Application of Thermodynamic Calculations in the Study of Slag Melting Characteristics and Aluminum Loss Control

Abstract

According to the production process requirements of oriented silicon steel in a certain steel mill, optimization of the slag composition ratio is studied through thermodynamic calculations. The CaO-SiO₂-Al₂O₃-FeO-MgO slag system is studied using FactSage thermodynamic software (FactSage 8.1), and a slag optimization plan is proposed based on industrial experiments involving changes in the composition ratio of the slag, calculation and analysis of the melting characteristics of RH refining slag, further verification through orthogonal experiments, and observations of the slag state, temperature, and composition relationship through phase diagrams. This study provides theoretical guidance for finding a suitable slag composition ratio based on the influence of slag on dissolved aluminum in steel liquid. Research has shown that, combined with thermodynamic analysis, slag melting characteristics, component content calculations, and industrial experiments, the range of RH refining slag composition suitable for production in this steel mill is slag in the range of 1.3~1.5 alkalinity, 25~30% Al₂O₃, 5~6% MgO, and 1–2% FeO.

1. Introduction

Oriented silicon steel is an Fe-Si soft magnetic material containing approximately 3% (2.9% to 3.5%) Si, with a (110)<001> Gaussian texture and coarse secondary recrystallization grains. Oriented silicon steel has the characteristics of high magnetic induction and low iron loss. Its manufacturing process is complex. To obtain high-quality oriented silicon steel products, it is necessary to improve the magnetic properties of silicon steel products. The key to obtaining excellent stable magnetic properties is to control the content of inhibitor-forming elements in the steel [1,2]. The dissolved aluminum content in oriented silicon steel is a key element in the formation of inhibitors, and the main factors affecting the dissolved aluminum content in molten steel at the RH refining endpoint include the temperature of the molten steel, the dissolved oxygen content in the molten steel, the FeO content in the slag, the net circulation time of the molten steel, and the amount of aluminum added to the ladle [3,4,5].

Under fixed conditions for the net circulation time of molten steel and the amount of aluminum added to the ladle, the dissolved aluminum content of molten steel is affected by changing the temperature of the molten steel and the FeO content of the slag. Su Duxing et al. [6], focusing on the smelting of low-carbon aluminum killed steel, investigated the effects of slag composition, slag oxidizability, and steel cleanliness on the aluminum loss during the RH refining process, thereby enhancing the production stability and product quality in practical operations. According to the actual production composition, the CaO-SiO₂-Al₂O₃-FeO-MgO slag system was selected as the research object to explore the influence of different composition ratios of slag on the slag melting characteristics. The FeO content in the slag was selected as the starting point to study its impact on aluminum loss in the RH process. Finally, various factors were considered to achieve optimization of the slag in the RH process of oriented silicon steel [7,8,9].

2. Methods

Slag research is often constrained by experimental limitations, while operational steel plants are generally unable to frequently modify stable production lines for small-scale testing. Therefore, thermodynamic calculations using FactSage software (version 8.1) are widely employed to guide practical production owing to their advantages in efficiency, cost-effectiveness, and environmental benefits. In this study, the Phase Diagram, Equilibrium, and Viscosity calculation modules in the FactSage 8.1 calculation software are used to calculate the phase equilibrium and physicochemical properties of the CaO-SiO₂-Al₂O₃-FeO-MgO five component slag system. Calculations are conducted using the FToxyd oxide database, FactPS pure substance database, and FTmisc comprehensive database at standard atmospheric pressure [10,11]. Because the slag system used contains the multivalent metal oxide FeO, O₂ is added to the system during the calculation in order to define the oxygen partial pressure and determine the two iron ion contents [12].

