in molten steel during the tundish changeover process—the final step before the steel enters the mold—remain unclear. This study investigated a two-strand tundish equipped with a turbulence inhibitor and weir-dam baffle at a plant. Using the volume of fluid (VOF) model coupled with the discrete phase model (DPM), numerical simulations were performed to analyze the three-phase flow of steel, slag, and air, as well as the removal of inclusions ranging from 2 to 200 μm, during the three stages of tundish changeover: casting stop, tundish opening, and re-steady state. The results indicate that during the casting stop and tundish opening stages, the total removal rate of 10 μm inclusions is the lowest, at 16.28% and 41.7%, respectively. In the re-steady state stage, the total removal rate of 5 μm inclusions is the lowest, measuring 67.4%. Regarding the escape rate of inclusions, during the casting stop and tundish opening stages, the escape rates of 10 μm inclusions are 17.49% and 7.94%, respectively. In the re-steady state stage, 5 μm inclusions exhibit the highest escape rate, at 21.4%. Across all stages of the ladle changeover process, inclusions larger than 100 μm are readily captured by the top slag, with total removal rates reaching between 93.4% and 98.5%. In summary, the study on the movement and removal laws of inclusions with different particle sizes and the flow state of molten steel is of great significance for improving molten steel cleanliness.
To investigate the feasibility of preheating ladles using oxygen-enriched combustion of converter gas for the purpose of recovering and utilizing converter gas, and to explore the effects of factors such as combustion time, epoxy ratio, and nozzle height on the preheating performance, this study employs Fluent software to conduct simulation calculations on the oxygen-enriched combustion preheating of ladles with converter gas. By analyzing the velocity distribution, residual oxygen distribution, and temperature distribution inside the ladle after preheating under different operating conditions, the optimal preheating operating condition was determined, and the influence laws of factors including combustion time, epoxy ratio, and nozzle height on the preheating performance were analyzed. The results show that the stability of the combustion state proves that the longer the combustion time is, the more sufficient the combustion is, and the better the preheating effect of the hot metal ladle is. However, the increase of combustion time will increase the combustion cost. The combustion state tends to be stable when burning for 5 min, and the optimal combustion time is 5 min. As the epoxy ratio increases, oxygen can be more evenly distributed in the combustion zone, promoting full combustion of the fuel and improving heat transfer efficiency, resulting in better temperature uniformity. However, if the epoxy ratio is too large, the oxygen concentration at the center decreases, leading to a decrease in combustion efficiency. When the epoxy ratio is 4∶1, the average temperature of the ladle is highest and the preheating effect is optimal. As the nozzle height decreases, high-temperature flue gas can penetrate deeper into the bottom of the ladle, enhancing the preheating effect at the bottom. However, lowering the nozzle position will compress the combustion space of the converter gas, which is not conducive to gas combustion. If the nozzle position is too low, the flame impact will be too concentrated, resulting in local high temperatures, while other areas have relatively low temperatures, affecting the overall preheating effect. Overall, the best preheating effect is achieved when the nozzle is positioned 1/4 of the height inside the bag from the bag mouth.
With the acceleration of the green and low-carbon transformation in the steel industry, direct reduced iron (DRI), as a clean raw material for electric arc furnaces (EAF) smelting, plays a significant role in reducing carbon emissions and energy consumption. However, the impact of a large proportion of DRI in the EAF on smelting parameters and costs is still unclear. Based on production data, a material energy balance model based on the law of conservation of mass and energy was established. The influence of a large proportion of DRI on the smelting parameters was studied, and the smelting costs were also calculated. The analysis shows that, for every 20% increase in the mass fraction of DRI, the slag volume, steel material consumption, oxygen and lime consumption increase by 28.7, 26.5, 7.2 and 8.3 kg/t respectively, and the electricity consumption increases by 30.2 kW·h/t, and the metal yield rate decreases by 2.1%. When the ratio of DRI is 100%, for every 1% increase in carbon content, the oxygen consumption increases by 14.6 kg/t, the electricity consumption decreases by 8.3 kW·h/t; for every 2% increase in metalization rate, the electricity consumption decreases by 25.2 kW·h/t; for every 100 ℃ increase in the preheating temperature, the electricity consumption decreases by 19.4 kW·h/t. When the scrap price is 2.4, 2.7, 3.0 and 3.3 yuan/kg, according to the smelting cost calculation, the price of DRI is below 2.03, 2.3, 2.56 and 2.83 yuan/kg, respectively, the EAF using DRI will be more cost-competitive.
