Materials Processing Fundamentals 2021

In this paper, the influence of the structure of turbulence inhibitor on the flow of molten steel in tundish is studied. By means of numerical simulation, by changing the shape of turbulence inhibitor and adding eaves on the top, the different influences on the flow field, temperature field, inclusion removal rate, molten steel quality, and residence time distribution curve of the whole tundish and the injection zone are obtained. By comparing the effects of four different shapes of turbulence inhibitors, this paper provides guidance for the design of flow control device in tundish.

In metallurgical processing, non-metallic inclusions contaminating metallic materials are one highly relevant challenge. Bubble injection into molten metals boosts the inclusion control and removal, thus enhancing metal homogenisation and purification. Although this principle of bubble flotation has been used for a long time, the effects of bubble–inclusion interactions in molten metals are not yet well researched. Imaging measurements of multiphase metal flows are challenging for two main reasons: the metals’ high melting temperatures and their opaqueness for visible light. This work focuses on X-ray and neutron radiographic experiments employing low-melting gallium alloys laden with model particles smaller than 1 mm in diameter. Both, bubbles and particles, are visualised simultaneously with high spatial and temporal resolution in order to analyse their motions by tracking algorithms. We demonstrate the capability of time-resolved X-ray and neutron radiography to image multiphase flows in particle-laden and optically opaque liquid metals, thus contributing to pave the way for systematic investigations on bubble–inclusion interactions in molten metals.

In this paper, a single strand tundish is taken as the research object, and a three-dimensional mathematical model is established to analyze the effect of the long shroud structure on the flow field, temperature field, pressure field, and inclusions trajectories of molten steel in the tundish. The fluid flow patterns in the tundish are analyzed by comparing the residence time distribution curves of the four types of long shrouds: traditional long shroud vortex long shroud, dissipative long shroud, and trumpet-shaped long shroud. At the same time, the effects of different long shroud structures on the fluctuation of free surface are compared. The reasonable long shroud structure is conducive to obtaining uniform molten steel and provides theoretical guidance for actual production.

The aeration leaching process is a three-phase reaction of gas, liquid and solid. The solid phase is black opaque particles. It is difficult to observe the flow field and the distribution of each phase. The water model experiment does not involve chemical reactions, and the physical properties of the solid particles remain unchanged. The effect of particle changes on the flow field cannot be explored. This study is based on computational fluid dynamics (CFD) software, using the Euler–Euler three-fluid model. A 60 L stirred reactor was simulated, and the solid–liquid ratio in the tank was 1:4, ventilation rate Q = 1–2.5 m³/h, speed 400–500 rpm. This study simulated the particle changes in the real reaction process and analyzed the phase distribution and flow field.

The evolution behaviors of the type, amount, and size of inclusions in tinplate were studied by industrial experiments and thermodynamic calculations during RH refining process. The results indicated that Al₂O₃ inclusions were generated at first after Al addition. With the slag-steel and refractory-steel reactions, inclusions varied with the route as Al₂O₃ inclusions → CaO–Al₂O₃ system inclusions → CaO–MgO–Al₂O₃ system inclusions and Al₂O₃ inclusions → MgO–Al₂O₃ spinel inclusions → CaO–MgO–Al₂O₃ system inclusions. From Al addition to the end of RH refining, the number density of inclusions increased from 9.32 to 4.09 per mm². The total oxygen content in steel could be decreased from 450 to 37 ppm. The size of inclusions varied mainly less than 10 μm, the proportion of inclusions with size less than 5 μm gradually decreased by 15% and inclusions in the range of 5–10 μm increased by 10%. These inclusions with size less than 10 μm were difficult to be removed from liquid steel through the floatation. The inclusions were Al₂O₃ and CaO·Al₂O₃ inclusions with a high melting point at the end of RH refining. In order to obtain inclusions with a low melting point, the reasonable C/A mass ratio of the final RH slag is 1.10 through theoretical calculations.

Flocculants type polyacrylamide (PAM) plays an important role in mineral processing circuits. It is common practice to reuse water from the thickening stages to other unit operations such as grinding and flotation. In the case of molybdenite flotation, an important depressing effect on mineral particles has been demonstrated by the residual flocculant dissolved in the recycled water. The aim of this work was to evaluate the effect of the degradation of a PAM type of flocculant on the flotation recovery of molybdenite. The study considered experiments of microflotation, adsorption isotherms, electrophoretic mobility, and intrinsic viscosity measurements. Molecular dynamic simulations were used to look for mechanisms of interactions between PAM and molybdenite. The results showed that molybdenite recovery was less affected by the flocculant when PAM molecules were subjected to more intense conditions of mechanical shearing. This behavior is related to the adsorption of flocculant on the mineral and the disposition and shape anisotropy of the flocculant chains.

