Dephosphorization of Hot Metal Containing High Phosphorus Using F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ Slag

Abstract

With the extensive usage of domestic low-grade iron ore with high phosphorus, abundant high (> 0.35%) phosphorus hot metal generates in China. To solve the dephosphorization issue of high-P hot metal under the premise of avoiding the environmental problem caused by CaF₂ flux agent, a type of F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag was designed for the dephosphorization of high-P hot metal and systematically tested and analyzed combined with FactSage software under 1400 °C in this paper. The results show that it is an effective approach to remove phosphorus from hot metal-containing high phosphorus using F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag. The major driving force for this dephosphorization comes from the precipitation of ${\text{Ca}}_{{2}} {\text{SiO}}_{{4}} \cdot n{\text{Ca}}_{{3}} {\text{(PO}}_{{4}}^{{}} {)}_{{2}}$ (C2S-C3P) phase which lowers the activity of phosphorus in liquid slag by providing a site to stabilize phosphorus in slag. The increase of basicity enhances the dephosphorization process by promoting the precipitation of the P-rich C2S-C3P phase. The Fe₂O₃ in slag favors the removal of phosphorus from hot metal-containing high phosphorus using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag by improving the oxidability of slag. The Al₂O₃ in slag will compete with P₂O₅ to combine with CaO and SiO₂, which inhibits the formation of C2S-C3P and thus weakens the dephosphorization ability of slag.

1 Introduction

With the massive exploitation of mineral resources all over the world, the high-quality iron ore is at the point of exhaustion and its cost is soaring. In this context, more and more domestic steel enterprises in China have no choice but to use more domestic low-grade iron ore with high phosphorus as an alternative [1]. As a consequence, the phosphorus content in the produced hot metal is getting to a much higher level, generating a great deal of high (> 0.35%) phosphorus hot metal [2]. For the end products of steelmaking process, phosphorus is a harmful impurity that should be controlled to be a very low level. To remove phosphorus from high phosphorus hot metal efficiently, using conventional slag with high basicity and good fluidity as a dephosphorization agent is usually a solution. To lower the melting point of the dephosphorization slag and enhance the kinetic condition, CaF₂ is generally added as a flux agent.

However, CaF₂ in dephosphorization slag not only accelerates the erosion of the lining of the converter and ladle [3,4], but also reacts with SiO₂ in slag to produce SiF₄ gas which is harmful to the human body [5,6,7]. Moreover, CaF₂ in slag is very likely to dissolve into the groundwater system to cause serious pollution. For these reasons, the research on F-free dephosphorization slag is attracting more and more attention. As a result, various types of F-free dephosphorization slag were developed by replacing CaF₂ with oxides with a low melting point, such as Na₂O, Li₂O, K₂O and B₂O₃ [8,9,10,11]. However, most of these slags were designed for the dephosphorization of hot metal with medium or low phosphorus, and the research concerning hot metal with high phosphorus has hardly been reported.

It is widely believed that 2CaO·SiO₂ will combine with 3CaO·P₂O₅ in slag to form 2CaO·SiO₂–3CaO·P₂O₅ solid solution which plays a significant role in improving the dephosphorization efficiency from hot metal [12,13,14]. Based on this idea, CaO–SiO₂–Al₂O₃–Fe₂O₃ slag was designed for the dephosphorization of hot metal-containing high phosphorus in this paper. Among the components in the slag, the CaO and SiO₂ are added to form 2CaO·SiO2 which provides a site for phosphorus enrichment, the Fe₂O₃ is added to guarantee a strong oxidability, and a proper amount of Al₂O₃ is added to promote the melting performance of the slag [15]. To verify this thought, a series of high-temperature (1400 °C) experiments of removing phosphorus from high-P hot metal using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag with various basicity (CaO/SiO₂), Fe₂O₃ content and Al₂O₃ content were carried out in the present study. The microstructure of the dephosphorization slag was characterized using a scanning electron microscope (SEM) equipped with energy-dispersive spectrometry (EDS) and the key factors (basicity, Fe₂O₃ content and Al₂O₃ content) were discussed with the help of a thermodynamic software named FactSage.

