Effect of Single Power Two Circuits Electroslag Remelting Process on the Cleanliness of the Remelted Ingot

Abstract

Single power two circuits electroslag remelting process with current carrying mould (ESR-STCCM) has been developed to remelt high alloy. In the present work, the laboratory experiments, physical simulations and numerical simulations were set up to systematically investigate the droplet size and cleanliness of the remelted ingot for ESR withdrawing process (ESRW) and ESR-STCCM. The results indicated that ESR-STCCM can change the distribution of electromagnetic force, thereby reducing the droplet size in the case of the same remelting power. ASPEX explorer was utilized to investigate the non-metallic inclusions of the remelted ingot for different remelting processes, and the result indicated that the types of the non-metallic inclusions for the different remelting processes were not changed, however, the number decreased by 42.3% for ESR-STCCM. Compared with the ESRW, the deoxidation ability of ESR-STCCM increased by 10.7% meanwhile, the desulfurization ability increased by 24.5%.

1. Introduction

ESR is one of the most effective secondary refining methods and use for remelting a multitude of steel.¹⁾ During the ESR process, metal droplet was formed and dripped from the tip of the consumable electrode under the influence of gravity, interfacial tension, electromagnetic force, thermal buoyancy and other forces.^2,3) During the formation and detachment processes, metal droplet occurred complex physicochemical reactions, which result in the favorable ability to remove non-metallic inclusions. The process of removing non-metallic inclusions for ESR process can be divided into three stages: (a) metal droplets form at the tip of the consumable electrode; (b) metal droplets drip through the liquid slag pool; (c) non-metallic inclusions float upward from the molten metal pool. According to the experiment results done by Li Z B, et al.,⁴⁾ the formation and detachment processes are the main stages to remove non-metallic inclusions, especially during the formation stage.

The sulfur content of the remelted ingot is one of the critical factors affecting its mechanical properties. S can form MnS non-metallic inclusion at the grain boundary with manganese (Mn) which can be one of the factors that cause the ductile fractures.⁵⁾ Therefore, the sulfur content is an important index of the ingot quality. There are two methods to desulfurize during the ESR process - gasifying desulfurization and slag-metal reaction desulfurization. Previous study^6,7) revealed that gasifying desulfurization plays a more dominant role than slag-metal reaction desulfurization. Gasifying desulfurization is the reaction between liquid slag and air, in addition, the metal droplet formation and detachment processes are the main processes for slag-metal reaction desulfurization.

The diameter of the metal droplet has an extremely important effect on the slag-metal reaction. The smaller diameter of the metal droplet will result in the larger interface area, therefore, the efficiency of removing non-metallic inclusions through the slag-metal reaction will be increased. According to the previous studies,^8,9,10) the diameter of the metal droplet was affected by several parameters such as interfacial tension, power, slag composition, etc.

With the advancement and development of ESR, current carrying mould (CCM) was proposed^11,12) and used for electroslag surfacing with liquid metal (ESS LM),¹³⁾ bimetallic composite roll^14,15) and solid ingot.^16,17) Besides, single power two circuits electroslag remelting process with current carrying mould (ESR-STCCM) had been proved to be an effective method to improve the ingot solidification quality.^18,19) The electromagnetic field, flow field, temperature field of the liquid slag pool and shape of the molten metal pool for ESR-STCCM were different from that of conventional ESR, therefore, the shape of the consumable electrode tip and the metal droplet size must be different. Hence, the remove non-metallic inclusions abilities for different remelting processes were no doubt different.

As ESR is the high temperature and opacity process, it is impossible to observe the behavior of the metal droplet directly. Therefore, the transparent container was used for visibly observing the formation and detachment behavior of the metal droplet. Cao Y L, et al.²⁰⁾ simulated the metal droplet behavior of conventional ESR as well as the influence of the process parameters on the metal droplet with wood alloy and NaCl solution. Wang H, et al.²¹⁾ investigated the influence of the current frequency on the metal droplet evolution during magnetic-field-controlled ESR process with zinc bar and ZnCl₂ solution.

Previous researches shown that the current can influence the numbers of the inclusions in the remelted ingot.^22,23) Wang Q, et al.’s work²²⁾ found that, the final oxygen mass percent in the metal increases with a higher current. Li Q, et al.²³⁾ found the size of the droplet and the number of the inclusions were dramatically changed with different static magnetic field for magnetically controlled electroslag remelting process. Wang H, et al.’s work²⁴⁾ found ESR can dramatically decrease the size of the inclusions, and it is hard for ESR to remove Al₂O₃ inclusions. Li S J, et al.²⁵⁾ investigated the evolution of oxide inclusions in G20CrNi2Mo carburized bearing steel during industrial ESR. They found that most inclusions were absorbed at the tip of the electrode, and some pure Al₂O₃ inclusions can generate in the dripping process because of the oxygen transfer. Hou D, et al.²⁶⁾ found that the content of electrode and slag can strongly effect the inclusions. However, these researches are focus on the traditional ESR. For ESR-STCCM, the electromagnetic field was dramatically different with conventional ESR.^16,17) Therefore, the cleanliness of the remelted ingot must be significantly different. According to the existing knowledge, the effects of the remelting process on the cleanliness of the remelted ingot are still scarce. In the present work, the laboratory experiments, physical simulations and numerical simulations were developed to investigated the remove non-metallic inclusions ability for different remelting processes. The laboratory experiments were used for revealing the cleanliness of the remelted ingot for the different processes. The dimension of the T shape mould for the laboratory experiments was 333 mm/260 mm. Besides, the physical simulation was used for simulating the formation and detachment processes of the metal droplet by wood alloy and NaCl solution. Furthermore, electromagnetic field, flow field and temperature field were calculated by numerical simulation.

