Materials Chemistry
Distribution of Platinum between the SiO2–CaO–Al2O3–TiO2 Slag System and Molten Copper
2024 Volume 65 Issue 6 Pages 652-656
Details
2024 Volume 65 Issue 6 Pages 652-656
A residue containing TiO2 and PGMs is generated in the hydrometallurgical process used for recycling platinum-group metals (PGMs). In this study, a pyrometallurgical process was considered in which PGMs from the residue generated in the hydrometallurgical process were concentrated in a molten copper phase as a collector metal and TiO2 was separated into the SiO2–CaO–TiO2 slag phase with SiO2 and CaO flux. The dissolution of PGMs must be reduced to minimize the loss of PGMs to the slag. Therefore, the distribution ratios of Pt as representative PGMs between the liquid SiO2–CaO–Al2O3–TiO2 or liquid SiO2–CaO–TiO2 slag and molten copper were measured at 1773 K under an oxygen partial pressure of $p_{\text{O}_{2}} = 10^{ - 10}$. The experimental results showed that the distribution ratio of Pt increased with TiO2 concentration in the slag, and the distribution ratio of Pt reached a maximum value at a TiO2 concentration of approximately 10 mass%, and decreased with a further increase in TiO2 concentration with the SiO2–CaO–Al2O3–TiO2 slag. However, as TiO2 concentration in the slag increased, the distribution ratio of Pt decreased with the SiO2–CaO–TiO2 slag. Additionally, the experimental results showed that the distribution ratio of Pt between the SiO2–CaO–Al2O3–TiO2 slag and liquid copper increased with the slag basicity B, defined as B = (mass%CaO)/(mass%SiO2) when the TiO2 concentration in the slag was greater than 10 mass%.
Platinum-group metals (PGMs) are classified as precious metals. They are important elements in a wide range of fields owing to their unique catalytic properties, heat resistance, acid resistance, and chemical stability. However, supply sources of PGMs are unevenly distributed, with most existing reserves located in South Africa and Russia, and ensuring a stable supply remains a concern [1–5]. Although sufficient minable reserves do exist, the ores of these metals are present only in low concentrations. Furthermore, production of PGMs from ores requires a great deal of energy and has a wide environmental impact [1, 2].
Pt, a PGM, is used as a catalyst for automotive exhaust gas purification, crucibles for glass production, and electrodes for fuel cells. Secondary supplies of Pt have been reported as 26% of primary supplies, and almost all of it is recycled from automobiles [4]. In addition, large amounts of PGMs are recycled between several closed companies in a business alliance. These are not reported because the ownership of metal is retained by the industrial consumer, who returns PGM-bearing materials to specialist companies for refining and reuse in the same application [4].
Therefore, recycling PGMs is important, including for Pt itself. Currently, residues containing TiO2 and PGMs is generated from one of the hydrometallurgical process used to recycle PGMs. To selectively recover PGMs from the residuals, in this study, we considered an additional pyrometallurgical process of residue treatment in which PGMs were concentrated in the metal phase with liquid Cu as the collector and Ti was separated into a slag phase. The slag and metal were separated in a furnace; however, the chemical dissolution of Pt in the slag resulted in the loss of Pt. Therefore, to investigate the operating conditions of the pyrometallurgical process for Ti removal with minimal PGM loss, we examined the distribution of Pt, representing PGMs, between the slag and molten copper. A SiO2–CaO–TiO2 slag system was used owing to its low flux cost of SiO2 and CaO. Based on the thermodynamic data, $p_{\text{O}_{2}}$ was set to 10−10, which oxidized Ti but hardly oxidizes PGMs, where $p_{\text{O}_{2}}$ is a dimensionless number obtained by dividing the partial pressure of oxygen by the standard pressure of 1.01325 × 105 Pa.