2 Experimental

2.1 Material Preparation

The initial charge material for the experimental melts was taken from on-site pig iron, and the composition is shown in Table 1. To obtain targeted molten iron containing high phosphorus (around 0.48%), a certain amount of ferrophosphorus was added together with the initial charge material. The composition of the ferrophosphorus is listed in Table 2. All slags were prepared by mixing analytically pure CaO, Fe₂O₃, SiO₂, and Al₂O₃. Before experiments, these slag components were oven-dried at 473 K for 10 h to remove moisture.

Table 1 Chemical composition of the initial iron charge material (wt-%)

Table 2 Chemical composition of the ferrophosphorus for experimental use (wt.%)

2.2 Experimental Procedure

The melting experiments were performed in a molybdenum disilicide resistance furnace with six MoSi₂ rods as heating elements, as shown in Fig. 1. Before the experiments, an empty corundum crucible was put into the constant temperature zone of the molybdenum disilicide resistance furnace and the inner temperature of the furnace was raised to the experimental temperature. Then, a B-type (Pt-30% Rh/Pt) calibration thermocouple was used to measure the temperature inside the crucible. And the temperature near the MoSi₂ rod is automatically measured by the build-in thermocouple of the furnace and displays on the dashboard to indicate the experimental temperature. Finally, the temperature displayed on the dashboard and measured using calibration thermocouple was compared to eliminate the temperature measurement error caused by the position difference between the internal thermocouple and the inside of crucible .

Fig. 1

Schematic diagram of molybdenum disilicide resistance furnace

Throughout the experiments, Ar gas (99.9% purity) was introduced from the top of the furnace to prevent the oxidation of the hot metal. Firstly, 500 g of pig iron together with ferrophosphorus was melted in an alumina crucible held by an outer protective graphite crucible within a MoSi₂ furnace at 1673 K. After 10 min holding time which is enough for homogenizing the composition of hot metal, an iron sample was withdrawn from the hot metal using a quartz tube coupled with a rubber suction bulb and rapidly quenched in cold water. Then, 100 g of pre-mixed dephosphorization slag with various constitute was added on the liquid level of hot metal through a quartz tube. The designed compositions of slags for different heats are shown in Table 3. After the reaction between slag and hot metal for 25 min, an iron sample was taken as described previously and a slag sample was withdrawn using a Mo rod and rapidly quenched in cold water. It should be noted that the compositions of slag and hot metal as well as the reaction temperature and time were determined according to the practical parameter in production.

Table 3 Initial composition of dephosphorization slag

2.3 Analysis

Each iron sample was broken into powder and then examined by inductively coupled plasma mass spectrometry (ICP-MS) to determine the [P] content. Each slag sample was divided into two portions. One portion was ground into powder and analyzed by X-ray diffraction (XRD), the other was mounted in resin, ground, polished, gold-sprayed and then analyzed using a scanning electron microscope equipped with an energy-dispersive spectrometer (SEM–EDS) to characterize the microstructure of dephosphorization slag.

3 Results and Discussion

3.1 Effectiveness of Dephosphorization

The dephosphorization efficiency $\eta_{\text{P}}$ is calculated with the following formula:

$$\eta_{\text{P}} = \frac{{[{\text{P}}]_{i} - [{\text{P}}]_{t} }}{{[{\text{P}}]_{i} }} \times 100\%$$

(1)

where $[{\text{P}}]_{i}$ represents the phosphorus content in hot metal before dephosphorization, $[{\text{P}}]_{t}$ represents the phosphorus content in hot metal before dephosphorization.

The calculated results are plotted in Fig. 2. It can be seen that most of the heats obtained dephosphorization efficiency higher than 60%, which indicates good dephosphorization effectiveness. This proves the feasibility of removing phosphorus from hot metal-containing high phosphorus using F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag.

Fig. 2

Dephosphorization efficiency in different heats

The typical XRD diffraction pattern of dephosphorization slag is shown in Fig. 3. It indicates that amounts of Ca₇Si₂P₂O₆, as well as little Fe₃O₄ and Ca₂Al₂SiO₇ phases, precipitate from molten slag after dephosphorization reaction.

