Regular Article
Copper Removal of Liquid Steel Containing 0.25% Carbon Using Fe2O3–CaCl2–SiO2 Flux
2025 Volume 65 Issue 11 Pages 1597-1606
Details
2025 Volume 65 Issue 11 Pages 1597-1606
Copper is recognized as one of the most detrimental residual elements in steelmaking due to its adverse effects during the thermo-mechanical processing of steel. Selective chlorination volatilization has emerged as a promising method for removing residual copper from liquid steel. In this study, the selective chlorination behavior of copper was systematically investigated using Fe2O3–CaCl2–SiO2 flux at 1873 K, focusing on optimizing flux composition parameters: Fe2O3/Cu molar ratio (n(Fe2O3)/n(Cu)), CaCl2/Cu molar ratio (n(CaCl2)/n(Cu)), and SiO2/CaCl2 molar ratio (n(SiO2)/n(CaCl2)). Experimental results demonstrated an ideal copper removal ratio of 39.68% under optimized conditions, achieved with a flux composition of n(Fe2O3)/n(Cu) = 10.0, n(CaCl2)/n(Cu) = 1.0, and n(SiO2)/n(CaCl2) = 1.0. Phase analysis revealed that the primary chlorination volatiles consisted of CuOHCl, Cu2Cl(OH)3, and FeCl2(H2O)2, confirming the mechanistic pathway of copper removal. This work substantiates the technical feasibility of selective chlorination for copper elimination from liquid steel and provides a viable strategy for enhancing scrap steel recycling in sustainable metallurgical practices. The proposed method demonstrates significant potential for industrial application in residual element control.
In recent years, China has emerged as the world’s largest steel producer, driven by rapid expansion in steel output. Concurrently, the impending end-of-life phase of steel products is projected to generate substantial volumes of scrap steel for recycling.1,2,3) Amid China’s carbon reduction initiatives, optimizing scrap steel utilization has become an essential strategy for steel manufacturers pursuing sustainable development and aligning with national green industrial policies.4,5,6) However, the accumulation of residual elements during recycling, particularly copper (Cu), poses a critical challenge to steel quality. Copper impurities induce interfacial segregation, exacerbate hot shortness, and diminish ductility and deep-drawing properties in steel products.7,8,9) These detrimental effects underscore the urgent need for innovative metallurgical processes capable of efficiently removing copper from liquid steel.10) Developing such technologies is pivotal to maintaining copper concentrations within permissible limits during steel production, thereby enabling the manufacture of high-value steel grades while supporting circular economy goals. Several methods, include dilution method,11,12) sorting method,13) sulfide treatment,14,15) vacuum distillation16,17) and selective chlorination,18) have been developed for residual elements removal from liquid steel. However, above approaches suffer from quality inconsistency, incomplete removal, or environmental burdens.19)
Selective chlorination emerges as a promising alternative, leveraging differential chloride volatility for separating and extracting nonferrous metals.20,21,22) It is characterized by its low-cost, high efficiency,23) low energy consumption,24) and exceptional selectivity.25) In general, the high volatility of metal chlorides makes this method particularly attractive for extracting metals from ores and electronic waste.26) Conventional chlorinating agents include Cl2, HCl, and solid chlorides. Due to the low cost, high volatility, low corrosivity, easy availability and controllability of the chlorine source, the solid chlorinating agent often acts as a chlorine donor in the reaction.27) According to the literature, calcium chloride is more advantageous solid chlorinating agent.28,29,30) Hu Xiaojun31,32) proposed to remove residual element copper from liquid steel by selective chlorination method, and demonstrated the feasibility of this approach for the separation of copper in Fe–Cu–O system through both thermodynamic analysis and experimental validation. The selective chlorination process offers significant advantages in terms of efficiency, overcoming the limitations of existing technologies, and providing a promising solution for scrap steel recycling. Based on the previous work of our group,33) approximately 95.0% of copper was removed from FexOy–Cu2O–CaCl2–SiO2 system under specific experimental conditions and the primary phases identified in chlorination products were iron oxides and silicates. Although the feasibility of removing residual copper from liquid steel via selective chlorination has been demonstrated, no published studies have systematically investigated the influence of various chlorination parameters on copper removal efficiency in liquid steel. It is well known that the composition and dosage of chlorinated flux are critical factors influencing the effectiveness of the chlorination process. Therefore, it is highly necessary to explore how chlorination parameters impact copper removal efficiency to achieve effective removal of residual element copper.
