The corrosion behaviors of corundum brick and carbon composite brick used in blast furnace hearth by CaO–SiO2–MgO–Al2O3–Cr2O3(-CaF2) slags were studied in the present work. The degradation of the corundum brick in slag was a result of slag infiltration and brick dissolution, and the corrosion of the brick became more serious with the addition of CaF2 due to the decrease of slag viscosity. The disintegration of carbon composite brick in CaF2-containing slag was caused by the combination of slag penetration, brick dissolution and reaction between slag and brick. By comparing the corrosion behavior in CaF2-containing slag between the corundum brick and carbon composite brick, the corrosion degree of the corundum brick was greater than that of the carbon composite brick. To the blast furnace operation in which a low grade iron ore such as laterite ore and CaF2 containing slag (about 2 wt%) are used, it was found that the carbon composite brick with better slag corrosion resistance can be selected as a hearth refractory so as to improve the operation performance and ensure the longer campaign life of blast furnace.
In order to extend the viscosity estimation of the CaO–MgO–Al2O3–FeO–SiO2 system and provide new methods to establish the more comprehensive and high-precision viscosity model, a structurally-based viscosity model for fully liquid aluminosilicate slags was established for the CaO–MgO–Al2O3–FeO–SiO2 system, and the results showed that the model could well link the slag viscosity to the internal structure of melts and the concentrations of components by determination of the integral activation energy of slag using the basic oxide, acidic oxide and electronegativity of structural units, the viscosity model showed overall superior performance for the CaO–MgO–Al2O3–FeO–SiO2 system, with the mean deviation for the quinary system of 10.4%. Meanwhile, the viscosity model also had good forecast effect for the quaternary system, ternary system and binary system containing silicate or aluminates, with the mean deviation for the three systems of 14.1%, 11.8% and 12.8%, respectively. The new model was expected to be widely used for different systems containing silicate or aluminate.
The influences of Na2O, K2O and Li2O additions on the electrical conductivity of CaO–SiO2–(Al2O3) melts are investigated by the four-electrode method. It is concluded that the electrical conductivity monotonously increases as increasing R2O (R=Li, Na, K) content at a fixed CaO/SiO2 ratio of 1.1. As gradually substituting Na2O or K2O for Li2O, respectively, the mixed alkali effect occurs, and there is a minimum value of electrical conductivity. Furthermore, the presence of Al2O3 promotes the mixed alkali effect. Raman spectroscopy is used to illustrate the relationship between electrical conductivity and melt structure. The electrical conductivity decreased with the increase of the degree of polymerization of melt.
The mixed alkali effect of silicate melts was investigated by performing 29Si magic-angle spinning–nuclear magnetic resonance (MAS-NMR) measurements and impedance spectroscopy analysis. The obtained 29Si MAS-NMR results showed that the Qn species of quenched glass was monotonically dependent on the ratio of the alkali mixture. In contrast, the equivalent circuit components given by impedance measurements of silicate melts composed of mixed alkalis were found to have extreme values. The 29Si MAS-NMR results suggested that these extreme values were unrelated to the structure of the silicate, but were related to the alkali present in the melt influencing the equivalent circuit components. This is thought to be caused by the increased pre-exponential factor, which is related to the concentration of ions with movement in the melt that is initiated by alkali mixing. It was also found that the equivalent circuit components of the super-cooled melt (i.e., the homogeneous melt just below the liquidus temperature) were significantly different from the equivalent circuit components of the homogeneous melt above the liquidus temperature. This is considered to be due to the fact that the activation energy of ion conduction in the super-cooled melt was increased by ~18%.
Increasing sinter productivity, product yield, and sinter strength are important for more efficient processes. Because the sintering bed structure should strongly affect such sintering performances, evaluation techniques and controlling methods have been studied. X-ray computed tomography (CT) was used in several studies, because a sintering bed can be observed without destroying it. Sintering bed shrinkage has been also focused on as an index of product yield or sinter strength, because shrinkage should indicate structural change of the sintering bed. However, shrinkage was only evaluated as a cumulative value at the end of sintering, and material behaviour in the sintering bed has not been clarified. In this study, the correlation of shrinkage and sintering bed structure by X-ray CT observation was investigated. Tracers were located in the sintering bed, and the vertical shrinkage distribution was derived from their height change. The result was compared with the shrinkage change with time. Several factors for shrinkage were considered, and their influence degree was discussed. The main factor was sintering with melt formation. This result, however, contradicted the positive effect of ‘stand’, which prevents shrinkage and improves productivity and product yield. The reason is discussed, focusing on the support effect of the stand and difference in shrinkage distribution in the sintering bed. A quench test at a certain time during sintering was also carried out, and the transition state from granule to sinter was evaluated.