Several studies have focused on the dissolution of Pt in slags. In a prior work, we showed the effect of the SiO2–CaO–Al2O3–CrOx slag composition on the dissolution behavior of Pt between the slag and Cu at 1773 K [6]. Nishizima and Yamaguchi investigated the distribution behavior of Pt between molten Cu and the Al2O3–CaO–SiO2–Cu2O slag at 1723 K [7]. Borisov and Danyushevsky measured the solubility of Pt in the SiO2–Al2O3–MgO–CaO slag at 1723 K [8]. Baba and Yamaguchi studied the solubility behavior of Pt between Pt or the Cu–Pt alloy and the FeOx–MgO–SiO2 slag at 1573 K [9]. Avarmaa et al. measured the distribution behavior between molten Cu and the Al2O3–CaO–FeOx–SiO2 slag at 1573 K [10]. Klemettinen et al. measured the distribution of platinum between the solid Pt–Fe alloy and the Al2O3–FeOx–SiO2 slag at 1573 K [11]. Wiraseranee et al. investigated the solubility of Pt in the Na2O–SiO2-based slags at 1473 K [12]. Nakamura et al. considered the solubility of Pt in BaO–Al2O3, BaO–SiO2, CaO–Al2O3, CaO–SiO2, Na2O–SiO2, CaO–Al2O3–SiO2, BaO–CuOx, BaO–MnOx, CaO–SiO2–FeOx, K2O–SiO2, and Na2O–P2O5 slags at 1373–1873 K [13, 14]. Ertel et al. determined the solubility of Pt in the SiO2–CaO–MgO–Al2O3 slag at 1573 K [15]. Dable et al. examined the solubility of Pt in the CaO–Al2O3–SiO2 slag at 1700 K [16]. Murata and Yamaguchi studied the distribution behavior of Pt between Cu–Cu2O- and Pb–PbO-based slags at 1523 K [17]. However, to the best of our knowledge, no investigations of the solubility of Pt in slag systems containing TiO2 have been reported in the literature. Therefore, in this work, we examined the distribution behavior of Pt in Al2O3 or SiO2 crucibles by equilibrating SiO2–CaO–Al2O3–TiO2 and SiO2–CaO–TiO2 slag systems at 1773 K with Pt-containing Cu alloys at $p_{\text{O}_{2}} = 10^{ - 10}$ by changing the composition of the slag.
Two types of crucibles composed of SiO2 (no purity statement) and Al2O3 (>99.6 mass% purity) were used for the measurements. Pt (0.2 g; >99.95 mass% purity) and Cu (1.8 g; >99.5 mass% purity) wires were added to the SiO2 or Al2O3 crucible. Additionally, 8 g of a mixture of SiO2 (>99.5 mass% purity), CaO (>99.9 mass% purity), and TiO2 (>99.99 mass% purity) was added to the crucible to cover the Pt and Cu metals. In the case of the Al2O3 crucible, excess Al2O3 powder (>99 mass% purity) was added to the sample to prevent melting of the crucible. Thereafter, a sample was inserted into a reaction tube, maintained at 1773 K (±3 K) in a furnace, and heated using SiC heating elements. Oxygen partial pressure was set to $p_{\text{O}_{2}} = 10^{ - 10}$ with a 140.9 mL/min CO and 9.1 mL/min CO2 gas mixture for 20 h. The mixing ratio of CO and CO2 is calculated from the reaction and ΔG0 reported in the thermodynamic database FactSage shown in eq. (1) [18].
| \begin{equation} \text{2CO(g)} + \text{O$_{2}$(g)} = \text{CO$_{2}$(g)}:\Delta G_{1773\,\text{K}}^{0} = -258{,}576\ (\text{J}) \end{equation} | (1) |
After equilibrium was reached, the samples were quenched with water. Following a previous study, the slag composition was determined by chemical analysis [6].