In this study, the thermodynamic feasibility of selectively chlorinating residual elemental copper in copper-containing liquid steel using Fe2O3–CaCl2–SiO2 flux was investigated using Factsage software. Additionally, the effects of various chlorination parameters on copper removal from liquid steel were examined in detail at 1873 K under an argon atmosphere. The results confirmed the feasibility of selectively chlorinating copper-containing liquid steel with Fe2O3–CaCl2–SiO2 flux. This method proved to be an effective technique for removing copper from liquid steel, offering advantages such as lower economic cost and shorter reaction time, which enhance its potential for scaling up to industrial applications.
Copper-containing steel was prepared using a vacuum induction heating furnace. The specific procedure involved proportionally placing electrolytic iron, pure copper particles, and high-purity graphite into a crucible, followed by melting under dynamic vacuum with electromagnetic stirring. After solidification, the copper-containing steel was sectioned to coupons using wire electrical discharge machining for subsequent trials. Its main components are shown in Table 1.
| No. | Cu | C | Fe | S | O | N | H |
|---|---|---|---|---|---|---|---|
| 1 | 0.50 | 0.25 | 99.24 | 0.0002 | 0.0019 | 0.0001 | 0.0001 |
| 2 | 0.25 | 0.25 | 99.49 | 0.0009 | 0.0037 | 0.0003 | 0.0001 |
The analytical reagents employed in the study included Fe2O3, CaCl2, and SiO2. The calcium chloride (CaCl2) was dehydrated at 473 K for 24 hours to serve as the chlorinating agent.34) The chemical reagents Fe2O3, CaCl2, and SiO2 were thoroughly mixed before being placed into an iron crucible, which served as the feeding device for the selective chlorination process. As illustrated schematically in Fig. 1, 150.0 g of a copper-containing steel sample was placed in an alumina crucible and heated in a tube furnace under a high-purity argon gas atmosphere (99.999%) at a flow rate of 100 mL/min. Once the furnace temperature reached 1873 K according to the heating procedure, it was maintained for 30 min to ensure complete melting of the copper-containing steel. Subsequently, the iron crucible containing the flux was inserted into copper-containing liquid steel located in the hot zone of tube furnace through a feeding channel, allowing the reaction to occur. To determine the optimal chemical proportion of Fe2O3–CaCl2–SiO2 flux to remove copper from liquid steel, the effect of Fe2O3 content (n(Fe2O3)/n(Cu)), CaCl2 content (n(CaCl2)/n(Cu)), and SiO2 content (n(SiO2)/n(CaCl2)) on the copper removal efficiency were investigated systematically in accordance (in Table 2). Previous studies have indicated that a chlorination time of 10 min is sufficient to achieve equilibrium,18) so the reaction time was set to 10 min following each addition of flux. At 1873 K, the flux was added to No.1 liquid steel in five additions over a period of 50 min. Similarly, the flux was added to No.2 liquid steel in three additions over a period of 30 min. Systematic sampling was conducted at 10-minute intervals, with collected specimens undergoing immediate quenching, proper labeling, and archival storage. After the reaction, the crucible was allowed to cool to room temperature under an argon atmosphere following the pre-set procedures. The condensed gas products inside the quartz tube were promptly collected and stored. The obtained steel samples and chlorinated products were properly preserved for subsequent analysis.