A three-dimensional steady state mathematical model, considering the chemical reactions and the transfers of momentum, heat and mass between the gas and solid phases, is developed to investigate the performance of COREX central-gas-distribution (CGD) shaft furnace with top gas recycling (TGR). The model is validated first by data from practical measurement and then is used to study the performance of CGD shaft furnace with TGR. The results reveal that, compared with the process of 15% CGD gas input without TGR, the TGR can reduce the fresh gas consumption from 1058 Nm3/tBurden to 392 Nm3/tBurden. The reduction potential of top gas with TGR increases from 0.5755 to 0.6293, the utilization rate of top gas decreases from 34.18% to 28.03%, the metallization rate of solid product increases from 61.80% to 70.05%, and the standard deviation of metallization rate decreases from 0.8% to 0.4%.
Ironmaking process by Blast Furnace (BF) method contributes about 70% of CO2 emission and energy consumption in steel works. To aim at the sustainable development and environmental protection, decreasing CO2 emission and energy consumption is always an indispensable task. Using iron ore-coal composite as the raw material of BF is a promising way to mitigate the above issue.
In this study, we conducted an experiment procedure which is to simulate the reduction of an iron ore-coal composite travelling from the top zone to the cohesive zone of BF. Reduction behavior, morphology changes and carburization characteristic of the composites with the variables of C/O (Carbon to Oxygen) ratio were investigated.
As C/O ratio was increased from 0 to 0.6, the reduction rate of composite was accordingly enhanced. It was found that the composite is swelling severely when C/O ratio was lower than 0.4. The swelling occurred in the temperature range from 800°C to 1100°C which was just under the stage of reduction from wustite to iron. The composite is shattering obviously starting from 1160°C when C/O ratio was higher than 0.4. It is because the remained free carbon is surplus in reduced iron, causing shattering and powdering. Hence, C/O = 0.4 was suggested to the composite for being charged into BF.
The actual state and packing condition of deadman in blast furnace (BF) hearth have always been a matter of great concern. For it directly affects thermal state of BF hearth and relates to the smooth operation and longevity of BF. In this study, a commercial BF was dissected and then core drilling and image processing methods were used to obtain typical samples and characteristics of deadman. The results show that the deadman, with a radius of 80.09% of hearth radius, floated in hearth and surrounded by molten iron. The sample in horizontal direction has an average coke size of 28.04 mm and a void fraction of 0.52 while the vertical sample has an average coke size of 30.55 mm and a void fraction of 0.50. It should be noted that the average diameter of coke is the reference value which will be different depending on different assumptions. The coke size of vertical sample shows decreasing trendfrom topside to middle part and then keeps slight fluctuations.. And the void fraction in the middle of the vertical sample is the largest and gradually decreases toward both the upper and lower ends. Changes in these features of deadman are caused by the renewal of deadman and the flow of molten iron.
Changes in particle size distribution, mineral yield and strength of coke samples from various locations of two Chinese blast furnaces as well as deadman porosity were investigated in the present study for an in-depth understanding about the blast furnace hearth phenomenon. It was found that the percentage of <10 mm coke fines varied from 20% to 49% in majority of the hearth-level regions. The average size of hearth coke was about 20 mm–31 mm. Compared with the feed coke, the hearth coke size was observed to decrease by 43%–63%. The average size of hearth coke particles of a 2800 m3 blast furnace in diameter direction distributed in “M-shape” in majority of the hearth-level regions while that of a 5500 m3 blast furnace distributed in inversed “V-shape”. The hearth coke mass was 1.43–2.21 times of the feed coke under the same conditions. The M10 of hearth coke with size larger than 40 mm after drum test was about 11%–18% and the M40 was 75%–79%. The M10 increased with the increasing distance to the tuyere level while the M40 decreased with the distance. Due to the catalytic effect of hot metal on coke graphitization, the M10 of hearth coke in the lower part was increased by 63.6% compared with the coke in the upper part. The average porosity of the edge, the middle and the center areas was 0.334, 0.299 and 0.250, respectively. The average porosity of deadman decreased with the increase of distance to the center line of the taphole and the increasing distance to the furnace wall.