2. Experiments

2.1. ESR Experiments

The schematic diagram of different remelting processes was shown in Fig. 1. ESRW and ESR-STCCM experiments were carried out using a T shape current carrying mould with the dimension of Φ333 mm/260 mm. GCr15 bearing steel was used as the consumable electrode with the dimension of Φ150 mm×2000 mm (manufactured by Benxi Iron & Steel Corporation) and the component was shown in Table 1. There was only one circuit for ESRW (the circuit is power → consumable electrode → liquid slag pool → ingot → power). However, there were two circuits for ESR-STCCM (one of the circuits is the ESRW and the other is power → conductor → liquid slag pool → ingot → power). The details of the remelting processes had been described in our previous study.¹⁹⁾ When the slag and air interface (slag-surface) was submerging the up mould (Φ333 mm) more than 40 mm, the withdrawing operation was started. The remelting parameters were shown in Table 2.

Fig. 1.

Schematic diagram of the remelting process: (a) ESRW, (b) ESR-STCCM. (Online version in color.)

Table 1. Chemical compositions of the consumable electrode (Mass, %).

Element	C	Si	Mn	Cr	S	O	Fe
Content	0.99	0.20	0.29	1.47	0.00350	0.00163	Bal.

Table 2. Parameters of ESR experiments and numerical simulation.

Parameters	Value
Voltage/V	39
Current/kA	4.8
Electrode diameter/mm	150
Slag weigh/kg	20
Mould diameter/mm	333/260
Ingot height/mm	450
Slag depth/mm	110
Electrode insert depth/mm	10
Conductor height/mm	30

After ESR experiments, the remelted ingots were cut for detailed analysis. First, the sulfur print test was used for marking the profile of the molten steel pool. Then, 10 mm × 10 mm × 10 mm samples were cut in the center of the remelted ingot and the height was 240 mm from the bottom of the remelted ingot. The samples were ground and polished to observe the non-metallic inclusions by automated scanning electron microscope (ASPEX) explorer system. The size of the inclusions was bigger than 1 μm, and the scan area was 30 mm² per sample. The morphology of the non-metallic inclusions was analyzed by scanning electron microscope (JSM-7800F, JEOL. Japan, SEM) and the compositions of the typical particles were analyzed by Energy Dispersive Spectrometer (EDS). Besides, the content of the oxygen and sulfur were detected by TC500C Nitrogen/Oxygen Determinator and Leco CS344 C-S analyzer.

2.2. Physical Simulation Experiments

Metal droplet size has an important influence on remove non-metallic inclusions ability. Hence, it is very meaningful to study the influence of different remelting processes on the metal droplet size. However, ESR is the high temperature, high power and invisible process. To the authors’ best knowledge, there was no effective methods to describe the formation and detachment of the metal droplet. Furthermore, it was extremely dangerous to collect the metal droplets in the laboratory experiment. Previous research indicated that the diameter of the metal droplet for ESR depends on both alloy system and slag composition.²⁷⁾ However, in the present work, both the component of the consumable electrode (GCr15 bearing steel) and slag composition (50%CaF₂-17%CaO-25%Al₂O₃-3%MgO-5%SiO₂) are identical, therefore, the alloy system and slag composition have no effect on the metal droplet size for the different processes. Therefore, in order to investigate the influence of different remelting processes on the metal droplet size, low melting point and visible physical simulation experiment was adopted, and the schematic diagram was shown in Fig. 2.

Fig. 2.

Schematic diagram of the physical simulation. (Online version in color.)

Basing on the similarity principle, a quartz beaker was used as the copper mould with the dimension of 80 mm. To considerate the safety and simplicity of experimental operation, wood alloy, whose melting point was just 348 K (75°C), was used as the consumable electrode, therefore, the alloy can be melted and dripped at a safety temperature. The diameter of the consumable electrode was 40 mm (the value of the consumable electrode/mould of the physical simulation is equivalent with ESR experiment). According to Cao and Dong et al.’s work,^14,16) the conductivity of 0.1221 mol·L⁻¹ NaCl solution is approximately equal to that of the molten slag used in electroslag metallurgy. Besides, it will not change with temperature besides a break at temperature of 357 K. Therefore, 0.1221 mol·L⁻¹ sodium chloride solution was used as the slag. NaCl solution of certain concentration was used as the liquid slag because the conductivity was similar to the premelted slag and the depth of the solution was 40 mm. Molybdenum sheets was used as the baseplate and the conductor, moreover the insert depth of the conductor was 10 mm. There were one digital voltmeter and two digital ammeters in the experiments, wherein, the digital voltmeter was used for detecting the total voltage, one of the digital ammeter (A₁) was used for detecting the total current, and the other (A₂) was used for detecting the current which flows through the conductor.