The FactSage software uses the method of minimizing the Gibbs free energy of the entire system to calculate the state of a given element or compound system when it reaches chemical equilibrium under certain conditions [13]. The thermodynamic properties and phase equilibrium are calculated using the Gibbs free energy minimization algorithm, which means that in an isothermal and isobaric multiphase system, the equilibrium condition is that the Gibbs free energy (G) of the system is minimized. The Gibbs free energy is expressed as Equation (1).

$$\mathrm{G}={\sum }_{\mathrm{i}}{\mathrm{n}}_{\mathrm{i}}{\mathsf{\mu }}_{\mathrm{i}}={\sum }_{\mathrm{i}}{\mathrm{n}}_{\mathrm{j}}({\mathsf{\mu }}_{\mathrm{i}}^0+\mathrm{R}\mathrm{T}\mathrm{l}\mathrm{n}{\partial }_{\mathrm{i}}),$$

(1)

where i is the component in the system; j is the element in the component i; n_i is the amount of the component i; n_j is the amount of the element j; μ_i is the chemical potential for the component i; ${\mathsf{\mu }}_{\mathrm{i}}^0$ is the standard chemical potential of the component i; ∂_i is the number of atoms of the component i at temperature T; R is the ideal gas constant.

The quality constraint conditions are expressed as Equation (2).

$${\sum }_{\mathrm{i}}{\partial }_{\mathrm{i}\mathrm{j}}{\mathrm{n}}_{\mathrm{i}}={\mathrm{b}}_{\mathrm{j}},$$

(2)

where ∂_ij is the number of atoms of the element j in the component i, and b_j is the total amount of the element j.

According to the multiple measurements of the steel mill, the average content of the slag composition is shown in

Table 1. Slag composition (ω%).

Al2O3	FeO	MgO	CaO	SiO2	S	TiO2	P2O5	R
32	1	5.5	36	24	≤0.03	≤0.4	≤0.05	1.5

.

3. Results and Discussion

3.1. The Influence of Slag Composition on the Melting Characteristics of Slag

The melting characteristics of slag mainly include its melting point and viscosity. These are influenced by the slag’s oxidation potential and alkalinity and together form its physicochemical properties, which vary with changes in slag composition. As a fundamental condition for steelmaking, these physicochemical properties must be within a suitable range.

The oxidizing power of the slag, which reflects its capacity to transfer oxygen, significantly influences several key aspects of the steelmaking process. These include the slag formation rate, dephosphorization, desulfurization, decarburization, splash behavior, metal recovery yield, oxygen content in the endpoint steel liquid, and the erosion rate of the furnace lining. The FeO content in the slag serves as a direct indicator of this oxidizing power.

The removal of inclusions in slag is influenced by the alkalinity of the slag, which also affects the activity of the components in the slag. In this study, the alkalinity of the slag is calculated based on the binary alkalinity R2 = ω (CaO)/ω (SiO₂). The slag locks a fixed melting point, which is replaced by the melting temperature in this paper. The melting temperature of the slag primarily depends on the composition of the slag. The greater the content of the high-melting-point substances, the higher the slag melting temperature [14,15,16].

The dissolved oxygen content in molten steel is closely related to the oxidizability and fluidity of the slag. The viscosity and fluidity of the slag vary inversely, and the fluidity of the slag is related to the proportion of refractory substances in the slag. The fluidity of the slag directly affects the smooth operation of the blast furnace and the quality of pig iron. High-viscosity slag can lead to operational difficulties in smelting or affect the quality of finished products [17].

Figure 1 shows the equilibrium phase diagram of CaO-SiO₂-Al₂O₃-0.01FeO, where multiple components precipitate in each equilibrium region, directly determining the physicochemical properties of the slag. The liquidus shows an extensive region in the low melting point range, and a gap exists in the high melting point range of CaO and Al₂O₃. If these high-melting-point solid compounds precipitate in large quantities in the slag, it will inevitably lead to a decrease in the fluidity of the slag, and viscous slag will cause slogging phenomenon and losses. The formation of various complex compounds in the low melting point range effectively reduces the slag melting temperature [18].

To study the influence of individual components on the melting characteristics of slag, the Equilibrium and Viscosity calculation modules in FactSage thermodynamic software are used to simulate and calculate the effects of each component on the melting temperature and the viscosity of slag. The melting temperature is reported in °C, and the viscosity is measured in Pa·s.