Non-metallic inclusions, as foreign phases within bearing steel, tend to generate stress concentration and plastic strain accumulation in their surrounding regions under external loads, making them important initiation sites for rolling contact fatigue (RCF) cracks. To examine the effects of inclusion type and mechanical parameters on RCF damage in bearing steel, a rolling-contact polycrystalline finite-element model was established on the Abaqus platform in conjunction with the abqVoronoi3D plug-in, and a UMAT user subroutine was employed to incorporate a crystal-plasticity constitutive law. The stress distribution and equivalent fatigue damage characteristics of single inclusions, inclusions with different elastic parameters, and composite inclusions under Hertzian loading were analyzed. The results indicated that different types of inclusions induced varying levels of equivalent fatigue damage, after ten cycles, hard inclusions such as Al2O3 produced damage values up to 0.04, whereas deformable inclusions such as MnS remained below 0.03. Fatigue damage increased markedly with the mismatch in elastic modulus between inclusion and matrix, while the influence of Poisson′s ratio was minor. Composite inclusions exhibited a "damage-transfer" mechanism, whereby the soft outer layer absorbed part of the deformation energy during early cycles, but stress transmission to the rigid core amplified damage, leading to an overall damage level higher than that caused by single soft inclusions. These findings provide a theoretical basis for the design and quality control of long-life bearing steels.
The new energy vehicle trend is driving the development of automotive structural components towards lightweight and high safety. The dual-phase steel, which is the most widely used material for automotive steels, faces the challenge of increasing strength but deteriorating hole-expansion performance, which leads to vulnerable edge cracking phenomenon of automotive parts during the flanging and forming process. In this study, a 1 000 MPa grade Ti-bearing dual-phase steel was developed with the aim of improving the strength and hole-expansion properties by refining the grain size and reducing the strength difference between ferrite and martensite. The results show that a large number of TiC particles with an average diameter of 8 nm are successfully introduced into the designed DP steel (Ti bearing-DP) after the addition of Ti, resulting in the refinement of the ferrite grain size from 3.2 μm in the initial DP steel (Ti free-DP) to 1.7 μm. Due to the effects of precipitation strengthening and fine grain strengthening of TiC, the yield strength and ultimate tensile strength of Ti bearing-DP are significantly increased from 643 MPa and 1 023 MPa to 841 MPa and 1 140 MPa, respectively, compared with Ti free-DP. Although the introduction of TiC reduced the strength difference between the two phases, thereby improving their compatibility in deformation, it also decreased the work hardening capacity and uniform elongation of the Ti-bearing DP steel. However, the enhanced deformation compatibility and microstructural refinement significantly suppressed interface decohesion and martensite fracture during necking deformation, resulting in improved post-uniform elongation. Consequently, the Ti-bearing DP steel exhibited a total elongation comparable to that of the Ti-free DP steel. Furthermore, the improved deformation compatibility and refined microstructure effectively inhibited interface decohesion and martensite fracture during the hole expansion process, leading to a notable increase in the hole expansion ratio of the Ti-bearing DP steel from 13% to 21%, an improvement of 61.5% compared to the Ti-free DP steel.
The roll bonding process of U3Si2-Al fuel plates is critical to achieving a metallurgical bond between the core and cladding. However, uneven thickness at the front and tail of the core seriously affects the geometric accuracy and service reliability of the fuel plate. Its precise control is influenced by the coupling of multiple process parameters, and there is currently no effective regulation method available. To address this, a two-dimensional simulation model of the fuel plate roll bonding process was established based on simulation software to systematically investigate the influence of key process parameters—such as initial core thickness, single-pass reduction rate, roll bonding temperature, and pass reduction distribution—on the thickened regions at the front and tail of the core. The results indicate that a single-pass reduction of 20%-25% and a roll-bonding temperature of 400-450 ℃ significantly improve the thickness uniformity of the U3Si2-Al fuel plate core. Multi-pass reduction distribution further enhances thickness homogeneity and facilitates maintaining the front and tail thickness within acceptable tolerances. Because thickening at the front is less pronounced than at the tail, reciprocating rolling effectively suppresses the formation of "dog-bone" and "fish-tail" defects in the core. The study systematically reveals the influence of roll bonding process parameters on core thickness uniformity, providing theoretical guidance for the optimization of actual production processes. This is of great significance for improving the forming accuracy of fuel plates and ensuring their service safety.