During the pyrometallurgical production of industrial metals such as ferromanganese in electric smelting furnaces, molten slag and alloy phases are removed from the unit by tapping at regular intervals. Intermixing of the two phases may occur during the latter stages of the tapping process, and if not carefully managed, it can result in significant alloy being lost to the waste slag by entrainment. This paper presents the results of a computational fluid dynamics study of the multiphase free surface fluid flow in tapping ladles, with a specific focus on the impact of various design and operational parameters on alloy losses to the slag.

Carbothermic reduction was conducted using coarse (sample A) and fine (sample B) Brazilian Linz–Donawitz (LD) steel sludges and a sample of Peruvian anthracitic metallurgical coke (AMC) at 82.5% of fixed carbon. Specimens of 1 g of sample A and sample B preliminarily mixed with AMC at Fe/C: 1/2 were weighted. Sample A/AMC and sample B/AMC, reduction time and reduction temperature were identified as factors or independent variables. Conversions (α) or % reduction were selected as dependent variables. The values of weight ratios for sample A/AMC and sample B/AMC (3/7 and 7/3, respectively), reduction temperatures of 600 and 900 °C and reduction times lapses of 30 and 60 min, corresponding to their minimum as well as their maximum level of fluctuation respectively, were carried out via two design of experiments based in the factorial method 2³; one for each sample (A and B); using a Brazilian statistics software called COLMEIA—Snedecor algorithm F in order to evaluate the effects simple, double and multiple of factors over the conversions (α) or % Reduction. Carbothermic reduction tests were performed at the optimal weight ratio (sample A/AMC: 3/7 and sample B/AMC: 3/7), reduction temperatures: 600, 700, 800 and 900 °C and reduction times: 20, 30, 40, 50 and 60 min in order to estimate k-specific reaction rate constant, E_a-apparent activation energy and the A-Arrhenius pre-exponential frequency factor. The kinetic models that better fitted to the conversions (α) of both samples (A and B) were: boundary chemical reaction model for spherical symmetry (BCRM-ss): \(1-{\left(1-\mathrm{\alpha }\right)}^{1/3}=\mathrm{kt}\) and the model of simple exponential continuous reaction (MSECR): \(-\mathrm{ln}\left(1-\mathrm{\alpha }\right)=\mathrm{kt}\). The kinetic parameters obtained were: (1) sample A: E_a = 7. 45 − 8.08 kJ/mol and A = 0.009 − 0.032 Hz for a linear correlation between 0.8241 and 0.8276 and (2) sample B: E_a = 19.36 − 21.94 kJ/mol and A = 0.05 − 0.21 Hz for a linear correlation between 0.9758 and 0.9777.

Gibbs energy minimization is a powerful tool for modeling liquid–liquid equilibrium (LLE) in multicomponent systems, but it requires key thermodynamic parameters that may not be readily available. Our earlier work demonstrated an approach for estimating these properties via regression to equilibrium isotherm data but identified limitations regarding applicability to higher-concentration systems, sensitivity to experimental data, and ease of implementation. To extend the accessibility and applicability of Gibbs energy minimization for modeling LLE, this study presents the open-source, liquid–liquid equilibrium parameter estimator (LLEPE) package for estimating thermodynamic parameters. LLEPE extends parameter estimation to include aqueous-phase Pitzer coefficients that capture the effects of ionic interactions and offers analysis and visualization tools that assess experimental data quality and model prediction accuracy. We apply LLEPE to a rare earth extraction case study where we incorporate data filtering and Pitzer parameter fitting to reduce the effect of outlier data, capture the contribution of ion interactions, and ultimately improve overall prediction accuracy for multicomponent data. Our results show that filtering data decreases the prediction error by removing outliers, and fitting Pitzer parameters increases overall prediction accuracy, especially for high-concentration systems. Iteratively fitting to a combined dataset increases prediction accuracy for multicomponent data for low to medium feed concentrations.