The slag and alloy phases in the crucible were separated carefully. Two types of dissolution methods including an alkali fusion method using Li2B4O7 and a method employing aqua regia were employed to achieve total dissolution of the sample for complete and accurate characterization using instrumental methods. Si, Ca, Ti, and Al in the slag were dissolved using the alkali fusion method with Li2B4O7 as the flux. Cu and Pt in the slag, which were present in low concentrations and could not be analyzed by alkali fusion, were dissolved in aqua regia. Cu, Pt, and Ti in the metal phase were dissolved in aqua regia. For details on the analytical method used in this study, please refer to our previous report [6]. Pt195 was monitored using inductively coupled plasma mass spectrometry (ICP–MS; Agilent Technologies, 7700x) as the mass of Pt. Additionally, Ti wavelengths (334.188 nm) were used to determine concentrations using inductively coupled plasma optical emission spectrometry (ICP–OES; Agilent Technologies, 5100). For the initial composition, the TiO2 concentration and basicity were varied based on the phase diagrams reported by DeVries et al. and Murata et al. [19, 20] as given in eq. (2).
| \begin{equation} B = \frac{(\text{mass%CaO})}{(\text{mass%SiO$_{2}$})} \end{equation} | (2) |
Table 1 lists the results of the chemical analysis of the slag and Cu–Pt alloy phases. The concentration of Ti in the molten copper was below the lower limit of quantification. Here, ( ) and [ ] represent the X values of the slag and alloy phases, respectively. The oxidation state of Ti in the slag was assumed to be 4 based on previous studies [21–24].
;%0A%09%09%09newWindow.document.open();%0A%09%09%09newWindow.document.write('<img src=%22./Graphics/MT-M2024002tab01.jpg%22>');%0A%09%09%09newWindow.document.close();%0A%09%09)
In the experiment using the SiO2 crucibles, SiO2 in the crucible dissolved into the slag, and the slag became saturated with SiO2. Thus, a SiO2 saturated liquid phase was obtained in the SiO2–CaO–TiO2 slag at 1773 K and $p_{\text{O}_{2}} = 10^{ - 10}$. The solubility of SiO2 in the SiO2–CaO–TiO2 slag is shown in Fig. 1. The liquid-phase line on the SiO2 saturated side of the SiO2–CaO–TiO2 ternary phase diagram presented in this study is consistent with that reported by DeVries et al. [19]. Compared with SiO2, in the experiment that used Al2O3 crucibles, Al2O3 in the crucible also dissolves in the slag, and the slag becomes saturated with Al2O3. Thus, Al2O3-saturated liquid phase was obtained in the SiO2–CaO–Al2O3–TiO2 slag at 1773 K and $p_{\text{O}_{2}} = 10^{ - 10}$. The solubility of Al2O3 in the SiO2–CaO–TiO2 slag as a function of B is shown in Fig. 2. The solubility of Al2O3 in the SiO2–CaO–Al2O3–TiO2 slag is shown in Fig. 3. The solubility of Al2O3 in the SiO2–CaO–TiO2 slag is 30.7–44.9 mass%. The solubility of Al2O3 increased with B. This trend is consistent with the fact that the liquid phase line of CaAl12O19, Al2O3 or CaAl2Si2O8 saturation in SiO2–CaO–Al2O3 slag approaches a high Al2O3 concentration with increasing basicity 0.62 ≤ B ≤ 1.20 [25].
Solubility of SiO2 in the SiO2–CaO–TiO2 system (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
Relationship between solubility of Al2O3 in the SiO2–CaO–TiO2 system and basicity of the slag (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
Solubility of Al2O3 in the SiO2–CaO–TiO2 system (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
In this study, the distribution ratios of Pt and Ti between the slag phase and liquid copper are defined by eq. (3).