The evolution mechanism of oxide inclusions in Cr–Mn–Ni stainless steel was investigated by industrial trials and thermodynamic calculation. The morphology, composition, and size distribution of inclusions in steel specimens were analyzed by scanning electron microscopy and energy dispersive spectroscopy. During the LF refining process, there were mainly liquid Ca–Si–Mg–Al–O inclusions in molten steel deoxidized with FeSi alloy. Combined with the Al–Si–O phase diagram, the specimen compositions were also located in the liquid oxide phase. At the same Al content, increasing Si content could make the steel compositions in the liquid oxide phase to avoid the formation of Al2O3. After continuous casting, the number density of Ca–Si–Mg–Al–O inclusions decreased to 1.81 mm−2. On the contrary, the number density of Mn–(Al–Ti)–O inclusions increased to 4.62 mm−2. The MnO contents of most Mn–(Al–Ti)–O inclusions were higher than 40%. The size of most Mn–(Al–Ti)–O inclusions was smaller than 3 µm. The formation of these inclusions was consistent with thermodynamic calculation, which indicated that Mn–Al–O and Mn–Ti–O inclusions were formed during the solidification of Cr–Mn–Ni stainless steel. The effects of different Al and Ti contents on the formation of oxide inclusions during continuous casting process were discussed.
Laboratory experiments and thermodynamic calculation for the Al-killed Ti-stabilized 20Cr stainless steel with several CaO–MgO–SiO2–Al2O3–TiO2–CaF2 slags containing different CaF2 contents were performed to investigate the effect of slag composition on inclusions in molten steel. The titanium and magnesium contents were higher in the steel samples reacted with the slag samples containing higher CaF2 contents. The thermodynamic results based on the ion and molecule coexistence theory (IMCT) also indicated that the log (αSiO2/αTiO2), log(α2Al2O3/α3TiO2) and log(αAl2O3/α3MgO) decrease with the increase of CaF2 content in slag, which would make the molten steels have higher titanium and magnesium content. The increase of magnesium content in steel due to the increase of CaF2 in slag led to the increase of MgO content of inclusions. Due to the highest CaF2 content in slag, some inclusions in the steel were located in (liquid + MgO + TiSp) phase field. Reducing the CaF2 content in slag to 5.18 mass%, the MgO content of inclusions in steel was reduced, which made most of the inclusions located in or close to liquid oxide phase field. The reasonable CaF2 content in slag was discussed with the consideration of controlling the titanium content and the formation of inclusions in molten steel.
A kinetic model was developed to describe multicomponent reactions and mass transfer at the steel/molten flux interface under the effect of the interfacial tension. This model mainly describes the following interfacial physicochemical phenomena: i) Silica decomposition and oxygen adsorption at the interface, ii) Oxygen and titanium reactions at the interface, iii) Oxygen and aluminum reaction at the interface, iv) Silica mass transfer from the flux bulk to the interface, and v) Dissolution of the formed titanium dioxide and alumina into the flux and its transfer in flux. With this model, the dynamic changes of the mold flux composition, steel composition, interfacial oxygen content and interfacial tension for different mold flux compositions were predicted. Overall, the dynamic composition changes of the mold fluxes in a casting mold were reproduced. The basicity of the mold flux shows a large influence on the dynamic change of its composition. The initial composition change of the mold flux is fast when the flux with a high basicity was used, compared with the case of the mold flux with a low basicity. The interfacial oxygen content and the interfacial tension were found to reach a constant value after the steel/flux reaches a metastable state. In addition, the interfacial adsorption of oxygen due to the interfacial tension effect was found to significantly accelerate the dynamic change process of the steel/mold flux system.
Solidification shell deformations within the mold during continuous casting have been calculated in order to clarify the influence of mold flux infiltration variability on the cooling rate, the width of the low heat flux region, the height of air gap, the unevenness of solidified shell, and the resulting strain in the solidified shell. A sequentially coupled thermal-mechanical finite element model has been developed to perform the calculations. The simulation includes heat transfer and shell deformation in a growing solidified shell, along with the delta-to-gamma transformation. Further, it takes into account the effects of variability in mold flux infiltration and air gap formation on heat transfer into the mold, as well as the effect of cooling rate on the thermal expansion resulting from delta-to-gamma transformation. The results showed that mild cooling and small width of low heat flux region (i.e. low variability in mold flux infiltration) strongly decrease the height of the air gap, the unevenness in the solidified shell and the strain in the solidified shell. It is confirmed that it is important to prevent the variation and optimize the cooling rate in mold flux infiltration, especially at near the meniscus region of δ to γ transformation in order to minimize longitudinal crack formation.