| No. | Liquid steel | n(SiO2)/n(CaCl2) | n(Fe2O3)/n(Cu) | n(CaCl2)/n(Cu) | Mass/g | Final [Cu] content/% |
|---|---|---|---|---|---|---|
| 1-1 | No.1 | 0.5 | 5.0 | 1.0 | 11.28 | 0.29 |
| 1-2 | 1.0 | 5.0 | 1.0 | 11.71 | 0.28 | |
| 1-3 | 1.5 | 5.0 | 1.0 | 12.08 | 0.30 | |
| 1-4 | 2.0 | 5.0 | 1.0 | 12.74 | 0.28 | |
| 1-5 | 1.0 | 2.5 | 1.0 | 6.80 | 0.33 | |
| 1-6 | 1.0 | 7.5 | 1.0 | 16.10 | 0.28 | |
| 1-7 | 1.0 | 10.0 | 1.0 | 20.75 | 0.22 | |
| 1-8 | 1.0 | 15.0 | 1.0 | 30.19 | 0.21 | |
| 1-9 | 1.0 | 10.0 | 0.5 | 19.97 | 0.27 | |
| 1-10 | 1.0 | 10.0 | 1.5 | 22.06 | 0.22 | |
| 1-11 | 1.0 | 10.0 | 2.0 | 22.97 | 0.20 | |
| 2-1 | No.2 | 1.0 | 2.5 | 1.0 | 3.45 | 0.17 |
| 2-2 | 1.0 | 5.0 | 1.0 | 5.87 | 0.17 | |
| 2-3 | 1.0 | 7.5 | 1.0 | 8.28 | 0.10 | |
| 2-4 | 1.0 | 10.0 | 1.0 | 10.70 | 0.09 | |
| 2-5 | 1.0 | 15.0 | 1.0 | 16.10 | 0.13 |
In this study, the oxygen and nitrogen content in the steel were determined by an oxygen-nitrogen analyzer (EMGA-830, Horiba, Japan), while the carbon and sulfur content were determined by a carbon-sulfur analyzer (EMIA-920V2, Horiba, Japan). The phase compositions of the collected gas products were identified at ambient temperature using X-ray diffraction (XRD, D8 Advance, Bruker, Germany) with a Cu Kα radiation source, scanning diffraction angles from 10° to 70°. The composition analysis of samples using wavelength dispersive X-ray fluoroscopy (XRF, EDX-8000, Shimadzu, Japan). The copper content in liquid steel was quantified by inductively coupled plasma-mass spectroscopy (ICP-MS, iCAP-RQ, Thermo Scientific, Germany).
The potential chemical reactions involved in copper removal from liquid steel through the Fe2O3–CaCl2–SiO2 flux are summarized in Table 3. The removal mechanism was systematically determined through thermodynamic calculations performed using FactSage 8.0 software, with the computational results presented in Fig. 2. Standard Gibbs free energy change (ΔGθ) refers to the change in Gibbs free energy of a system when reactants are completely converted into products under standard conditions. As demonstrated in Fig. 2(a), reactions (1)–(4) exhibit positive values of standard Gibbs free energy change, rendering them thermodynamically unfavorable but showing increasing feasibility with temperature elevation. Crucially, comparative analysis reveals copper’s preferential oxidation pathway via Fe2O3 (Reaction (5)), forming Cu2O intermediate prior to chlorination. Furthermore, compared to other oxides of copper and iron in Fig. 2(b), Cu2O reacts more readily with CaCl2 to form CuCl at stoichiometric ratio (n(CaCl2)/n(Cu) = 0.5). Meanwhile, the volatilization of copper removal products from liquid steel further promotes the chlorination of copper, achieving the goal of removing residual copper from the system. In addition, it is also crucial to inhibit the chlorination of iron oxides during the reaction process. As illustrated in Fig. 2(c), the standard Gibbs free energy change values at 1873 K for reactions (10)–(13) are negative, indicating that SiO2 readily reacts with CaO and FeO to form silicates. The copper removal from liquid steel by selective chlorination using Fe2O3–CaCl2–SiO2 flux is shown in reactions (14)–(17). Figure 2(d) reveals that the standard Gibbs free energy change values of reactions (14)–(15) decrease with increasing reaction temperature, while that of reaction (17) increases. At 1873 K, the standard Gibbs free energy change values for reactions (14) and (15) are both negative, indicating that the reactions proceed spontaneously under these conditions. Notably, the formation of calcium silicates (Ca2SiO4 and Ca3SiO4) emerges as the predominant reaction pathway in this system. These thermodynamic calculations substantiate the technical viability of employing Fe2O3–CaCl2–SiO2 ternary flux for efficient removal of residual elemental copper from liquid steel. In order to promote the volatilization of copper during chlorination process, SiO2 was incorporated into the selective chlorination flux.