When the switch ‘S’ was disconnected, the conventional ESR worked. When the switch ‘S’ was connected, it was the ESR-STCCM. Power was supplied after all of the preparations completed, and the total voltage and the total current were 13.0 V and 7.5 A (The values of voltage and current for obtaining an appropriate melting rate were determined by preliminary tests). The total voltage and total current should be kept same to guarantee same power supply. Then the consumable electrode melted and formed the metal droplets, and the metal droplets were collected using a spatula, followed by weighted using an electronic balance with the accuracy of 0.001 g.

2.3. Numerical Simulation

During the ESR process, the remove non-metallic inclusions ability was extremely related to the flow and temperature distribution of the liquid slag pool. However, it is impossible to observe the flow and temperature distribution of liquid slag pool by experiment, therefore, numerical simulation was used for deeply investigating the mechanism of non-metallic inclusion removal for different remelting processes.

In the present work, the electromagnetic field, flow field and temperature field for different remelting processes were investigated by the FLUENT software. Figure 3 shows the geometrical model of the ESR-STCCM and the geometrical model is completely consistent with the laboratory experiments. The shape of the consumable electrodes is obtained from the ESR experiments.

Fig. 3.

Geometrical models of the numerical simulation. (Online version in color.)

The height of the remelted ingot, the depth of the slag pool, the diameter of the consumable electrode and the diameter of the T shape current carrying mould were 450 mm (Z₀≤Z≤Z₁), 110 mm (Z₁≤Z≤Z₅), Φ150 mm (2R_e), Φ333 mm/Φ260 mm (2R_s and 2R_i) and 30 mm (Z₃≤Z≤Z₄), respectively.

2.3.1. Governing Equations and Boundary Conditions

The electromagnetic equations could be described by the Maxwell equation, Ampere law, Electromagnetic induction law and Ohms law. The electromagnetic, heat transport equations and boundary conditions have been reported in the previous researches.^17,18) However, the dimensions and parameters of physical schematic are different. Moreover, the position and the parameter of the conductor are completely different. Besides, the physical property parameters of the steel are the Inconel 718 superalloy, however, GCr15 bearing steel was remelted in the present work. Therefore, the physical property parameters are significantly different. In a word, all of these parameters will result in the different electromagnetic field, flow field and temperature field.

In the present work, the governing equations and the boundary conditions of the flow field were briefly discussed as it has significant influence on the fluid flow of the liquid slag. The fluid flow of the liquid slag and steel could be described by continuity equation and Navier–Stokes equation:   

$$\frac{\partial \rho }{\partial \tau }+\nabla \cdot \overrightarrow{v}=0$$(1)

  

$$\begin{array}{l}\frac{\partial \rho \overrightarrow{v}}{\partial \tau }+\rho \left(\overrightarrow{v}\cdot \nabla \right)\overrightarrow{v}=\\ -\nabla P+\nabla \left({\mu }_{\text{eff}}\cdot \nabla \overrightarrow{v}\right)+\rho \beta \overrightarrow{g}\left(T-T_0\right)+{\overrightarrow{F}}_{\text{loc}}+{\overrightarrow{F}}_{\text{i}}\end{array}$$(2)

where ρ is the density of liquid slag, kg·m⁻³. τ is the time, s. $\overrightarrow{v}$ is the velocity, m·s⁻¹. P is the pressure, Pa. μ_eff is the effective viscosity including molecular viscosity (μ) and turbulent viscosity (μ_t), Pa·s. β is the thermal expansion coefficient. $\overrightarrow{g}$ is the gravitational acceleration, m·s⁻². T is the temperature, K. ${\overrightarrow{F}}_{\text{loc}}$ is the Lorentz force, N. ${\overrightarrow{F}}_{\text{i}}$ is the interfacial tension between the molten steel and the liquid slag (metal droplet and slag interface), N.

Choudhary M, et al.²⁸⁾ found that the Reynolds number of the industrial electroslag furnace was 5900 and there were several vortices in the liquid slag pool, therefore, the realizable κ–ε turbulence mode was appropriate in the present work. The slag-surface was regarded as the free slip condition and no slip condition was applied to other boundaries of the slag pool.

In the present work, VOF model was used for calculating the motion of metal droplets for different remelting processes and the volume fraction of each phases could be descripted by Eqs. (3) and (4).   

$$\frac{\partial \left({\alpha }_{\text{q}}{\rho }_{\text{q}}\right)}{\partial \tau }+\nabla \cdot \left({\alpha }_{\text{q}}{\rho }_{\text{q}}{\overrightarrow{v}}_{\text{q}}\right)=0$$(3)

  

$$\phi =f{\phi }_{\text{m}}+\left(1-f\right){\phi }_{\text{s}}$$(4)

where α_q is the volume fraction of the q^th phase. ρ_q is the volume averaged density of the q^th phase, kg·m⁻³. ${\overrightarrow{v}}_{\text{q}}$ is the velocity vector of the q^th phase, m·s⁻¹. φ is the mixture phase property, φ_m is the metal property and φ_s is the slag property. f is the volume fraction.

The tip of the consumable electrode was modeled as the mass-flow inlet and the mass flow rate was related to the remelting rate which was obtained from the ESR experiments. The bottom of the remelted ingot was modeled as the press-outlet.