3.1.1. Effect of Al₂O₃ on the Melting Temperature and Viscosity of Slag

Figure 2 shows the changes in the melting temperature and viscosity of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system with increasing Al₂O₃ content at different alkalinity levels. As shown in Figure 2a, the melting temperature of the slag increases with the increase in the Al₂O₃ mass fraction, and the growth rate gradually slows. When the mass fraction of Al₂O₃ is 25% to 30%, the melting temperature of the slag increases more rapidly, while when the mass fraction of Al₂O₃ is 30% to 38%, the melting temperature of the slag increases more slowly. When the alkalinity of the slag increases from 1.2 to 1.6, the overall melting temperature of the slag decreases, and the rate of increase in the slag melting temperature also decreases. As shown in Figure 2b, the viscosity of the slag consistently increases with the increase in the Al₂O₃ mass fraction. Al₂O₃ primarily exists as [AlO₄]⁵⁻ tetrahedra in the slag, which significantly enhances the degree of polymerization of the slag structure, thereby leading to a uniform increase in the viscosity of the slag. When the alkalinity of the slag increases from 1.2 to 1.6, the overall viscosity of the slag decreases, and the rate of increase in the slag viscosity also decreases [19].

3.1.2. Effect of MgO on the Melting Temperature and Viscosity of Slag

Figure 3 shows the changes in the melting temperature and viscosity of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system with increasing MgO content at different alkalinity levels. As shown in Figure 3a, when the alkalinity is 1.2, the melting temperature of the slag increases with the increase in the MgO mass fraction. When the alkalinity is 1.6, the melting temperature of the slag decreases first and then increases with the increase in the MgO mass fraction. The lowest value is achieved at a mass fraction of 5.8%, and the growth rate of the slag melting temperature slows down when the MgO mass fraction exceeds 5.8%. As shown in Figure 3b, the viscosity of the slag decreases with the increase in the MgO mass fraction, and the rate of viscosity decrease gradually slows down. When the alkalinity of the slag increases from 1.2 to 1.6, the overall viscosity of the slag decreases, and the rate of decrease in slag viscosity also decreases. With the increase in the mass fraction of MgO in the slag, the amount of oxygen ions supplied to the slag increases, which enhances the activity of oxygen ions in the slag. This causes the complex silicate network structure in the slag to depolymerize into simple units, thereby reducing the viscosity of the slag.

3.1.3. Effect of Alkalinity on the Melting Temperature and Viscosity of Slag

Figure 4 shows the changes in the melting temperature and viscosity of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system at different alkalinity levels. The alkalinity of the second group of slag follows the first group, but the content of other components is reduced. As shown in Figure 4a, the melting temperature of the slag first increases and then decreases with increasing alkalinity. The increasing trend of the melting temperature is obvious, while the decreasing trend is slow. The first group achieved its maximum value at the alkalinity of 1.6, while the second group achieved its maximum value at the alkalinity of 1.8. The overall melting temperature of the first group is higher than that of the second group. As shown in Figure 4b, the viscosity of the slag decreases with the increasing alkalinity, and the downward trend gradually slows down. The overall viscosity of the second group is higher than that of the first group. Within a certain range, as the basicity of the slag increases, the content of free oxygen ions in the slag rises. These free oxygen ions interact with bridging oxygen, disrupting the network structure of the slag and thereby reducing the viscosity of the slag.

3.1.4. Effect of FeO on the Melting Temperature and Viscosity of Slag

Figure 5 shows the changes in the melting temperature and viscosity of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system with increasing FeO content at different alkalinity levels. As shown in Figure 5a, when the alkalinity is 1.2, the melting temperature of the slag first increases and then decreases with the increase in the FeO mass fraction, reaching its maximum value at a mass fraction of 5%. When the alkalinity is 1.6, the melting temperature of the slag decreases with the increase in the FeO mass fraction, and the rate of decrease accelerates. As the FeO mass fraction increases, the slag melting temperature fluctuates moderately. As shown in Figure 5b, the viscosity of the slag decreases with the increase in the FeO mass fraction, and the deceleration gradually increases. When the alkalinity of slag increases from 1.2 to 1.6, the overall viscosity of slag decreases, and the rate of decrease in slag viscosity also decreases.