Solvent extraction and electrowinning (SX/EW) play an important role during the hydrometallurgical production of copper for low-grade ores. To better understand the rapid Cu extraction/strip kinetics for producing a high-purity CuSO₄–H₂SO₄ electrolyte and elucidate the method to maximize current efficiency by preventing leach carryover to electrolyte during SX and minimizing the iron concentration, a detailed speciation study is required to quantitatively describe the iron and copper distribution in the SX/EW solutions. In the present study, by virtue of our previous modeling work, thermodynamic modeling of the SX/EW solutions is carried out to quantify the distribution of iron (in the form of ferric and ferrous), copper, and sulfuric acid. The developed model provides a mathematical tool capable of quantifying the concentrations of free ions and complexes in terms of temperature (25–60 °C) and solution composition under simulated industrial conditions. Finally, oxidation reduction potential measurements are employed to validate the model.

The silicate-based molten slag plays a crucial role in the process of continuous casting as functional materials. In the present study, the influences of electromagnetic field intensity and frequency on melting temperature of mould flux are revealed by an experimental method. To achieve this, a conductivity electrode and a measurement device are implemented to measure the melting temperatures under various magnetic field intensities and frequencies. The tested results show that with the increasing of magnetic field intensity (0–30 mT), the melting temperature of mould flux increases by 11–12 K. When the magnetic field intensity is 20 mT, the frequency increases by 6 Hz, and the melting temperature of the slag is increased by 0.4 –1.6 K. Additionally, it is found that according to the melt solidification theory, the potential barrier requires to be overcome when the flux melts from liquid phase to solid phase, and the barrier can be weakened when electromagnetic force is applied.

In this work, the dissolution of Ti ions from a sacrificial Ti anode during electrolysis on the reduction behavior of Ti–Al alloy electrodeposits from a Lewis acidic eutectic mixture of 1-ethyl-3-methylimidazolium chloride (EMIC) and a 0.667-mol fraction of aluminum chloride (AlCl₃) is investigated. The Ti ions are dissolved in EMIC-AlCl₃ ionic liquid (IL) by potentiostatic and galvanostatic electrolysis using chronoamperometry (CA) and chronopotentiometry (CP) techniques, respectively. At the same time, the electrodeposition of the Ti–Al alloy is accomplished on the copper cathode electrode at 383 K using the Ti anode. The dissolution, concentration, and deposition of Ti species are controlled by varying the electrolysis current, potential, and the electrolysis duration (1–3 h). The electrochemical reduction behavior of Ti and Al ions is studied on all Pt wire electrodes using cyclic voltammetry (CV). SEM studies revealed homogeneous and crystalline Ti–Al electrodeposits for CP-electrolysis. EDS and XRD revealed 16 at %. Ti with a cubic Ti_0.12Al_0.88 phase of Ti–Al alloy obtained from 1 h CP-electrolysis. The Ti content in Ti–Al alloy decreased with an increase in electrolysis time.

The ZrO₂ ceramic filter in the steelmaking plant in Nanjing Iron and Steel United Co., Ltd. was studied. Samples were taken from the tundish that ceramic filter installed. The total oxygen content and the type, shape, size, and composition of the non-metallic inclusions in steel were analyzed. Then the inclusion-removing capability of the tundish ceramic filter was evaluated. The results indicate that the effect of the ceramic filter on the filtration of inclusions is striking. The inclusion removal rate was 41.6% when the diameter of inclusions was more than 30 μm. The tundish guaranteed the production of high-quality steel. However, the calcium treatment process should be reasonably controlled to decrease secondary oxidation and improve the cleanliness of molten steel.

A hybrid model based on the combination of k-means method, BP neural network and decision tree algorithm is proposed for predicting the end-point phosphorus content of electric arc furnace. The industrial data from electric arc furnace is filtered firstly by the box-plot method, and the processed data of end-point phosphorus content is classified into three clusters by k-means analysis method. Then, three BP neural networks with different parameters for each cluster are established to deal with the data overlapping problem and increase the model accuracy. In order to obtain the optimum prediction result, a new method combined with the posterior knowledge of dephosphorization ratio and the decision tree algorithm is employed. With this method, the results predicted, respectively, by the three different BP neural networks are merged according to the merging rule set established by decision tree algorithm and the identified result is taken as the final end-point phosphorus content. In comparison with the traditional BP neural network and deep layer neural network, the hybrid model increases the prediction accuracy of end-point P content to 97.8% with ±0.006% error range, and meanwhile for the error ranges of ±0.005% and ±0.004%, the prediction accuracy is 94.2% and 83.0%, respectively.