To understand the phase transformation behavior of cryogenic steel bars for liquefied natural gas (LNG) storage tanks under different temperatures and cooling rates, dilatation-temperature curves at various cooling rates were measured using a Gleeble-3800 thermal simulation tester, and phase transformation temperatures were determined by the tangent method. Combined with optical microscopy (OM), scanning electron microscopy (SEM), and hardness testing, the microstructural evolution and austenite phase transformation behavior during continuous cooling were systematically investigated. The continuous cooling transformation (CCT) curve was plotted, and the kinetics of bainitic transformation were analyzed. The results show that the microstructure consists of ferrite and bainite in the cooling rate range of 0.1-5 ℃/s. As the cooling rate increases, the bainite content increases continuously, and the hardness rises from 180HV to 223HV. In the range of 5-20 ℃/s, the increase in undercooling provides a greater driving force for phase transformation, resulting in a continuous increase in the bainite nucleation rate and microstructural refinement, with hardness increasing from 223HV to 238HV. When the cooling rate exceeds 25 ℃/s, the combined effect of increased martensite content and refined bainite leads to a peak hardness of 280HV at 40 ℃/s. The optimal controlled cooling range for LNG cryogenic steel bars is 5-20 ℃/s, within which a fully bainitic microstructure and stable hardness are obtained. This study provides a theoretical basis and process guidance for the development and production of cryogenic steel bars used in LNG storage tanks.
Non-metallic inclusions significantly affect the mechanical properties of steel. Accurate characterization of inclusions in steel is essential for achieving precise control during production. Both 2D and 3D methods were employed to comprehensively analyze inclusions in 30MnSi steel bars in the current study. The research outcomes provide technical support for optimizing the production process and enhancing material performance of 30MnSi steel bars, thereby ensuring their safe and reliable application in engineering practices. The 2D characterization method will destroy the integrity of inclusions during sample preparation, but it can be used to statistically measure the sizes of inclusions on the sample surface and determine the type of inclusions through energy spectrum analysis. In contrast, 3D characterization method preserves the complete 3D morphological and spatial distribution information of inclusions, though it cannot accurately determine their types. 2D characterization result shows that the number densities of inclusions are 2.76 pieces/mm2 on the cross section and 3.20 pieces/mm2 on the longitudinal section, and the average equivalent circular diameters of inclusions are 2.51 μm on the cross section and 2.53 μm on the longitudinal section for sample 1. Moreover, the number densities of inclusions are 15.89 pieces/mm2 on the cross section and 62.15 pieces/mm2 on the longitudinal section, and the average equivalent circular diameters of inclusions are 2.05 μm on the cross section and 1.67 μm on the longitudinal section for sample 2. 3D characterization reveals that the number densities of inclusions, average volume, median Feret diameter of inclusion are 714.3 pieces/mm2, 67.7 μm3, and 5.29 μm for sample 1, respectively. However, the number densities of inclusions, average volume, median Feret diameter of inclusion are 668.4 pieces/mm2, 130.0 μm3, and 6.12 μm for sample 2, respectively. The results indicate that large inclusions (>100 μm) in bars are predominantly Mn-Ca-Si-O type inclusions deformed along the rolling direction. However, the presence of non-deformable cores within these inclusions is identified as the primary factor contributing to decreased yield strength.