3.1.5. Orthogonal Experimental Study on the Melting Characteristics of Slag

An orthogonal experiment is a design method that uses statistical principles to study multi-factor and multi-level problems. Based on orthogonality, representative points are selected from a comprehensive experiment for testing; these points are uniformly dispersed and comparable. The approach can obtain sufficient and meaningful data by minimizing the number of experiments, making it an efficient, fast, and economical experimental design method [20]. In this study, the orthogonal experimental method is used to further compare the degree of influence of each component on the melting temperature and viscosity of the slag. The design of the slag composition is shown in

Table 2. Design of slag composition (ω%).

ω(Al2O3)	ω(FeO)	ω(MgO)	Alkalinity
26	0.5	4	1
30	1	5	1.2
34	1.5	6	1.6
38	2	7	2

.

Using FactSage software to calculate the effects of various components on the melting temperature and viscosity of slag, an orthogonal experimental table L16(44) was created. A total of 16 experiments were conducted, and the results obtained are shown in

Table 3. Results of orthogonal experiments.

Group	ω(Al2O3)	ω(FeO)	ω(MgO)	Alkalinity	Melt Temperature	1600 °C Viscosity	1500 °C Viscosity
1	26	0.5	4	1	1434.1	3.564	6.916
2	30	0.5	5	1.2	1498	2.864	5.437
3	34	0.5	6	1.6	1534.6	2.12	3.884
4	38	0.5	7	2	1569.5	1.862	3.364
5	26	1	5	1.6	1513.9	1.735	3.093
6	30	1	7	2	1546.4	1.49	2.614
7	34	1	4	1	1432.5	4.354	8.803
8	38	1	6	1.2	1529.4	3.192	6.19
9	26	1.5	6	2	1510.1	1.352	2.349
10	30	1.5	7	1.6	1547.3	1.737	3.089
11	34	1.5	5	1.2	1489	3.014	5.785
12	38	1.5	4	1	1459.4	4.647	9.547
13	26	2	7	1.2	1506.9	2.006	3.725
14	30	2	6	1	1457.3	3.099	5.901
15	34	2	5	2	1509.7	1.718	3.076
16	38	2	4	1.6	1529.4	2.434	4.576

. The melting temperature is measured in °C, and the viscosity is measured in Pa·s.

Based on the analysis of the orthogonal experimental results using Minitab21 software, the mean response table of each component for the melting temperature is shown in

Table 4. Mean response of each component to melting temperature.

Factor	Al2O3	FeO	MgO	Alkalinity
Mean 1	1491	1509	1500	1446
Mean 2	1512	1506	1495	1506
Mean 3	1491	1501	1508	1531
Mean 4	1522	1501	1514	1534
Range (R)	31	8	19	88
Rank	2	4	3	1

. The range (R) can reflect the influence of each factor on the response, and the magnitude of the rank is the degree of influence [21]. The range in the table is ranked in descending order: R (Alkalinity) > R (Al₂O₃) > R (MgO) > R (FeO). The melting temperature of slag is most significantly affected by alkalinity, followed by Al₂O₃ content and finally FeO content. When the melting temperature of the slag is the lowest, the component levels are: 34% Al₂O₃-1% FeO-4% MgO-1 alkalinity. At the highest melting temperature of the slag, the values of each component are: 38% Al₂O₃-0.5% FeO-7% MgO-2 alkalinity.

The mean response graph of each component to the melting temperature is shown in Figure 6, which more intuitively shows the influence of each component on the slag melting temperature and facilitates the comparison of the degree of influence of each component for the slag melting temperature. It can be clearly seen from Figure 6 that the melting temperature is most affected by alkalinity, and the melting temperature continues to increase as the alkalinity increases. The melting temperature is least affected by FeO and continues to decrease as the FeO content increases. As the Al₂O₃ content increases, the melting temperature first increases, then decreases, and then increases again, showing significant fluctuation. As the MgO content increases, the melting temperature first decreases and then increases. Comparing the results calculated in Section 3.1.1, Section 3.1.2, Section 3.1.3 and Section 3.1.4, except for the different trend of the influence of Al₂O₃, the influence trend of each component on the melting temperature of the slag is relatively consistent.