Cleanliness control in steel has always been a topic discussed by researchers. Many methods were used to remove residual elements from steel. Adding rare earth elements to molten steel is one of the potential ways. However, to remove arsenic from steel using lanthanum, many issues need considerations. The effects of steel-crucible reactions, vaporation, and slags on the removal of arsenic from the molten steel were investigated. The results show that using the alumina crucible is more conducive to removing arsenic from the molten steel by rare earth elements, because LaAlO₃ can inhibit the further reactions between lanthanum and the crucible. Magnesia crucible, however, does not have similar effects. BN crucibles drop powders and decrease the cleanliness of steel. Besides, when the concentration of arsenic is low, arsenic can be barely removed by volatilizing under normal pressure. Furthermore, due to the existence of the interfacial tension, the molten slag is easy to separate from the molten steel and makes the fluidity of slag become worse. Although the molten slag can remove arsenic from the molten steel to some extent, it also has harmful corrosion effects on the crucibles.

Electromagnetic and electrostatic levitation techniques have been utilized for the last few decades to process metastable and highly reactive materials. Main fields of application include thermophysical properties, solidification, and transport phenomena of molten metals, which are of critical importance to better understand and control manufacturing processes dealing with molten metals, such as casting, welding, and additive manufacturing. This paper addresses part of our international collaborative work with NASA, European Space Agency, German Aerospace Center, and Japanese Aerospace Exploration Agency to support containerless experiments aboard the International Space Station. Along with space experiments, our recent work using an electrostatic levitator to unveil the liquid structure of aqueous sodium sulfate solutions (the most damaging salt to civil structures) is also introduced.

Due to unbalanced phases’ contents in welds of duplex stainless steel, their mechanical properties and corrosion resistance should be remarkably degraded from base metals’ ones. It is reported that both the precipitation and growth of the austenite phase in the welds are affected by the nitrogen content as well as thermal cycles of welding. Especially, in the case of laser beam welding (LBW), these effects might be drastically changed as compared with them in conventional welding methods. From the microstructure observation in the LB welds using argon shield gases varied with nitrogen contents of 0, 10, and 20%, austenite phases growth at grain boundaries are promoted by the increase in nitrogen content in the shield gases. Based on elements’ distributions measured in the previous report by an EPMA analysis, nitrogen is powerfully concentrated in austenite in each case. In addition, the in-situ observation by the micro-focused X-ray transmission imaging system revealed that porosity in the LB welds is suppressed by the increase of nitrogen in argon shield gases. These results suggested that under suitable thermal history conditions, the addition of nitrogen in argon shield gas for LBW might be effective to increase in austenite contents in welds of duplex stainless steels.

The viscosity evolution of vanadium slag using photovoltaic cutting waste as a reductant during the pre-reduction process was very important to kinetic condition and separation of Fe. However, there was little work reported on viscosity and model of FeO−SiO₂−V₂O₃−CaO−MnO−Cr₂O₃−TiO₂ system for the pre-reduction process of vanadium slag. In this study, the viscosity of FeO−SiO₂−V₂O₃−CaO−MnO−Cr₂O₃−TiO₂ system was experimentally measured by the cylinder method, and the Iida model was optimized by fitting experimental viscosity data. The results showed that the viscosity of the slag system increased from 0.120 to 0.473 Pa s as the FeO mass fraction decreased from 37 to 11% at 1873 K. Besides, the optimized Iida model estimated the viscosity, which had an average error value (Δ) of less than 11.40% from the experimentally measured viscosity, and the optimized Iida model also well predicted its subsidiary system of FeO−SiO₂−V₂O₃−TiO₂−Cr₂O₃ and FeO−SiO₂−V₂O₃−TiO₂, with Δ below 13.62% and 11.25%, respectively. For the vanadium slag with Cr₂O₃ mass% less than 5, the optimized Iida model could provide theoretical viscosity and it had a satisfactory agreement with experimentally measured results.