Recognition of plate identification numbers (PINs) is of much importance for the automation of the iron and steel production. The recognition of PINs in industrial site is a challenging problem due to complicated background, low quality of characters. Conventional image processing algorithms are employed to extract the numbers, but it is difficult for these methods to locate and recognize the numbers on the plates in complex industrial production by manually designed features. The end-to-end convolution neural network is employed to solve these problems by automatically extracted features. These features seldom combine the real production rules. A delicate recognition method of PINs using convolution neural network and characters distribution rules is proposed. The PINs are roughly recognized by convolution neural network with Non-Maximum Suppression (NMS), and the PINs are exactly processed using the character distribution rules. Experiment results demonstrate that the method proposed can arrive at a very high recognition rate 96.32% and improve the recognition rate by 54.07% compared with the end-to-end convolution neural network.
The effects of process conditions, material properties, and initial shape of flaw on the deformation behavior of surface flaw during wire drawing process have been investigated to understand the deformation behavior of surface flaw and to find solutions to decrease them using the finite element (FE) analysis. The surface flaw decreased with decreasing die angle and friction coefficient, and with increasing reduction of area (RA) per pass. Interestingly, the surface flaw and strain inhomogeneity of wire rod were simultaneously decreased at the same process conditions. The shape of surface flaw was somewhat varied with strain hardening exponent (n) and almost same regardless of strain hardening coefficient (K), which means the effect of material properties such as K and n values is not critical. The higher flaw angle has a positive effect on removing or reducing the surface flaws during wire drawing process. The V shape surface flaw in wire rod was relatively well removed; whereas, U shape flaw tends to deform in the manner of overlap with increasing total RA, which means U shape surface flaw was much more detrimental than V shape flaw. To reduce or remove the surface flaw during wire drawing, homogeneous plastic deformation along the radial direction was necessary. Based on above results, seven strategies were proposed to remove or reduce the surface flaw of wire rod during wire drawing process, which can be helpful in determining the guideline to design the wire drawing process.
The influence of Si content in steel on tool wear in turning of 0.8 mass% C hardened steels with TiAlN coated CBN cutting tools is investigated. The Si contents in the steels are varied between 0.05 and 0.6 mass%. Although these steels have similar microstructure, hardness and volume fraction of retained austenite, the width of the flank wear of the tool increases with increasing the Si content. Adhered oxides are formed on the flank face after cutting, and the amounts, compositions and crystal structures of these oxides are changed as the Si content is varied. The higher Si content results in large amounts of adhered oxides. The crystalline oxide containing a large amount of Fe is mainly formed when cutting the 0.2 mass% Si steel, while the amorphous oxide containing a large amount of Si is mainly formed when cutting the 0.6 mass% Si steel. At the interfaces between the tool coating and the adhered oxides, the Al element of the tool coating tends to diffuse more easily into the Si containing amorphous oxide than into the Fe containing crystal oxide. This indicates that the Si containing amorphous oxide, formed with the higher Si added steel, promotes diffusion wear, resulting in increased tool wear.
Resistance spot welding (RSW), as one of the most widely used processes in sheet metal fabrication, is a complex electromechanical coupled nonlinear process, and weld quality is influenced by various process conditions, noise and errors. Therefore, inconsistent quality from weld to weld is a major problem for RSW process. However, so far there is no satisfactory non-destructive quality estimation method to evaluate the weld quality.
The objective of this study is to explore a quality estimation system for RSW. In order to build the quality estimation system, the relationship between various variables during RSW process and weld quality was studied by lots of experiments. Test results showed that the system built in this study could accurately estimate the weld quality and the maximum estimation error was only 5.6%.
Bead-on- plate welding was performed on 20 mm thick cast ductile iron plate in shielded metal arc welding using three different levels (0.05, 0.1, 0.2%) of Ce and one without Ce containing developed coated electrodes after establishing weld procedure as per AWS (D11). As-welded DI weldments were given isothermal heat treatment at two different austempering temperatures (300°C and 350°C) for 1.5 h, 2 h and 2.5 h holding time. Microstructures of as-deposited and austempered weld metals were characterized by OM, SEM, TEM, XRD analysis and microhardness testing. Results show that weld metal containing 0.1% Ce attributed lowest amount of ledeburitic carbide in as-deposited condition and maximum vol.% of retained austenite (46.7%) after austempering heat treatment compare to other two levels of Ce. Since maximum amount of retained austenite is the desirable microstructure after austempering heat treatment, 0.1% Ce is considered as the optimum one in DI weld metal.