To address the issues of low alumina dissolution rate in the Bayer process due to the intercalation of aluminum and iron in high-iron bauxite, and the difficulties in red mud settling caused by the presence of goethite, oxalic acid and ammonium oxalate are proposed as iron removal agents. The response surface methodology was used to optimize the process conditions for the removal of interlocked iron and aluminum, and the resource utilization pathways of the leaching solution were systematically explored. The test results show that the optimal conditions for iron removal are as follows. The molar ratio of H2C2O4 to (NH4)2C2O4 is 0.52. The total amount of C2O42- is 2.40 times the required amount. The liquid-to-solid ratio is 6.13 mL/g. The temperature is 95 ℃, and the time is 90 min. Under these conditions, the mass fractions of iron and aluminum in the leached residue are 0.27% and 28.00%, respectively. The leaching rates of iron and aluminum are 88.19% and 17.78%, respectively. In the leaching solution, Fe(Ⅲ) and Al(Ⅲ) exist in the forms of Fe(C2O4)33- and Al(C2O4)33-. Ammonia water can be used to precipitate iron, aluminum, and silicon from the leachate while simultaneously regenerating (NH4)2C2O4. Al(OH)3 can be removed from the mixed precipitate using NaOH solution to obtain sodium aluminate solution. Finally, Fe3O4 can be prepared by hydrothermal reduction of Fe(OH)3 with iron powder, while regenerating NaOH. The leaching residue after iron removal and the sodium aluminate solution can be returned to the Bayer process for pulp preparation or sintering process and dissolution process of sintered clinker, achieving effective integration with the Bayer process or sintering process. (NH4)2C2O4is used in the iron removal process for high-iron bauxite, and the regenerated NaOH solution is used in the hydrothermal reduction process to prepare Fe3O4. (NH4)2C2O4 is used in the iron removal process of high-iron bauxite, while the regenerated NaOH is used for the hydrothermal reduction to prepare Fe3O4. This process uses H2C2O4 and (NH4)2C2O4 as iron removal agents. Not only are the reaction conditions mild, but the process also enables the regeneration of reagents in the leaching solution and the recovery of iron resources. It is a green and efficient resource utilization technology, providing an innovative idea for the full hydrometallurgical treatment of high-iron bauxite.
Scenarios in the special steel industry exhibit both discrete and continuous characteristics, featuring dense material distribution, variable specifications, complex process routes, and frequent abnormal operating conditions. Traditional logical tracking and simple identification methods are challenging to achieve accurate positioning, while conventional visual inspection is also limited to single-scenario judgment. To address the challenge of multi-target material tracking under complex boundary conditions, this study proposes a multi-target tracking method based on state modeling and multi-source fusion. Firstly, a state machine-based model for state reasoning and trajectory deduction is constructed. Combined with the Drools rule engine for standardized management of scenario tracking business logic, this model enables process-driven real-time positioning and state reasoning. Secondly, in regions with frequent abnormalities, the visual perception scheme DTE (Detection-Tracking-Enhanced) is proposed to achieve accurate multi-target recognition and cross-frame association. Finally, an improved Dempster-Shafer evidence fusion strategy is designed, and mechanisms for spatiotemporal alignment, dynamic weighting, and hierarchical anomaly response are further established. These mechanisms conduct bidirectional verification on logical and visual perception results, thereby enhancing the system′s fault tolerance and tracking reliability. A big data platform and a tracking system were built based on the actual steel pipe production line of a special steel enterprise, and experimental verification was conducted using real data from the steel pipe hot working process. Results show that the overall tracking accuracy reaches 99.98%, and the inference accuracy under complex operating conditions (e.g., high temperature, occlusion, and multi-process overlap) increases by more than 9.0%. This performance is significantly superior to that of single-modal algorithms and traditional fusion strategies, verifying the effectiveness and superiority of the proposed algorithm in resolving information conflicts. The method holds important application value for improving refined production management and optimizing multi-business operations.