The presence of high content of Cr₂O₃ introduces a major impact on the smelting reduction and slagging processes in the blast furnace process. Therefore, the effect of Cr₂O₃ content on the phase of chromium-containing high-titanium blast furnace slag has been investigated in this work via XRD. The results show that when the added amount of Cr₂O₃ was a minimum of 1.5 wt.%, perovskite and spinel phases were formed in slag, resulting in increased viscosity. For good fluidity and thermal stability of blast furnace slag, the content of Cr₂O₃ in blast furnace slag should not exceed 0.5 wt.%.

The temperature and heat loss of molten slag during transportation has guiding significance for molten slag heating and heat preservation. In the present work, the effects of the thickness of a solidified slag layer and heat-dissipating caused by natural convection were numerically simulated. The results showed that the temperature of molten slag decreased and the slag density increased near the pot wall. Slag with higher density flowed down the pot wall and the slag with lower density flowed up to the slag surface, respectively. Circulation was formed between the pot wall and the center, which caused the higher temperature by ~200 K in the middle part of pot wall than the bottom. The maximum thickness of the slag layer ~28 mm appeared at the bottom, much higher than the thicknesses of 16.6 mm on the side. The slag upper surface is the main path of the energy dissipation and accounts for 81% of the total.

The dissolution behavior of chromium ore has severe effects on reduction rate of blast-furnace slag. Therefore, the study of dissolution behavior of Cr₂O₃ in the slag has great significance for accelerating the melting reduction rate of chromium ore and optimizing the blast-furnace smelting process. The effects of different rotation speed and temperature on the dissolution rate of Cr₂O₃ were studied by using the rod sample rotation method in the temperature range from 1500 to 1540 °C. The results of the ICP analysis show that the dissolution rate of Cr₂O₃ in the slag increases with rotating speed and temperature.

A number of finite element (FE) models with Lagrangian formulation have been developed to simulate the high-speed particle impact. Although the Lagrangian formation-based FE models developed so far contributed significantly to unveiling the important questions on the mechanics of particle impact, they have a critical weakness. The Lagrangian models often diverge under large deformation due to mesh distortion. Recently, FE models based on Eulerian formation have been developed. This research describes a part of such efforts. A thin three-dimensional finite element model of high-speed impact of a micron-sized metallic particle was developed using Eulerian formulation. The meshes of a finite element model based on Eulerian formulation do not deform and therefore allow for dealing with a case with extreme plastic deformation. The two cases from the literature were reproduced to evaluate the validity of the developed model. After comparison, the model was used to simulate the impact of a stainless steel particle on a stainless steel substrate at an impact speed of 1000 m/s, which could not be readily simulated using the Lagrangian model.

To increase on both of the capacity and defectless thin slab production, the submerged entry nozle (SEN) structure design and technology parameters of continuous casting were modified by studying the fluid flow and temperature distribution characteristics using physical and numerical simulation method. The results showed that the mold level fluctuation was strongly dependent to caster speed, and the average wave height exceeded 7 mm while the casting speed increased from 2.0 to 2.2 m/min for the section of 150 mm × 1540 mm. Compared to increase the angle of side-outlet or submerged depth of the SEN, the increase of the bottom port diameter of the three-port SEN was more effective on modification of level fluctuation although the maximal surface temperature was slightly decreased. Practical application also showed the effective performance of SEN design on improvement of the slab surface quality.

The equilibrium phase space of the Ag–Fe–X–Se (X: Ge, Sn) systems in the parts Ag₈XSe₆–XSe–FeSe₂–AgFeSe₂–Ag₈XSe₆ consists of four quaternary-phase regions formed with the participation of low-temperature modifications of the Ag₂FeGeSe₄ and Ag₂FeSnSe₄ compounds. The kinetic barriers of the formation of equilibrium four-phase regions that are observed under conditions of vacuum ampoule synthesis below 600 K were overcome by synthesis of phases at the positive electrodes of electrochemical cells (ECCs): (−)C | Ag | SE | R (Ag⁺) | PE | C(+), where C is graphite, Ag is left (negative) electrode, SE is the solid-state electrolyte, PE is the right (positive) electrode, and R (Ag⁺) is the region of Ag⁺ diffusion into PE. Silver cations Ag⁺ that shifted from the left to the right electrode of ECCs acted as the seed centers of an equilibrium set of phases. Based on the temperature dependences of the EMF of the cells in the temperature range 430–485 K, the standard thermodynamic functions of the Ag₂FeGeSe₄ and Ag₂FeSnSe₄ compounds were calculated for the first time. The observed experimental results and thermodynamic calculations are in good agreement.