| Reaction | No. |
|---|---|
| 0.5Fe3O4 + Cu = 0.5Cu2O + 1.5FeO | (1) |
| Fe3O4 + Cu = CuO + 3FeO | (2) |
| 0.5Fe2O3 + Cu = 0.5Cu2O + FeO | (3) |
| Fe2O3 + Cu = CuO + 2FeO | (4) |
| Cu2O + CaCl2 = 2CuCl + CaO | (5) |
| CuO + CaCl2 = CuCl2 + CaO | (6) |
| FeO + CaCl2 = FeCl2 + CaO | (7) |
| 0.25Fe3O4 + CaCl2 = 0.25FeCl2 + 0.5FeCl3+ CaO | (8) |
| 0.33Fe2O3 + CaCl2 = 0.67FeCl3 + CaO | (9) |
| 2FeO + SiO2 = Fe2SiO4 | (10) |
| 2CaO + SiO2 = Ca2SiO4 | (11) |
| 3CaO + SiO2 = Ca3SiO5 | (12) |
| 0.5CaO + 0.5FeO + SiO2 = 0.5CaFeSi2O6 | (13) |
| Cu + 0.5Fe2O3 + 0.5CaCl2 + 0.25SiO2 = CuCl + 0.25Ca2SiO4 + FeO | (14) |
| Cu + 0.5Fe2O3 + 0.5CaCl2 + 0.17SiO2 = CuCl + 0.17Ca3SiO5 + FeO | (15) |
| Cu + 0.5Fe2O3 + 0.5CaCl2 + 0.5SiO2 = CuCl + 0.25Ca2SiO4 + 0.25Fe2SiO4 + 0.5FeO | (16) |
| Cu + 0.5Fe2O3 + 0.5CaCl2 + SiO2 = CuCl + 0.5CaFeSi2O6 + 0.5FeO | (17) |
The feasibility of selective chlorination for copper removal has been confirmed through thermodynamic calculations. A series of experiments were conducted on liquid steel containing 0.50% Cu using different Fe2O3–CaCl2–SiO2 fluxes at 1873 K. To assess the effectiveness of copper removal, the removal ratio of copper was calculated based on a simple mass balance, as expressed in Eq. (1).
| (1) |
where, RCu is the removal ratio of element Cu, %; m0,steel and mt,steel are the mass of liquid steel in the initial sample and residue, g; w0,Cu and wt,Cu are the content of element Cu in the initial sample and residue, %.