| \begin{equation} L_{\text{X}} = \frac{(\text{mass%X in Slag})}{[\text{mass%X in Alloy}]},\quad \text{X: Pt or Ti} \end{equation} | (3) |
Figure 4 shows the relationship between the distribution ratio of Pt and the concentration of TiO2 in the slag. Figure 5 shows the relationship between the distribution ratio of Pt and basicity of the slag. The distribution ratio of Pt between the SiO2–CaO–Al2O3–TiO2 or SiO2–CaO–TiO2 slag and liquid copper was 3.6 × 10−7–7.9 × 10−6. Figure 4 shows that the distribution ratio of Pt increased with TiO2 concentration in the slag. The distribution ratio of Pt reached a maximum value at a TiO2 concentration of approximately 10 mass% and decreased with a further increase in TiO2 concentration in the SiO2–CaO–Al2O3–TiO2 slag. Similarly, as the TiO2 concentration in the slag increased, the distribution ratio of Pt decreased in the case of the SiO2–CaO–TiO2 slag. Figure 5 shows that the distribution ratio of Pt between the SiO2–CaO–Al2O3–TiO2 slag and liquid copper increased with the basicity of the slag, whereas the TiO2 concentration in the slag was more than 10 mass%, similar to reports of Nishizima et al. and Nakamura et al. [7, 13, 14]. However, the distribution ratio of Pt between the SiO2–CaO–Al2O3 ternary slag with a 0 mass% of TiO2 concentration and the molten copper decreased as B increased. When the concentration of TiO2 in the slag was less than 10 mass%, the increase in the distribution ratio of Pt between the SiO2–CaO–Al2O3–TiO2 slag and molten copper as a function of the TiO2 concentration in the slag was similar to that in the SiO2–CaO–Al2O3–TiO2 slag as a function of the Cr2O3 concentration.
Relationship between the distribution ratio of Pt and concentration of TiO2 in the slag (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
Relationship between the distribution ratio of Pt and basicity of the slag (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
Figure 6 shows the relationship between the concentrations of Cu and TiO2 in the slag. Figure 7 shows the relationship between the concentration of Cu and basicity of the slag. Figure 6 shows that the Cu concentration increased with TiO2 concentration in the slag. Figure 7 shows that the Cu concentration tends to decrease with increasing slag basicity B.
Relationship between the concentration of Cu in the slag and concentration of TiO2 in the slag (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
Relationship between the concentration of Cu in the slag and basicity of the slag (1773 K, $p_{\text{O}_{2}} = 10^{ - 10}$).
The relationship between the distribution ratios of Ti could not be calculated because the concentration of Ti in the molten copper was below the lower limit of quantification. Therefore, the concentration of Ti in the molten copper phase was assumed to be the lower limit of quantification, and the concentration of Ti in the molten copper phase was 25 ppb. Ti was not distributed in the molten copper but in the slag phase. These results, together with those in Figs. 4 and 5, indicate that Pt is concentrated in the molten copper and Ti is removed from the slag. Thus, Pt and TiO2 can be separated.
When the TiO2 concentration in the slag was high, the amounts of SiO2, CaO, and Al2O3 added were small. Therefore, large amounts of the residue could be processed in a single operation, which is advantageous. Furthermore, the distribution ratio of Pt decreased with increasing TiO2 concentration when the TiO2 concentration in the slag exceeded 10 mass%. Therefore, a high concentration of TiO2 in the slag is advantageous for both the distribution behavior of Pt and its efficient operation.
The distribution of Pt between the liquid SiO2–CaO–Al2O3–TiO2 or liquid SiO2–CaO–TiO2 slag systems and molten Cu was investigated. Pt and Ti were concentrated in the alloy and slag phases, respectively. The experimental results showed that the distribution ratio of Pt increased with TiO2 concentration in the slag, and the distribution ratio of Pt reached a maximum value at a TiO2 concentration of approximately 10 mass%, and decreased with a further increase in TiO2 concentration with the SiO2–CaO–Al2O3–TiO2 slag. However, as TiO2 concentration in the slag increased, the distribution ratio of Pt decreased with the SiO2–CaO–TiO2 slag. Additionally, the solubility of Pt increased with slag basicity when the TiO2 concentration in the slag exceeded 10 mass%. Therefore, a high concentration of TiO2 in the slag is advantageous for both the distribution behavior of Pt and its efficient operation.
We would like to thank Tanaka Kikinzoku Kogyo K.K. for their helpful suggestions and discussions, the Environmental Safety Center of Waseda University for the use of ICP-OES and ICP-MS.