2.3.2. Model Solution

In the present work, a 2-D mathematical model was established to describe the complex process (Fig. 3). In order to guarantee the accuracy, the quadrilateral mesh was used with a 1 mm side length. Besides, the Joule heat, current density and electromagnetic force are focused on the corner of the electrode, therefore, the mesh of the slag pool edge were refined. Moreover, in order to ensure the convergence of solution, the typical calculation time step lies between 10⁻³~10⁻⁵ s. First, the governing equations of electric and magnetic fields were established by User Defined Scalar (UDS) to calculate the current density, electromagnetic force and Joule heat. Then, the electromagnetic force was applied as the body force to the momentum equation. Moreover, the Joule heat and solidification latent heat were applied as the source to the energy equation. The boundary conditions of the flow field and temperature field were implemented by User Defined Functions (UDF). Then, the flow field and temperature field were calculated. Besides, the motion of the metal droplets for different remelting processes was calculated via VOF model. The simplified flow diagram of the numerical simulation is shown in Fig. 4.

Fig. 4.

Simplified flow diagram for the numerical simulation.

During the calculating process, flow field and temperature field were not calculated until the normalized unscaled residuals of electric field and magnetic field less than 10⁻⁸. Besides, the iterative procedure continued of continuity was less than 10⁻⁴, and the others should be less than 10⁻⁶ (x-velocity, y-velocity, energy, etc.).

The metal and slag properties, the operating conditions and the geometry parameters of the present work are listed in the Table 3 (The influence of the temperature on the value of surface tension is neglected).

Table 3. Physical property parameter of steel and slag.

Properties	Metal	Slag
Density/(kg·m−3)	7840(s)/6950(l)	2690
Liquidus temperature/K	1730	1571
Solidus temperature/K	1617	—
Specific heat/(J·kg−1·K−1)	680	837
Viscosity/(kg·m−1·s−1)	0.006	0.02
Thermal conductivity(s)/(W·m−1·K−1)	34.13	10.45
Latent heat of fusion/(J·kg−1)	2.4×105	—
Magnetic permittivity/(H·m−1)	1.257×10−6	1.257×10−6
Slag/metal interfacial tension	0.9

3. Results

3.1. ESR Experiments

3.1.1. Content of Oxygen and Sulfur

Figure 5 shows the oxygen content of the consumable electrode and remelted ingot for different remelting processes. It could be clearly seen that the oxygen content is only 16.3×10⁻⁴% in the consumable electrode, and the oxygen content of ESRW and ESR-STCCM are 17.7×10⁻⁴% and 15.8×10⁻⁴%, respectively. Therefore, compared with the consumable electrode, the oxygen content increased by 8.6% for ESRW and decreased 3.1% for ESR-STCCM. Compared with the ESRW, the deoxidation ability for ESR-STCCM increased by 10.7%.

Fig. 5.

Content changes of oxygen and sulfur for different remelting processes. (a) Changes of oxygen content; (b) Changes of sulfur content. (Online version in color.)

Sulfur content of the consumable electrode and remelted ingot for different remelting processes is shown in Fig. 5. It can be observed that the sulfur content in the consumable electrode and the remelted ingot for ESRW and ESR-STCCM are 35.0×10⁻⁴%, 13.9×10⁻⁴% and 10.5×10⁻⁴%, respectively. The desulfurization ability of the ESRW and ESR-STCCM are 60.3% and 70.0%, respectively. Compared with the ESRW, the desulfurization ability of the ESR-STCCM increased by 24.5%. Therefore, the desulfurization ability for ESR-STCCM is better than that of ESRW.

3.1.2. Non-metallic Inclusions

Figure 6 shows the statistical results of the non-metallic inclusions for different remelting processes by the ASPEX explorer system. It can be found from Fig. 6(a) that the number of the non-metallic inclusions of the consumable electrode and remelted ingot for different remelting processes are 719, 991 and 572, respectively. Compared with the consumable electrode, the number of the non-metallic inclusions increased by 27.4% for ESRW, and decreased by 20.4% for ESR-STCCM. Compared with the ESRW, the number of the non-metallic inclusions decreased by 42.3% for ESR-STCCM. That is to say, ESR-STCCM could significantly reduce the number of the non-metallic inclusions.

Fig. 6.

Statistical results of the non-metallic inclusions by ASPEX for different remelting processes: (a) number of the non-metallic inclusions of the consumable electrode and the remelted ingot. (b) proportion of inclusions area. (c) size distribution of inclusions in the electrode and remelted ingots. (Online version in color.)

Figure 6(b) shows the proportion of inclusions area for different remelting processes. It is clearly shown that the total inclusion area of the consumable electrode and the remelted ingot for different remelting processes are 0.022%, 0.018% and 0.015%. Compared with the consumable electrode, the total inclusion area decreased by 18.2% and 31.2%, respectively.

Figure 6(c) shows size distribution of inclusions in the electrode and remelted ingots. It can be clearly found that the large size inclusions in the remelted ingots was dramatically decreased (>10 μm take up 3.2%, 0.81% and 1.04%, respectively), and the small size inclusions in the remelted ingots was markedly increased (1–2 μm take up 30.82%, 51.22% and 56.64%, respectively).

According to the ASPEX explorer system result, the typical inclusion is Al₂O₃, besides, some Al₂O₃–TiN, Al₂O₃–MnS, TiN and MnS inclusions can be found in the remelted ingot. The morphology and the content of the typical non-metallic inclusions in the remelted ingot by SEM are shown in Fig. 7. It is clearly found that the size of the Al₂O₃ inclusion in the consumable electrode is bigger than that in the remelted ingot, and Al₂O₃ is in the core of the complex inclusions (Al₂O₃–TiN and Al₂O₃–MnS).