3.2.1. Effect of SiO2 ContentSiO2 acts as a primary reaction promoter in chlorination reactions.35) The addition of SiO2 enhances the chlorination of copper, thereby increasing the chlorine partial pressure and improving the copper removal ratio. As shown in Fig. 3, the effect of SiO2 content on copper removal efficiency was investigated. The copper removal ratio increased steadily with the addition of flux and the extension of holding time. However, the copper removal ratio initially rose and then declined as the n(SiO2)/n(CaCl2) increased. Ultimately, the removal ratio of copper ranged between 22.11% and 27.10% for different n(SiO2)/n(CaCl2). Meanwhile, the reduction in copper content in the liquid steel was approximately 0.075% to 0.097%, as indicated by the initial and final copper contents in Fig. 3(b). In comparison, the copper removal ratio was less affected by variations in n(SiO2)/n(CaCl2) of Fe2O3–CaCl2–SiO2 fluxes. Although the addition of SiO2 facilitates the combination of calcium oxide (CaO) generated during the chlorination reaction, thereby enhancing chlorination efficiency and promoting copper removal, further increases in SiO2 content led to a decline in copper removal efficiency. Specifically, SiO2 can react with CaO, FeO, Fe3O4 and Fe2O3 to form silicates (Ca2SiO4, CaSi2O5, CaFeSi2O6) at high temperatures. With the increasing of SiO2 content in Fe2O3–CaCl2–SiO2 flux, the fluidity of the molten slag becomes weaker. At this time, the chlorination reaction of copper and the volatilization of chloride are inhibited, which is not conducive to further increase the removal ratio of copper. Therefore, appropriate SiO2 content is particularly important for the chlorination reaction of copper removal. n(SiO2)/n(CaCl2) of 1.0 was identified as the optimal parameter for subsequent experiments.
In Reactions (3) and (5), Cu2O serves both as the reaction product of metallic copper transferring from liquid steel into molten slag and as a necessary reactant for chlorination reaction to remove copper. Therefore, the Fe2O3 content of flux is a critical factor influencing the overall reaction during the removal of residual elemental copper from liquid steel. In order to enhance the removal ratio of copper, it is required to determine the optimal flux composition by varying Fe2O3 content at 1873 K. For this purpose, five fluxes with different Fe2O3 content were tested. Figure 4 illustrates the initial and final copper content in liquid steel, as well as the copper removal ratios for experiments using Fe2O3–CaCl2–SiO2 fluxes with varying Fe2O3 content. The copper removal ratio increased with prolonged reaction time and continued to rise with higher Fe2O3 content. For the five fluxes, the copper removal ratios were determined to be 15.88%, 27.10%, 26.70%, 39.68%, and 44.10%, respectively. The results demonstrate that the Fe2O3–CaCl2–SiO2 flux is effective in removing copper from liquid steel. During the copper removal process, the oxidation of copper is fundamental to the subsequent chlorination. The effect of Fe2O3 content on copper removal is primarily reflected in its role in oxidizing copper in liquid steel. When the Fe2O3 content is low, the partial pressure of oxygen is insufficient, leading to a decrease in the removal ratio of copper due to reduced Cu2O formation. The higher Fe2O3 content in Fe2O3–CaCl2–SiO2 flux, the higher partial pressure of oxygen available in the liquid steel to oxidize copper. Thus, the highest copper removal ratio of 44.10% was achieved when n(Fe2O3)/n(Cu) was increased to 15.0. However, when the Fe2O3 content of Fe2O3–CaCl2–SiO2 flux exceeds a certain threshold, the chlorination ratio of Fe2O3 increases, resulting in greater consumption of CaCl2 for iron chlorination. This not only leads to iron loss but also increases the cost of flux and generates more volatile by products. Based on these findings, the optimal Fe2O3 content of Fe2O3–CaCl2–SiO2 flux was determined to be n(Fe2O3)/n(Cu) of 10.0.