Continuing to analyze the results of the orthogonal experiment, the mean response table of each component for viscosity is shown in

Table 5. Mean response of each component to viscosity.

Factor	1600 °C Viscosity				1500 °C Viscosity
	Al2O3	FeO	MgO	Alkalinity	Al2O3	FeO	MgO	Alkalinity
Mean 1	2.164	2.603	2.625	3.916	4.021	4.9	4.973	7.792
Mean 2	2.298	2.693	2.741	2.769	4.26	5.175	5.288	5.284
Mean 3	2.801	2.688	2.441	2.006	5.387	5.193	4.581	3.66
Mean 4	3.034	2.314	2.49	1.606	5.919	4.319	4.745	2.851
Range (R)	0.87	0.378	0.3	2.31	1.898	0.873	0.707	4.941
Rank	2	2	4	1	2	3	4	1

. Two temperatures, 1500 °C and 1600 °C, were set here for comparative purposes. The range comparison in the table is as follows: R (Alkalinity) > R (Al₂O₃) > R (FeO) > R (MgO). The viscosity of slag is most significantly affected by the alkalinity, followed by the Al₂O₃ content and finally the MgO content. The compositions corresponding to the minimum and maximum viscosity are consistent at both temperatures. When the slag viscosity is the lowest, the optimal component is as follows: 26% Al₂O₃-1.5% FeO-6% MgO-2 alkalinity. When the viscosity of the slag is highest, the values of each component are as follows: 38% Al₂O₃-1.5% FeO-4% MgO-1 alkalinity.

The mean response graph of each component on the viscosity is shown in Figure 7, which can more intuitively show the influence of each component on slag viscosity and allows for the comparison of the degree of influence of each component on the viscosity. It can be clearly seen from the figure that with the increase in each component, the viscosity change trend at the two temperatures is highly consistent, differing only in the scale of the vertical axis. The slag is completely melted at 1600 °C, resulting in a further decrease in viscosity. The viscosity of the slag at 1600 °C is about half of that at 1500 °C.

The viscosity is most significantly influenced by the alkalinity, and the viscosity continues to decrease as the alkalinity increases and exhibits a wide fluctuation range, while the viscosity is least influenced by MgO, and the viscosity fluctuates first and then decreases and then increases as the MgO content increases, showing minimal variation. The viscosity increases with the increase in Al₂O₃; As the FeO content increases, the viscosity first decreases and then increases. Comparing the results calculated in Section 3.1.1, Section 3.1.2, Section 3.1.3 and Section 3.1.4, except for the divergent trend of the influence of MgO, the influence trend of each component on slag viscosity is relatively consistent.

3.2. The Influence of the Slag Composition on the Liquid Phase Zone

The phase diagram can be used to better determine the state, temperature, and composition relationship of the material phase. The Phase Diagram calculation module of FactSage thermodynamic software was used to draw the phase diagram of the CaO-SiO₂-Al₂O₃-xMgO-yFeO isotherms. The liquid phase zone below 1400 °C is termed the low-temperature liquid phase zone. Based on the composition of the slag and the limitations of the software ternary system phase diagram, phase diagrams were plotted with y = 0 and x = 0.05, 0.06, 0.07, respectively, and the results are shown in Figure 8a–c; phase diagrams were also plotted with x = 0 and y = 0.01, 0.02, 0.04, respectively, and the results are shown in Figure 8d–f [22].