Fig. 7.

Typical non-metallic inclusions of the remelted ingot: (a) Al₂O₃ of the consumable electrode; (b) Al₂O₃ of the remelted ingot; (c) Al₂O₃–MnS of the remelted ingot; (d) Al₂O₃–TiN of the remelted ingot. (Online version in color.)

3.1.3. Shape of the Consumable Electrode Tip after Experiments

Figure 8 shows the shape of the consumable electrode tip for different remelting processes after ESR experiments. It can be clearly seen that the shape of the consumable electrode tip for ESRW is like “V”, however, it is like a plane for ESR-STCCM. Therefore, the shape of the consumable electrode tip for ESR-STCCM is flatter.

Fig. 8.

Shape of consumable electrode tip for different remelting processes: (a) ESRW; (b) ESR-STCCM. (Online version in color.)

There are two circuits (power → consumable electrode → liquid slag pool → ingot → power and power → conductor → liquid slag pool → ingot → power) for the ESR-STCCM, and the total resistance of the ESR-STCCM is smaller than ESRW (it has the common resistance for the two circuits).¹⁷⁾ Therefore, the insert depth of the consumable electrode for ESRW is no doubt deeper than ESR-STCCM.

3.2. Physical Simulation

Figure 8 shows the consumable electrode and the metal droplet size after physical simulation for different remelting processes. The weights of the metal droplet for different remelting processes are shown in Fig. 9(d). It can be obviously noticed that the weights of the metal droplet for ESRW and ESR-STCCM are 0.636 g and 0.719 g, respectively. Compared with the ESR-STCCM, the weight of the metal droplet for ESRW decreased by 11.5%. Therefore, the metal droplet size for ESRW is smaller than that of ESR-STCCM.

Fig. 9.

Shape of the consumable electrode and the metal droplet size for different remelting processes: (a) shape of the consumable electrode before experiment. (b) metal droplet size of the ESRW. (c) metal droplet size of the ESR-STCCM. (d) weight of the metal droplet for different remelting processes by physical simulation. (Online version in color.)

3.3. Numerical Simulation

The velocity of the slag pool before the metal droplet detachment for different remelting processes is demonstrated in Figs. 10(a) and 10(b). There is only one vortex in the slag pool for ESRW with counterclockwise rotating, besides, the maximum velocity is located at the corner of the consumable electrode with the value of 0.06 m·s⁻¹. However, there are two vortices for ESR-STCCM, and the big one is near the conductor with clockwise rotating, besides, the maximum velocity is 0.05 m·s⁻¹, moreover, the other one below the consumable electrode.

Fig. 10.

Velocity and electromagnetic force distribution of the slag pool for different remelting processes: (a) velocity-ESRW; (b) velocity-ESR-STCCM; (c) electromagnetic force ESRW; (d) electromagnetic force- ESR-STCCM. (Online version in color.)

The difference of the velocity distribution is attributed to the difference of the electromagnetic force. Figures 10(c) and 10(d) shows the electromagnetic force distribution in the slag pool for different remelting processes. It can be clearly seen that the maximum electromagnetic force is at the corner of the consumable electrode with the value of 2420 N·m⁻³ and tilted downward for ESRW. Besides, the minimum electromagnetic force is located at the center of the slag pool and the upside of the mould. However, the maximum electromagnetic force for ESR-STCCM is located at the bottom of the conductor with the value of 1620 N·m⁻³. Moreover, the direction of the electromagnetic force for ESR-STCCM is tilted upward.

It is well known that the metal droplet would has a necking phenomenon at the tip of the consumable electrode because of the slag/metal interfacial tension, followed by dripping and passing through the slag pool. The area of the metal droplets was calculated by the Image Pro Plus (IPP 6.0) software. Table 4 shows the average areas of the metal droplet for the different remelting processes. According to the statistics results, the average areas of the metal droplet are 67.31 mm² and 77.74 mm², respectively. Consequently, the metal droplet size for ESRW is the smaller than that of ESR-STCCM. Therefore, the metal droplet size results of the physical simulation are completely consistent with the numerical simulation.

Table 4. Droplet size of the numerical simulation for different remelting processes.

Remelting process	Area/mm
ESRW	67.3±5.7
ESR-STCCM	77.7±5.2

4. Discussions

4.1. Influence of Different Remelting Processes on Metal Droplet Size

During the ESR process, consumable electrode was melted by the Joule heat, and the side of the consumable electrode melted firstly because of the skin effect. Then the molten metal flowed from the side of the consumable electrode to the center. Meanwhile, the molten metal was subjected to the gravity (G), interfacial tension (F_i), electromagnetic force (F_loc), thermal buoyancy (F_b) and other forces. Figure 11(b) shows the force analysis during the metal droplet formation process. Besides, the metal droplet was formed and dripped from the consumable electrode if the resultant force was downward. Consequently, the metal droplet passed through the liquid slag pool and formatted the molten metal pool.

Fig. 11.

Schematic diagram of the remove non-metallic inclusions for ESR process: (a) Total process of the remove non-metallic inclusions. (b) Motion of the non-metallic inclusions and the force analysis during the metal droplet formation process. (c) Float process of the non-metallic inclusion. (Online version in color.)