Fig. 3

Typical XRD diffraction pattern of dephosphorization slag

The typical mapping image of dephosphorization slag analyzed by SEM–EDS is shown in Fig. 4. It can be seen that amounts of black phases consisting of Ca, Si, P and O elements precipitate from the slag matrix. Combined with the XRD analysis results, the black phase can be identified as Ca₇Si₂P₂O₆. It should be noted that the phosphorus element in the dephosphorization slag enriches within the black phase. It suggests that the precipitation of the Ca₇Si₂P₂O₆ phase provides a site to stabilize phosphorus, which lowers the activity of phosphorus in the slag matrix, thus promotes the dephosphorization reaction between slag and hot metal.

Fig. 4

Typical mapping image of dephosphorization slag, a represents the original SEM image, b represents the mapping results

3.2 Effect of Dephosphorization Composition

3.2.1 Effect of Basicity

The relationship between dephosphorization efficiency and the basicity of dephosphorization slag (Al₂O₃ content fixed at 3% and Fe₂O₃ content fixed at 60%) is shown in Fig. 5. The dephosphorization efficiency of hot metal using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag shows an overall increasing trend with the increase of slag basicity. This suggests that high slag basicity favors the transfer of phosphorus from hot metal to slag.

Fig. 5

Relationship between dephosphorization efficiency and basicity of dephosphorization slag

To reveal the relevant mechanism, the reaction between slag and hot metal was simulated using the Equilib module of thermodynamic software FactSage 7.2. The initial condition for calculation was set as: w(Al₂O₃) = 3%, w(Fe₂O₃) = 60%, basicity ranges from 2 to 5.5, the composition of hot metal is 4.3%[C]–0.45%[Si]–0.87%[Mn]–0.48%[P]–0.045%[S], ratio of slag to hot metal is 8%. The simulated relationship between dephosphorization efficiency and the basicity of dephosphorization slag is shown in Fig. 6. It is evident that the dephosphorization efficiency increase with the increase of slag basicity, which is similar to the rules observed in the experimental conditions in Fig. 5. This verifies the relationship between dephosphorization efficiency and basicity, and also proves the reliability of the simulated results. The simulated calculation also gives us the changes in the activity of P₂O₅ in the liquid part of slag and precipitation amount of solid solution of ${\text{Ca}}_{{2}} {\text{SiO}}_{{4}} \cdot n{\text{Ca}}_{{3}} {\text{(PO}}_{{4}}^{{}} {)}_{{2}}$ (abbreviated to C2S-C3P) with increasing slag basicity, as shown in Fig. 7. As we all know, phosphorus in hot metal is removed by the following equilibrium Reactions (2) and (3) [16]. As shown in Fig. 7, when slag basicity increases, the amount of P-rich C2S-C3P phase increases gradually, which consumes more P₂O₅ through Reaction (3) and further leads to the decrease of the activity of P₂O₅ in the liquid slag. The decrease of the activity of P₂O₅ accelerates the proceeding of Reaction (2), and thus favors the removal of phosphorus. In brief, the increase of basicity enhances the dephosphorization process by promoting the precipitation of the P-rich C2S-C3P phase.

Fig. 6

Simulated relationship between dephosphorization efficiency and slag basicity

Fig. 7

Changes in the activity of P₂O₅ in the liquid part and precipitation amount of C2S-C3P with slag basicity

$$5\left( {{\text{FeO}}} \right)\, + \,2\left[ {\text{P}} \right]\, = \,5\left[ {{\text{Fe}}} \right]\, + \,\left( {{\text{P}}_{2} {\text{O}}_{5} } \right)$$

(2)

$$\left( {3n\, + \,2} \right){\text{CaO}}\, + \,n{\text{P}}_{2} {\text{O}}_{5} \, + \,{\text{SiO}}_{2} \, = \,{\text{Ca}}_{2} {\text{SiO}}_{4} \cdot n{\text{Ca}}_{3} \left( {{\text{PO}}_{4} } \right)_{2}$$

(3)

3.2.2 Effect of Fe₂O₃

The relationship between dephosphorization efficiency and Fe₂O₃ content in dephosphorization slag (Al₂O₃ content fixed at 3% and basicity fixed at 3.5) is shown in Fig. 8. It is evident that the dephosphorization efficiency of hot metal using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag shows an overall increasing trend with the increase of Fe₂O₃ content in the initial slag. This suggests that Fe₂O₃ favors the removal of phosphorus from hot metal. Quite understandably, it is because that Fe₂O₃ improves the oxidability of slag, thus accelerating the proceeding of Reaction (2) and enhancing the removal of phosphorus in hot metal.