Analysis of the phase diagram shows that the low-temperature liquid phase zone is concentrated in the low alkalinity and 10–30% Al₂O₃ region. Both the alkalinity and Al₂O₃ content have a significant impact on the melting temperature, which is consistent with the melting law obtained in the previous section. Meanwhile, as the MgO content increases from 5% to 7%, the overall liquid phase area decreases. While the low-temperature liquid phase zone shows no significant change, the high-temperature liquid phase zone changes significantly. As the FeO content increases from 1% to 4%, the overall liquid phase zone remains largely unchanged. When the FeO content is 4%, a new low-temperature liquid phase zone emerges in the 0–10% SiO₂ region, and the existing low-temperature liquid phase zone in the 70% SiO₂ region expands. In comparison, the variations in MgO content have a more significant impact on the melting temperature than the change in FeO content, which aligns with the melting trends observed previously.

In summary, to configure slag with suitable melting characteristics, it is necessary to control the content of each component within 1.2~1.6 alkalinity, 25~30% Al₂O₃, 4~6% MgO, and 1–2% FeO.

3.3. The Influence of Slag Oxidation on Aluminum Loss

In order to reduce the aluminum loss in steel, it is crucial to control the oxidation of slag. The reduction of dissolved aluminum in the RH refining process steel is defined as the loss value of the dissolved aluminum, denoted as ΔAls, and the oxidation of slag is generally represented by the level of FeO content in the slag [23]. Figure 9 shows the statistical charts of the FeO and ΔAls content in the slag of multiple groups after RH refining.

As shown in the figure, as the FeO content in the slag increases, the ΔAls in the steel also increases. This is because the FeO content in the ladle slag increases, and the oxidation of the slag increases. The slag will transfer more oxygen to the steel liquid, and the reaction between oxygen and dissolved aluminum in the steel liquid causes secondary oxidation, which increases the loss of dissolved aluminum and affects the cleanliness of the steel liquid. Therefore, it is necessary to control the FeO content at a lower range [24,25].

Using the Equilib calculation module in Factsage thermodynamic software, we simulated and calculated the effects of each component’s content on the FeO content in RH refined slag and changed the content of a single component to calculate the molar mass, mass fraction, and activity of FeO in the slag. The calculation results are shown in Figure 10.

As shown in Figure 10, the trend of the FeO molar mass and mass fraction changes in the refined slag remains consistent. The following text only explains the change in the FeO mass fraction, while the change in FeO activity shows another trend.

As the mass fraction of Al₂O₃ increases, the mass fraction of FeO also increases, while the activity of FeO decreases. As the MgO mass fraction increases, the FeO mass fraction decreases, with a sharp decrease observed at 9% MgO. The activity of FeO first increases and then decreases with the increase in the MgO mass fraction, peaking at 6% MgO. The FeO mass fraction decreases with the increasing CaO mass fraction, and the FeO activity first increases and then decreases with the increasing CaO mass fraction. At 30% CaO, the FeO activity reaches its peak. As the mass fraction of SiO₂ increases, the mass fraction of FeO also increases, while the activity of FeO gradually decreases. Within the range of 16% to 26% SiO₂, the FeO mass fraction increases at a relatively fast rate, and the trend of activity decline is slow. The growth rate of the FeO mass fraction is slow within the range of 26~36% SiO₂, and the decreasing trend of activity accelerates.

Overall, within the experimental content range, the following phenomena can be determined: both the Al₂O₃ and SiO₂ mass fractions are positively correlated with the FeO mass fraction in slag and negatively correlated with the FeO activity. A change in the SiO₂ content with a mass fraction between 16% and 28% has a significant impact on the FeO mass fraction in the slag but only a minor effect on the FeO activity. In contrast, a change in SiO₂ content with a mass fraction between 28% and 36% has little influence on the FeO mass fraction in the slag but substantially affects FeO activity. The mass fractions of MgO and CaO are negatively correlated with the mass fraction of FeO in the slag. An increase in their mass fraction causes the FeO activity to first increase and then decrease, with MgO having a more pronounced effect on the content of FeO in the slag [26,27].

After clarifying the influence of each component on the FeO content in the slag, five sets of slag were selected for composition determination, and the changes in the content of each component in the slag before and after RH exit were recorded. The specific data are shown in

Table 6. Content of various components in slag before and after RH exit from the station.