In the present work, the material of the consumable electrode and the slag composition for different remelting processes are the same, therefore, the interfacial tension for the different remelting processes is equivalent. Besides, the electromagnetic force for ESRW is higher than that of the ESR-STCCM, i.e. F_loc-ESRW>F_{loc-ESR-STCCM} (Fig. 10) Therefore, the gravity has the opposite tendency for different remelting processes if the resultant force is equivalent. That is to say, the weight of the metal droplet before dripping for different remelting processes is G_ESRW<G_ESR-STCCM, i.e. the metal droplet size for ESR-STCCM is the bigger than ESRW.

4.2. Influence of Different Remelting Processes on Cleanliness

ESR have the excellent ability to remove the non-metallic inclusions. Figure 11(a) shows the schematic diagram of remove non-metallic inclusions for ESR.

It is well known that the oxygen is mainly present in the form of inclusions in steel, therefore, the process of deoxidation is the process of remove the non-metallic inclusions. The inclusions in the remelted ingot mainly came from two parts - the non-metallic inclusions of the consumable electrode [(M_xO_y)_Original] and the oxidation of the alloy elements (Fe, Al, Mn and Si in the present work) [(M_xO_y)_Oxide].²⁹⁾ According to the previous research,^30,31) the type of the non-metallic inclusion in the remelted ingot is greatly influenced by the consumable electrodes, besides, the ability for ESR process to remove Al₂O₃ inclusions is particularly weak. In the present work, the consumable electrodes were manufactured by the continuous casting (CC) in Benxi Iron & Steel Corporation, and Al was used for reducing the oxygen content during the converter smelting and refining processes, therefore, the main inclusion of the consumable electrodes and the remelted ingot is Al₂O₃. Besides, part of the Fe, Al, Mn and Si were oxidized because of the high temperature of the consumable electrode and air interface (electrode-air interface).³²⁾ Figure 12 shows the temperature distribution of the electrode-air interface for different remelting processes by numerical simulation. It is clearly indicated that the temperature of the electrode-air interface for ESRW is slight higher than ESR-STCCM. Therefore, of element oxidation for ESRW is higher than ESR-STCCM. Therefore, the inclusions of (M_xO_y)_Oxide for ESRW is more than ESR-STCCM.

Fig. 12.

Temperature distribution of the electrode-air interface for different remelting processes. (Online version in color.)

As is mentioned above, the removing non-metallic inclusions process for ESR can be divided into three steps. And the first step (Fig. 11(b)), metal droplet formation process, is the most primary step to remove the non-metallic inclusion. Equation 5 is directly illustrated the relationship between the remelting rate and the content of the non-metallic inclusions in the remelted ingot. Where, C₀ is the initial content of the non-metallic inclusions in the consumable electrode, C is the content of the non-metallic inclusions in the remelted ingot, v_r is the remelting rate, r is the diameter of the inclusion, δ is the width of the liquid metal. Therefore, the metal droplet formation rate, i.e. the remelting rate, plays a decisive role in removing the non-metallic inclusion.³³⁾ In the present work, the remelting voltage and total current of the different remelting processes are 39 V and 4.8 kA, however, the current which flows through consumable electrode for ESRW is higher than ESR-STCCM (part of the current flows into the liquid slag from the conductor). The remelting rate of the ESRW (160 kg·h⁻¹) is higher than ESR-STCCM (140 kg·h⁻¹). Therefore, at the tip of the consumable electrode, the remove non-metallic inclusions ability for ESR-STCCM is higher than ESRW.   

$$\overline{C}=\frac{C_0{v}_r}{2D}{\left(1-\frac{r}{\delta }\right)}^2$$(5)

The second step ((b) in the Fig. 11(a)), drip process, is the secondary way. According to the previous knowledge,³⁴⁾ the specific surface area is higher for a smaller metal droplet. Therefore, the liquid slag will have a better ability to remove non-metallic inclusion for a smaller metal droplet. From the physical simulation and numerical simulation results, the metal droplet size for ESRW is the smaller than ESR-STCCM. Therefore, the remove non-metallic inclusions ability in dripping process for ESRW is higher than that of ESR-STCCM.

Because the total resistance of the ESR-STCCM is smaller than ESRW. Hence, the temperature of the slag pool for ESR-STCCM is higher than ESRW in the case of the same remelting power.¹⁸⁾ As is well known that high temperature is beneficial to remove non-metallic inclusion during the remelting process, therefore, the remove non-metallic inclusions ability for ESRW is higher than that of ESR-STCCM during the dripping process.

The third step (Fig. 11(c)), float process, is the weakest step.²⁾ Kelkar et al.’s³⁵⁾ work indicated that a slice of the inclusions passes into the molten metal pool, and part of the inclusions can float to the slag-metal interface and absorb by the liquid slag. According to the previous study,³⁶⁾ the shape of the molten metal pool has a directly effect on the removal of inclusions. Figure 13 is the result of the sulfur print experiment for different remelting processes, it is clearly demonstrated that the depths of the molten metal pool for the different processes are 152 mm and 85 mm, and the shape of the molten metal pool for ESR-STCCM is flatter and shallower (The effect of the remelting processes on the shape of the molten metal pool had been deeply discussed in our previous study¹⁹⁾). Therefore, ESR-STCCM is beneficial to the float process of the inclusions.

Fig. 13.

Shape of the molten metal pool for different remelting processes. (a) ESRW, (b) ESR-STCCM.