Fig. 8

Relationship between dephosphorization efficiency and initial Fe₂O₃ content in dephosphorization slag

3.2.3 Effect of Al₂O₃

The relationship between dephosphorization efficiency and Al₂O₃ content in dephosphorization slag (slag basicity fixed at 4 and Fe₂O₃ content fixed at 60%) is shown in Fig. 9. It is obvious that the dephosphorization efficiency of hot metal using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag shows an overall decreasing trend with the increase of Al₂O₃ content. This suggests that Al₂O₃ plays an adverse role in the process of dephosphorization from hot metal.

Fig. 9

Relationship between dephosphorization efficiency and Al₂O₃ content in initial slag

To better understand this phenomenon, the reaction between hot metal and slag containing various Al₂O₃ content was simulated using the Equilib module of thermodynamic software FactSage 7.2. The initial condition for calculation was set as: basicity is 4, w(Fe₂O₃) = 60%, Al2O3 content ranges from 0 to 10%, the composition of hot metal is 4.3%[C]–0.45%[Si]–0.87%[Mn]–0.48%[P]–0.045%[S], ratio of slag to hot metal is 8%. The simulated relationship between dephosphorization efficiency and the basicity of dephosphorization slag is shown in Fig. 10. It can be seen that the dephosphorization efficiency increase with the increase of Al₂O₃ content, which is similar to the rules observed in the experimental conditions in Fig. 9. This verifies the relationship between dephosphorization efficiency and Al₂O₃ content in slag, and also proves the reliability of the simulated results. The simulated calculation also gives us the changes in the activity of P₂O₅ in the liquid part of slag and the precipitation amount of solid solution of C2S-C3P with increasing slag basicity, as shown in Fig. 11. As the increase of Al₂O₃ content in slag, more Al₂O₃ competes with P₂O₅ to combine with CaO and SiO₂ by Reaction (4) [17], which inhibits the formation of C2S-C3P and also reduces the consumption of P₂O₅, leading to the increase of activity of P₂O₅. As a consequence, the dephosphorization efficiency gradually weakens with increasing Al₂O₃ content in slag.

$$a{\text{CaO}}\, + \,b{\text{A}}_{2} {\text{O}}_{5} \, + \,c{\text{SiO}}_{2} \, = \,a{\text{Al}}_{{2}} {\text{O}}_{{3}} \cdot b{\text{SiO}}_{{2}} \cdot c{\text{CaO}}$$

(4)

Fig. 10

Simulated relationship between dephosphorization efficiency and Al₂O₃ content in slag

Fig. 11

Changes in the activity of P₂O₅ in the liquid part and precipitation amount of C2S-C3P with Al₂O₃ content

4 Conclusions

In the present study, a type of F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag was designed for the dephosphorization of hot metal-containing high phosphorus, which turns to be an efficient approach in dephosphorization effect. Through a series of high-temperature experiments by employing F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag with different compositions, combined with the simulated thermodynamic calculation by FactSage, the dephosphorization mechanism and the key factors such as basicity, Fe₂O₃ content and Al₂O₃ content were discussed. The major results are listed below:

• 1. It is an effective approach to remove phosphorus from hot metal-containing high phosphorus using F-free CaO–SiO₂–Al₂O₃–Fe₂O₃ slag. The major driving force for this dephosphorization comes from the precipitation of the C2S-C3P phase which lowers the activity of phosphorus in liquid slag by providing a site to stabilize phosphorus in slag.

• 2. The dephosphorization efficiency of hot metal using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag shows an overall increasing trend with the increase of slag basicity. The increase of basicity enhances the dephosphorization process by promoting the precipitation of the P-rich C2S-C3P phase.

• 3. The Fe₂O₃ in slag favors the removal of phosphorus from hot metal-containing high phosphorus using CaO–SiO₂–Al₂O₃–Fe₂O₃ slag by improving the oxidability of slag.

• 4. The Al₂O₃ in slag weakens the dephosphorization ability. As the increase of Al₂O₃ content in slag, more Al₂O₃ competes with P₂O₅ to combine with CaO and SiO₂, which inhibits the formation of C2S-C3P and also reduces the consumption of P₂O₅. This leads to the increase of activity of P₂O₅ and thus weakens the dephosphorization ability of slag.