Vanadium-titanium magnetite as an important strategic resource, holds an irreplaceable position in the modern steel industry. However, the efficiency of vanadium resource recovery in the blast furnace smelting process still needs to be improved. To address this, an intelligent prediction and optimization method that integrates metallurgical mechanisms and data-driven approaches has been proposesd. A prediction and regulation model for vanadium content in blast furnace molten iron based on multiple feature methods and TimesNet. A bimodal correlation analysis framework was constructed to quantify the linear and nonlinear correlations between each parameter and the vanadium content in molten iron through Pearson correlation coefficients and mutual information. Combined with the feature importance ranking and forward stepwise regression of the XGBoost model, the optimal feature subset was screened out. The results of the comparative experiments show that the XGBoost feature selection method increases the model′s prediction accuracy by 32% and reduces the prediction error by 50%. In terms of prediction methods, a TimesNet time series prediction model optimized based on the Transformer architecture was developed. This model performs multi-period feature extraction through Fourier transform, effectively captures the long-term dependency of process parameters by using two-dimensional time-frequency reconstruction technology and multi-head attention mechanism, and realizes the fusion of multi-scale features by adopting an adaptive weighted aggregation strategy. At the application level, a closed-loop optimization regulation system of "prediction-analysis-regulation" was constructed. After the model outputs the predicted value of vanadium content in molten iron, the influence distribution and action mechanism of each parameter on the vanadium content in molten iron are quantified by combining SHAP interpretability analysis, Sobol sensitivity analysis, and metallurgical process mechanisms, providing a theoretical basis for the stable regulation of vanadium content in molten iron through multi-parameter coordinated regulation. Industrial verification shows that under the premise of maintaining stable smelting conditions, the prediction error of this method is 0.000 1, and it ranks first in terms of prediction accuracy. In all key performance indicators, it is significantly superior to other prediction methods. After the model was applied, the prediction accuracy rate of vanadium content in molten iron is as high as 90%, which has steadily improved compared to when the model was mature, and its stability rate reached over 80%, which was 9% higher than before. This method provides a practical and feasible technical solution for the intelligent control and efficient utilization of resources in the smelting process of vanadium-titanium magnetite in blast furnaces, and has significant engineering application value.
As a critical functional component in ladle refining, the distribution characteristics of bubble group generated by porous plugs significantly influence the dispersion uniformity of alloying elements in molten steel. However, high-temperature conditions and visual obstructions pose major challenges to quantitative research. In this study, a physical simulation experimental platform was independently established. Using the laser sheet method and high-speed imaging, bubble group from diffuse porous plugs were captured. An image processing algorithm based on multi-threshold Laplacian-Gaussian (LoG) edge detection was developed to overcome uneven background intensity caused by multiple scattering effects in the laser sheet method. The influence of bottom gas flow rate on bubble cluster characteristics was systematically investigated. Results indicate that at low flow rates, bubbles exhibit high roundness, small numbers, and small diameters. As the flow rate increases, the average bubble roundness decreases, the number of bubbles rises, the diameter distribution broadens significantly, and bubble overlap and adhesion become more pronounced. With further increases in flow rate, changes in average roundness, number, and diameter distribution diminish. After segmenting stacked bubbles, the average roundness and number of bubbles increase, while both the average diameter and Sauter mean diameter decrease, confirming that the gas flow rate significantly affects bubble roundness, number, and diameter. This study achieves feature extraction and quantitative analysis of bubble group generated by diffuse porous plugs, providing theoretical support for numerical simulation of bottom-blowing processes and structural optimization of porous plugs.
Accurately predicting disc shear force represents a core challenge for process optimization in cold-rolled strip finishing lines. Traditional methods are inadequate for deciphering the complex coupling mechanisms among multiple parameters, while existing machine learning models suffer from weak physical interpretability and inefficient feature utilization. To overcome these limitations, this paper proposes an artificial neural network (ANN) framework integrating feature augmentation (FA) and Bayesian optimization (BO), termed BO-FA-ANN. Using production data from a domestic steel enterprise, we systematically compared ANNs with five ensemble learning models and identified ANN as the optimal base model (R2=0.958 6). To address the information dilution of key physical features in deep network layers, a feature augmentation strategy was designed: SHAP (Shapley Additive exPlanations) multi-model consistency analysis identified entry tension and overlap as core features, which were explicitly incorporated into deeper network layers to enhance physics-guided decision-making. To tackle hyperparameter adaptation challenges caused by increased model complexity after feature augmentation, Bayesian optimization was employed to adaptively search for the optimal architecture. The resulting BO-FA-ANN model achieved superior performance (R2=0.985 7, EMS=0.013 6), significantly outperforming single-optimization models. This study demonstrates that feature augmentation mitigates the attenuation of physical significance in deep networks, while Bayesian optimization ensures the generalizability of the augmented structure. Their synergy offers a new paradigm for high-precision modeling of complex industrial processes and lays a technical foundation for the quantitative control and real-time optimization of shearing processes.
XML
Related Articles
File