In the process of removing residual copper from liquid steel, the CaCl2 content (n(CaCl2)/n(Cu)) of flux significantly influences the reaction rate and extent, making it a critical factor in the chlorination process. To further improve the copper removal ratio, it is necessary to explore more suitable flux compositions by varying the CaCl2 content at 1873 K. As shown in Fig. 5, the copper content of liquid steel decreases as n(CaCl2)/n(Cu) increases, while the copper removal ratio rises correspondingly. When n(CaCl2)/n(Cu) is 0.5, the copper removal ratio is 31.97%. As n(CaCl2)/n(Cu) increases to 1.0, the removal ratio reaches 39.68%. The copper removal ratio reaches 46.02% when (CaCl2)/n(Cu) was further increased to 2.0. The increase in CaCl2 content enhances chlorine availability, accelerates chlorine diffusion, and raises the chlorine partial pressure, thereby promoting the chlorination reaction of copper when n(CaCl2)/n(Cu) ranges from 0.5 to 2.0. These findings highlight that CaCl2 content is a key factor determining the copper removal efficiency. However, excessive CaCl2 content leads to increased volatilization of CaCl2 itself. Additionally, an excess of CaCl2 can cause higher iron loss and generate more volatile byproducts during the reaction. This behavior can be attributed to the combined effects of copper oxidation and chlorination efficiencies. Based on this analysis, the optimal composition of Fe2O3–CaCl2–SiO2 flux for achieving a high copper removal ratio is determined to be n(Fe2O3)/n(Cu) of 10.0, n(CaCl2)/n(Cu) of 1.0, and n(SiO2)/n(CaCl2) of 1.0.
The effectiveness of selective chlorination for removing copper from liquid steel containing 0.25% Cu by fluxes with different Fe2O3 content were investigated at 1873 K. Notably, even with a lower initial copper content in liquid steel, the copper removal effect remains significant. As shown in Fig. 6(a), the copper removal ratio increases with reaction time between 10 and 30 min at 1873 K. After continuous oxidation and chlorination for 30 min, the maximum copper removal ratio of 44.57% was achieved at n(Fe2O3)/n(Cu) of 10.0. When n(Fe2O3)/n(Cu) was increased from 2.5 to 5.0, the change in copper content in liquid steel was minimal, at 0.023% and 0.034%, respectively (Fig. 6(b)). Further increasing n(Fe2O3)/n(Cu) from 7.5 to 15.0 resulted in only a slight change in copper content, ranging from 0.063% to 0.068%. This indicates that a higher Fe2O3 content in flux enhances copper removal efficiency. Additionally, increasing the initial copper content from 0.25% to 0.50% led to a marginal increase in the change of copper content, indicating that higher initial copper levels can achieve a better copper removal effect. Based on these findings, it can be concluded that the copper removal ratio is influenced by both the initial copper content of liquid steel and the composition of flux.
The chlorination products were volatilized to the outside of furnace with the high purity argon gas during the experiment, and subsequently condensed on the inner wall of the quartz tube. XRD analysis was conducted to characterize the phase composition of volatiles generated during copper removal (initial Cu content: 0.50%) from liquid steel using fluxes with varying Fe2O3 and CaCl2 contents. The main constituent phases identified in the volatiles include CuOHCl, Cu2Cl(OH)3, and FeCl2(H2O)2 (in Fig. 7). Special consideration must be given to the unavoidable atmospheric exposure (oxygen and moisture) during post-experimental volatile collection. In view of the easy oxidation and water-absorbing properties of chlorides, it was judged that the appearance of above phases is attributed to the oxidation and hydration of volatiles during the sample collection process. As shown in Fig. 7(a), when n(Fe2O3)/n(Cu) of the flux is 2.5, the diffraction peaks of the resulting volatiles are relatively weak, indicating poor chlorination efficiency for both Cu and Fe. The overall intensity of the chloride diffraction peaks increased significantly as n(Fe2O3)/n(Cu) increased to 5.0. Notably, the increase in the diffraction peak intensity of Cu2Cl(OH)3 confirms that the Fe2O3 content of flux plays a crucial role in the chlorination reaction. As the n(Fe2O3)/n(Cu) increases to 10.0, the intensity of diffraction peaks for chlorides significantly increases, with the diffraction peaks of CuOHCl and Cu2Cl(OH)3 still more obvious. As n(Fe2O3)/n(Cu) continues to rise up to 15.0, the characteristic peak of FeCl2(H2O)2 stabilizes, while that of CuOHCl weakens slightly. To further determine the elemental composition of the chlorinated volatiles, XRF analysis was conducted, and the results are presented in Fig. 8(a). The analysis reveals that the volatiles are rich in elements such as Fe, Cl, Cu, and Ca. Across all samples with different n(Fe2O3)/n(Cu), the percentage of element Cl remains the highest, at approximately 