Compared with the ESRW and ESR-STCCM, the average metal droplets size of the ESR-STCCM is larger, which is harmful to remove the non-metallic inclusion. Nevertheless, the remelting rate of the ESR-STCCM is lower, which provide sufficient time for the reaction of the metal droplet and liquid slag during the formation step, besides, the shape of the molten metal pool for ESR-STCCM is flatter and shallower, moreover, the element oxidation of the consumable electrode for ESR-STCCM is lightened. Therefore, the oxygen content and the number of the non-metallic inclusions for ESRW is higher than ESR-STCCM, (Figs. 5 and 6).

ESR process provides a favorable desulfurization kinetic conditions because the metal droplets formed and dripped from the tip of the consumable electrode, followed by passing through the liquid slag pool, therefore, it can increase the contact area and reaction time between the slag and metal droplets. Besides, the liquid slag is subject to the intensively electromagnetic force and Joule heat, which will make liquid slag possess good fluidity and sufficient heat. All in all, ESR has a better desulfurization ability.

During the ESR process, there are two patterns to desulfurize (slag-metal reaction and gasifying reaction), and the reactions are indicated in Eqs. (6) and (7).   

$$[\mathrm{S}]+({\mathrm{O}}^{2-})=({\mathrm{S}}^{2-})+[\mathrm{O}]$$(6)

  

$$({\mathrm{S}}^{2-})+3/2\left\{{\mathrm{O}}_2\right\}=\left\{\mathrm{S}{\mathrm{O}}_2\right\}+({\mathrm{O}}^{2-})$$(7)

As is motioned above, ESR-STCCM has a better condition to desulfurize by slag-metal reaction. Nevertheless, the gasifying desulfurization plays a more dominant role in desulfurization.

As is well known, high temperature is beneficial to desulfurize, and the temperature distribution of the slag-surface for different remelting processes are shown in Fig. 14. It can be seen that the temperature of the slag-surface for ESR-STCCM is higher than ESRW, especially near the conductor. Besides, the numerical simulation indicates that the velocity of the slag pool where is near the conductor for ESR-STCCM is higher than ESRW (Fig. 10), therefore, it has a better dynamics condition to desulfurize. Taking all factors into consideration, ESR-STCCM has better desulfurization ability than ESRW, which will result a lower sulfur content in the remelted ingot. (Fig. 5)

Fig. 14.

Temperature of the slag-surface for different remelting processes. (Online version in color.)

5. Conclusions

The metal droplets size and cleanliness of the remelted ingot for ESRW, ESR-STCCM were investigated via the laboratory experiments, physical simulations and numerical simulations. The conclusions are as follows:

(1) ESR-STCCM can significantly change the electromagnetic force distribution, which can increase the metal droplet size in the case of the same remelting power.

(2) ESR-STCCM can significantly reduce the number and area proportion of the non-metallic inclusions. Nevertheless, the types of the non-metallic inclusions are not changed.

(3) ESR-STCCM can improve the cleanliness of the remelted ingot. The oxygen content and the sulfur content decreased for ESR-STCCM decreased by 10.7% and 24.5%, respectively.

Acknowledgement

This work was supported by National Science Foundation of China (No. U1560203, No. 51674070 and No. 51874084), Also, this research was supported by National key research and development program (2017YFB0305201).