The degree of sensitization in an austenitic stainless steel, has been measured using double loop electrochemical reactivation tests, and the measured values compared with predictions based on grain boundary chromium depletion characteristics obtained using the precipitation and diffusion modules of Thermo-Calc. In order to quantitatively predict Cr depletion, the precipitation of M23C6 carbides that are responsible for sensitization has been modelled for isothermal conditions by treating nucleation and growth separately. Based on a critical chromium concentration, a depletion parameter that predicts both sensitization and self-healing is given.
The effects of fine precipitates on the austenite grain refinement of micro-alloyed steel during cyclic heat treatment were investigated under different solution treatments. After three rounds of cyclic heat treatment of Ac3 and Ar3 transformations of the as-received steel rod with rapid heating and cooling, the austenite grain size was 3–10 µm. On the other hand, three rounds of cyclic heat treatment after solution treatment at 1300°C reduced the austenite grain size to 2–5 µm. The as-received sample included AlN–Ti(C,N)–MnS composite particles with a mean diameter of 30 nm and a number density of 11 µm−3, and the mean diameter did not change during cyclic heat treatment. Thus, it was considered that the reduction in austenite grain size without solution treatment was caused by the increase in the nucleation site of austenite phase with increasing number of cycles, due to the refinement of the prior austenite grains with martensitic structure during the cyclic heat treatment. When solution-treated at 1300°C, the AlN–Ti(C,N)–MnS composite particles were solved, and they were precipitated during the cyclic heat treatment with a mean diameter of 12 nm and an increased number density of 85 µm−3. Thus, it was considered that the further reduction in austenite grain size with solution treatment was caused by the pinning effect of the fine precipitates, in addition to the increase in the number of austenite phase nucleation sites with increasing number of heat treatment cycles.
HAADF-STEM observations of (a) solution-treated at 1300°C, (b) 780°C × 1 after solution treatment, and (c) 780°C × 3 after solution treatment.
In order to improve the mechanical properties of AISI304 stainless thin plate, a low-pressure, repeated explosive hardening process was carried out using the explosive treatment technique. Both the residual microstructures and mechanical properties of the treated material were investigated and compared with the undeformed sample. The results show that microhardness, yield strength and ultimate strength of the treated sample remarkably enhanced during a low-pressure, repeated explosive hardening. We propose that the strengthening mechanisms are closely related to the microstructure evolution during the low-pressure, repeated explosive hardening. The corresponding microstructures in the twice hardening samples were dominated by deformation twins, α martensites, with a few deformation bands. The microstructures in the four times hardening samples were characterized by deformation bands and α martensites, where volume fraction of martensite was larger than that in the twice hardening sample.
A novel method for treating the Sb-bearing slag using LDPE as a reductant under N2 atmosphere is proposed. The H2, CH4 and C2 (C2H6, C2H4, C2H2) are released quickly during the LDPE pyrolysis process, and some of them have escaped off the roasting system before the reaction with the Sb-bearing slag. The Sb, Pb and Fe phases in the original Sb-bearing slag could be selectively reduced to Sb4O6, Pb and Fe3O4 respectively by the pyrolysis gas at roasting temperature of 1000°C and LDPE amount of 25 wt%. The Sb4O6 could be separated and recovered effectively with the volatilization rate of 93.51% accompanying the Pb volatilization rate was only around 0.68% and the Fe phase was almost transformed into Fe3O4 retaining in the residue. Then the Fe3O4 can be separated and collected from the roasted residues using magnetic separation process.
In this study, we conducted trace element analysis using laser-ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) to quantitatively analyze the distribution of tramp elements and their coexistence states with other elements in recycled steel materials. The concentration distributions of Cu, Sn, Pb, Ag, and Au in a 455 µm × 270 µm region were measured using electric furnace steel as a sample by continuous line analysis. Although the concentration of Cu was the highest among the measured elements, the degree of variation in Cu concentration was the smallest. Meanwhile, not only were the degrees of variation of the Sn, Pb, and Ag concentrations large, but also the points of high concentration measurements tended to coincide. The coexistence states were investigated by determining the coefficient of correlation between the concentrations of two elements.
Microstructure of M2 high speed steel (HSS) after electropulsing treatment (EPT) was studied. The secondary carbides precipitated during pre-tempering were dissolved completely during EPT. But the secondary carbides in the starting microstructure cannot be dissolved completely upon conventional reheating to the temperature same as that during EPT. The austenite grain size was ultra-refined by EPT. This study provides a new way to refine austenite grain size in HSS without diminishing the dissolution amount of carbides.