Group		Al2O3	Alkalinity	MgO	FeO	∆FeO
1	before	27.54	1.29	5.3	1.24	−0.23
	after	28.49	1.27	6.7	1.01
2	before	29.91	1.81	6.17	0.62	−0.14
	after	30.01	1.81	6.65	0.48
3	before	29.78	1.73	6.23	0.89	0.01
	after	28.37	1.07	5.71	0.90
4	before	29.85	1.72	6.1	0.99	−0.07
	after	27.51	0.92	5.48	0.92
5	before	25.562	1.14	5.64	2.18	−0.86
	after	25.984	0.95	5.94	1.32

. Comparing five sets of data, it can be found that slag composed of a relatively low content Al₂O₃, MgO, and alkalinity and a higher content of FeO can lower the FeO content after refining; the influence of the Al₂O₃ content on the FeO content is greater than that of the MgO content and greater than that of the alkalinity [28].

Throughout the RH refining process, the oxidation of the ladle slag remains elevated. The free oxygen in the slag continuously transfers oxygen to the steel, causing secondary oxidation of the steel; as shown in Equations (3) and (4), part of [O] reacts with Als in the steel to generate more Al₂O₃ inclusions, and at the same time, Als in the steel also react with FeO in the slag to form additional Al₂O₃ inclusions. The reactions increase the loss of Als in the steel and generate a large number of inclusions that impair the cleanliness of the steel. The detrimental effect of oxidizing slag on steel cleanliness persists throughout the entire refining process until its conclusion. Therefore, additional measures should be implemented in subsequent processes to further reduce the FeO content in the slag after RH refining [29,30,31].

$$2\left[\mathrm{A}\mathrm{l}\right]+3\left[\mathrm{O}\right]=\left(\mathrm{A}{\mathrm{l}}_2{\mathrm{O}}_3\right)$$

(3)

2Al + 3FeO = Al₂O₃ + Fe

(4)

4. Conclusions

This study combines thermodynamic simulation and industrial verification to build an efficient framework for refining slag to reduce trial and error, clarifies the synergistic effects of slag components, identifies optimized slag compositions for grain-oriented silicon steel, bridges the gap between principles and plant parameters, and provides a solution for improving the quality and efficiency.

(1) Through the study of the composition of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system, it is found that alkalinity has a significant impact on the melting temperature of the slag, while the content of Al₂O₃, MgO, and FeO in the slag has a relatively small impact on the melting temperature of the slag. To ensure that the slag is in the lower melting temperature range, slag in the range of 1–1.4 alkalinity, 25–30% Al₂O₃, 4–6% MgO, and 0–2% FeO should be selected.

(2) Through the study of the composition of the CaO-SiO₂-Al₂O₃-FeO-MgO slag system, it is found that alkalinity has a significant impact on the viscosity of the slag, while the content of Al₂O₃, FeO, and MgO in the slag has a relatively small impact on the melting temperature of the slag. To ensure that the slag is in the lower viscosity range, slag in the range of 1–1.4 alkalinity, 25–30% Al₂O₃, 4–6% MgO, and 0–2% FeO should be selected.

(3) Oxidative slag will continue to harm the cleanliness of steel, leading to an increased loss of dissolved aluminum in the steel. The oxidation of slag mainly depends on the FeO content in the slag, and the higher the FeO content, the greater the loss of dissolved aluminum. The alkalinity of slag has a significant impact on the FeO content, while the influence of Al₂O₃ and MgO in slag on the FeO content is relatively small. The requirement for controlling the FeO content in slag should be to select slag in the range of 1~1.4 alkalinity, 25~30% Al₂O₃, 4~6% MgO, and 1–2% FeO.

(4) Based on the thermodynamic analysis, the influence of the slag melting characteristics, the calculation of the FeO content, and the industrial experiments, the suitable composition range of RH refining slag is slag in the range of 1.3~1.5 alkalinity, 25~30% Al₂O₃, 5~6% MgO, and 1–2% FeO.