50.0%. The maximum chlorine content of 52.34% is observed at n(Fe2O3)/n(Cu) of 5.0. As n(Fe2O3)/n(Cu) increases further, the iron content in the volatiles shows a slight increase, while the chlorine content gradually decreases. The copper content in the volatiles ranges from 0.38% to 0.61%. The enrichment of copper, iron, and chlorine in these substances confirms the practicality of using selective chlorination with Fe2O3–CaCl2–SiO2 flux to remove residual elemental copper from liquid steel. Similarly, the peaks of volatiles are relatively weak when n(CaCl2)/(Cu)) is 0.50 in Fig. 7(b), suggesting insufficient chlorination reactions for both copper and iron. As n(CaCl2)/(Cu) increases to 1.0, the intensity of chloride peaks increases, particularly CuOHCl, indicating enhanced chlorination of copper. When n(CaCl2)/(Cu) further rises to 2.0, the intensity of CuOHCl and Cu2Cl(OH)3 diffraction peaks in the collected volatiles remains relatively stable, while the intensity of FeCl2(H2O)2 diffraction peaks increases significantly. Although no diffraction peak for CaCl2 is observed, these results suggest that more iron is chlorinated alongside copper as n(CaCl2)/(Cu) increases. As illustrated in Fig. 8(b), the Fe and Cl contents in the volatiles obtained from fluxes with different n(CaCl2)/(Cu) remain high. However, when n(CaCl2)/(Cu) reaches to 1.5 and 2.0, the percentage of element chlorine slightly decreases, while the percentage of element iron increases. Concurrently, as n(CaCl2)/(Cu) increases from 0.5 to 2.0, the percentage of element copper remains stable at 0.38%, 0.68%, 0.78%, and 0.82%, respectively. Additionally, the Ca content remains at a low level, indicating that CaCl2 content of the flux is highly involved in chlorination reaction and is minimally lost. Furthermore, the generation of volatiles increases. From both economic and environmental perspectives, the n(CaCl2)/(Cu) of the flux should not be excessively high.
The molten slag obtained after the copper removal from liquid steel with an initial copper content of 0.50% using Fe2O3–CaCl2–SiO2 fluxes of varying compositions at 1873 K was analyzed by X-ray fluorescence (XRF), as shown in Fig. 9. The primary elements detected in the molten slag were Fe, Ca, Si, and Cu. Notably, no detectable element chlorine was found in the molten slag formed after any compositions of Fe2O3–CaCl2–SiO2 fluxes reaction. The calcium content of the molten slag was particularly significant, indicating that the calcium actively participated in the reaction as an effective chlorinating agent during the chlorination process. Under high-temperature conditions, the presence of Fe2O3–CaCl2–SiO2 flux provides an oxygen source that facilitates the decarburization reaction of liquid steel. To examine the influence of different flux compositions on the carbon content in liquid steel, we measured the carbon content of samples obtained from experiments using various compositions of Fe2O3–CaCl2–SiO2 fluxes to decopperization steel with an initial copper content of 0.50%. As illustrated in Fig. 10, the initial carbon content in liquid steel was 0.25%, and the introduction of flux resulted in a reduction of carbon content. Comparative analysis revealed that for different compositions of fluxes, the carbon content of liquid steel decreased significantly by approximately 0.20% during the first 10 min of reaction time. Subsequently, with continued flux addition and extended reaction time, the carbon content of liquid steel continued to decrease, albeit at a slower rate. The final carbon content of liquid steel predominantly ranged between 0.02% and 0.03% for 50 min. Specifically, as depicted in Fig. 10(a), when the Fe2O3 content of flux was low (with n(Fe2O3)/n(Cu) of 2.5), the carbon content measured 0.059% and 0.057% at 10 min and 50 min, respectively. As the Fe2O3 content of flux increased, the oxidizing capacity intensified, leading to a more pronounced decarburization reaction and a gradual decrease in carbon content. When n(Fe2O3)/n(Cu) reached 15.0, the carbon content dropped to 0.026% and 0.018% at 10 min and 50 min, respectively. Concurrently, the decarburization reaction not only reduced the carbon content in liquid steel but also generated CO gas. This CO gas formation could lead to the creation of gas bubbles, thereby enhancing the fluidity of liquid steel and improving kinetic conditions. In contrast, the CaCl2 content and SiO2 content in Fe2O3–CaCl2–SiO2 flux have little effect on the carbon content of liquid steel. Therefore, in the context of selective chlorination for the removal of residual element copper from liquid steel, better oxygen partial pressure, chloride partial pressure, and kinetic conditions can be achieved by carefully controlling the composition of flux and the carbon content of liquid steel. This ensures that the copper content of liquid steel remains within the desired range.