References

• 1)   J.  Fu: Acta Metall. Sin., 15 (1979), 526.
• 2)   Z. B.  Li: Theory and Practice of Electroslag Metallurgy, Metallurgical Industry Press, Beijing, (2010), 31 (in Chinese).
• 3)   Z. H.  Jiang,  Y. W.  Dong,  F. B.  Liu and  X.  Geng: Electroslag Metallurgy, Science Press, Beijing, (2015), 387 (in Chinese).
• 4)   Z. B.  Li,  W. H.  Zhou and  Y. D.  Li: Iron Steel, 15 (1980), 20 (in Chinese).
• 5)   K.  Yamamoto,  H.  Yamamura and  Y.  Suwa: ISIJ Int., 51 (2011), 1987.
• 6)   Y.  Liu,  Z.  Zhang,  G. Q.  Li,  Q.  Wang,  L.  Wang and  B. K.  Li: Steel Res. Int., 88 (2017), 1700058.
• 7)   Z. H.  Jiang: Physical Chemistry and Transport Phenomena for Electroslag Metallurgy, Northeastern University Press, Shenyang, (2000), 143 (in Chinese).
• 8)   Y. W.  Dong,  H. Y.  Chen,  X.  Li and  Z. X.  Wu: J. Northeast. Univ. (Nat. Sci.), 31 (2016), 1564 (in Chinese).
• 9)   A.  Kharicha,  A.  Ludwig and  M.  Wu: Proc. 2011 Int. Symp. on Liquid Metal Processing and Casting, TMS, Nancy, (2011), 73.
• 10)   Y.  Kojima,  M.  Kato,  T.  Toyoda and  M.  Inouye: Proc. 4th Int. Symp. on Electroslag Remelting Processes, The Iron and Steel Institute of Japan, Tokyo, (1973), 35.
• 11)   L. B.  Medovar,  A. K.  Tsykulenko,  V. Y.  Saenko,  A. V.  Chernets,  B. B.  Fedorovskii,  V. I.  Us and  I. A.  Lantsman: Proc. Medovar Memorial Symp., Elmet-Roll Medovar Group Company, Kyiv, (2001), 49.
• 12)   F. B.  Liu,  Z. H.  Jiang,  H. B.  Li,  X.  Geng,  X.  Chen,  H.  Feng and  X. M.  Zang: Ironmaking Steelmaking, 41 (2014), 791.
• 13)   Y. W.  Dong,  L. C.  Zheng and  Z. H.  Jiang: Ironmaking Steelmaking, 40 (2013), 153.
• 14)   Y. L.  Cao,  Z. H.  Jiang,  Y. W.  Dong,  X.  Deng,  L.  Medovar and  G.  Stovpchenko: ISIJ Int., 58 (2018), 1052.
• 15)   Z. W.  Hou,  Y. W.  Dong,  Z. H.  Jiang,  X.  Deng,  Y. L.  Cao,  H. B.  Cao,  L.  Medovar and  G.  Stovpchenko: Metals, 8 (2018), 504.
• 16)   Y. W.  Dong,  Z. H.  Jiang,  H. B.  Cao,  Z. W.  Hou and  K. A.  Yao: Metall. Mater. Trans. B, 47 (2016), 3575.
• 17)   Y. W.  Dong,  Z. H.  Jiang,  H. B.  Cao,  X.  Wang,  Y. L.  Cao and  D.  Hou: ISIJ Int., 56 (2016), 1386.
• 18)   Y. W.  Dong,  Z. W.  Hou,  Z. H.  Jiang,  H. B.  Cao,  Q. L.  Feng and  Y. L.  Cao: Metall. Mater. Trans. B, 49 (2018), 349.
• 19)   H. B.  Cao,  Z. H.  Jiang,  Y. W.  Dong,  F. B.  Liu,  Y. L.  Cao,  Z. W.  Hou,  K. A.  Yao and  M. X.  Xu: Steel Res. Int., 90 (2019), 1800337.
• 20)   Y. L.  Cao,  Y. W.  Dong,  Z. H.  Jiang,  H. B.  Cao,  D.  Hou and  Q. L.  Feng: Int. J. Min. Met. Mater., 48 (2017), 655.
• 21)   H.  Wang,  Y. B.  Zhong,  Q.  Li,  Y. P.  Fang,  W. L.  Ren,  Z. S.  Lei and  Z. M.  Ren: Metall. Mater. Trans. B, 48 (2017), 655.
• 22)   Q.  Wang,  F.  Wang,  G. Q.  Li,  Y. M.  Gao and  B. K.  Li: Int. J. Heat Mass Transf., 113 (2017), 1021.
• 23)   Q.  Li,  Y. B.  Zhong,  C. X.  Sun,  H.  Wang,  T. X.  Zheng,  W. L.  Ren and  Z. M.  Ren: Acta Metall. Sin., 31 (2018), 1311.
• 24)   H.  Wang,  J.  Li,  C. B.  Shi,  Y. F.  Qi and  Y. X.  Dai: ISIJ Int., 59 (2019), 828.
• 25)   S. J.  Li,  G. G.  Cheng,  Z. Q.  Miao,  W. X.  Dai,  L.  Chen and  Z. Q.  Liu: ISIJ Int., 58 (2018), 1781.
• 26)   H.  Dong,  F. B.  Liu,  T. P.  Qu,  Z. H.  Jiang,  D. Y.  Wang and  Y. W.  Dong: ISIJ Int., 58 (2018), 876.
• 27)   J.  Campbell: JOM, 22 (1970), 23.
• 28)   M.  Choudhary and  J.  Szekely: Ironmaking Steelmaking, 8 (1981), 225.
• 29)   Z. H.  Jiang,  Y. W.  Dong,  X.  Geng and  F. B.  Liu: Electroslag Metallurgy, Science Press, Beijing, (2016), 183 (in Chinese).
• 30)   Z. B.  Li,  W. H.  Zhou and  Q. H.  Wang: Iron Steel, 18 (1983), 13 (in Chinese).
• 31)   X.  Geng,  Z. H.  Jiang,  F. B.  Liu and  L. K.  Liang: Iron Steel, 44 (2009), 42 (in Chinese).
• 32)   X. C.  Huang,  B. K.  Li and  Z. Q.  Liu: Metall. Mater. Trans. B, 49 (2018), 709.
• 33)   Q.  Wang,  R. T.  Wang,  Z.  He,  G. Q.  Li,  B. K.  Li and  H. B.  Li: Int. J. Heat Mass Transf., 125 (2018), 1333.
• 34)   H.  Wang,  Y. B.  Zhong,  Q.  Li,  W. Q.  Li,  W. L.  Ren,  Z. S.  Lei,  Z. M.  Ren and  Q.  He: ISIJ Int., 57 (2017), 2157.
• 35)   K. M.  Kelka,  S. V.  Patankar,  S. K.  Srivatsa,  R. S.  Minisandram,  D. G.  Evans,  J. J.  deBarbadillo,  R. H.  Smith,  R. C.  Helmink,  A.  Mitchell and  H. A.  Sizek: Proc. 2013 Int. Symp. on Liquid Metal Processing and Casting, TMS, Texas, (2013), 3.
• 36)   Z. B.  Li: Theory and Practice of Electroslag Metallurgy, Metallurgical Industry Press, Beijing, (2010), 31 (in Chinese).