Based on the experimental results, the mechanism of copper removal from liquid steel using Fe2O3–CaCl2–SiO2 flux involves three key steps, as illustrated in Fig. 11: (1) the oxidation of metallic copper in the liquid steel to form copper oxides, (2) the chlorination of these metal oxides in the molten slag to form metal chlorides, and volatilization to the outside, and (3) the sddition of SiO2 promotes chlorination. This process ultimately converts copper in liquid steel into CuCl, while element iron remains stable in liquid steel. In fact, the presence of Fe2O3 in flux creates a localized high oxygen partial pressure, providing the necessary conditions for metallic copper to be oxidized to Cu2O. Simultaneously, the chlorine atoms in CaCl2 exhibit strong chemical reactivity (active [Cl])35,36) at high temperature. The Cu2O is promptly chlorinated by CaCl2 in flux, forming CuCl, which then diffuses into the gas phase and condenses on the inner wall of the volatiles collection device. As the increase of CaCl2 content in Fe2O3–CaCl2–SiO2 flux, the chlorination reaction of copper is progressively enhanced, leading to a higher copper removal efficiency. The gas-phase upwelling process further enhances the fluidity of liquid steel, promoting more complete oxidation and chlorination reactions. Additionally, silicates, known for their stability, are formed through the reaction of added silica with CaO and FeO in molten slag. The overall copper removal efficiency is likely a combination of multiple processes, and the method offers a practical and effective solution for controlling copper impurities in liquid steel.
In this study, an innovative separation strategy was developed for efficient copper removal from liquid steel through selective chlorination using calcium chloride (CaCl2). The methodology capitalizes on the volatile characteristics of metal chlorides under elevated temperatures to achieve effective copper elimination. The conclusions can be summarized as follows:
(1) Thermodynamic analysis revealed that copper chlorination becomes thermodynamically practicable at 1873 K, with SiO2 demonstrating significant catalytic potential in enhancing the CaCl2-mediated chlorination process at high temperatures.
(2) Systematic evaluation of chlorination parameters identified Fe2O3 and CaCl2 concentrations in the Fe2O3–CaCl2–SiO2 flux system as critical factors governing copper removal efficiency. Optimized flux composition parameters were established as: n(Fe2O3)/n(Cu) = 10.0, n(CaCl2)/n(Cu) = 1.0, and n(SiO2)/n(CaCl2) = 1.0. Under operational conditions of 1873 K, 50 min retention time, and 100 mL·min−1 inert gas flow rate, copper removal ratio reaches 39.68%.
(3) The proposed selective chlorination technique demonstrates substantial potential for copper impurity reduction in steelmaking processes. The irreversible volatilization of copper chlorides under protective atmosphere, coupled with the method’s cost-effectiveness, operational simplicity, and high efficiency, positions this approach as a promising solution for sustainable utilization of secondary metallurgical resources.
On behalf of all authors, the corresponding author states that there is no conflict of interest.
This work was supported by the National Natural Science Foundation of China (52293392) and University of Science and Technology Beijing State Key Laboratory of Advanced Metallurgy (41624009).
