Materials Chemistry
Volatile Separation and Recovery of Iridium from Oxygen Evolution Electrodes Using Calcium Oxide
2024 Volume 65 Issue 1 Pages 71-75
Details
2024 Volume 65 Issue 1 Pages 71-75
Iridium is a platinum-group metal with unique catalytic properties and chemical stability. Because of these characteristics, the metal is used in the form of iridium–tantalum oxides in the catalytic layer of oxygen evolution electrodes. Recovering Ir from end-of-life products is important because of its low production volume, the uneven geographical distribution of Ir sources, and high supply risks. However, Ir recovery requires its dissolution in aqueous solution using a strong acid, making the procedure not only dangerous but also hazardous to the environment. Moreover, if metals other than Ir are dissolved in the aqueous solution during its recovery, harmful effluents and gases could be generated. Separating Ir from such other metals would be difficult. Accordingly, we developed a method to extract Ir only from the catalyst layer and, simultaneously, recover Ir as a Ca–Ir composite oxide that is soluble in hydrochloric acid. Only iridium oxide was volatilized from the catalyst layer of the oxygen evolution electrodes and brought into contact with CaO in the gas phase. The composite oxide obtained was dissolved in hydrochloric acid and subsequently analyzed. The result revealed that Ir is highly soluble in hydrochloric acid, and the composite oxide does not contain Ta.
This Paper was Originally Published in Japanese in J. Japan Inst. Met. Mater. 87 (2023) 243–248.
Fig. 1 Schematic diagram of Ir separation and recovery in this experiment.
Iridium (Ir) is a platinum group metal (PGM) possessing heat and corrosion resistance and unique catalytic properties. This metal is an important oxygen evolution electrodes in industrial electrolysis processes. The electrode is coated with a catalyst layer comprising IrO2–Ta2O5 on a Ti substrate, and exhibits the highest catalytic activity when the Ir:Ta molar ratio is 7:3, resulting in a lower oxygen evolution overvoltage. The addition of Ta2O5 can improve durability against oxygen evolution, although it does not exhibit catalytic activity for oxygen evolution.1–3) Oxygen evolution electrodes using Ir contribute to reduced power consumption in industrial electrolysis processes because of their low oxygen evolution overpotential, and their high durability enables long-term use. In recent years, environmental and energy issues have become crucial concerns; consequently, the application of oxygen evolution electrodes is expected to expand even further. However, as Ir is a byproduct of platinum refining, only a few tons are produced annually, i.e., the production volume is quite low. The metal is scarce and production could hardly be increased. Moreover, resources are concentrated in specific regions such as South Africa, raising concerns about the stability of supplies. Recycling of used products is, therefore, critical. In the separation process for Ir recycling, the metal is dissolved and recovered through a wet process. However, because Ir and IrO2 are chemically stable, their dissolution necessitates using strong acid-containing oxidants, posing the problems of danger and detrimental environmental impact. When Ir is dissolved, other metallic elements could dissolve as well and could mix with the leaching solution. Hazardous effluents and gases could be generated, and the Ir recovery rate and purity during separation and refining could be reduced.4,5) Therefore, acid solubility must be improved while separating and recovering only Ir prior to solution.
In this study, we propose a method for recovering Ir in an easily soluble form in hydrochloric acid. The method comprises oxidizing and volatilizing IrO2 only from the catalyst layer of an oxygen evolution electrodes, separating it from Ta2O5, and converting the volatilized Ir oxide into a composite oxide with calcium oxide (CaO). Oxidizing and vaporizing Ir occur at high temperatures, which characteristic we considered could be utilized to ensure that only Ir is recovered. The recovery process comprises oxidizing and volatilizing IrO2 from the oxygen evolution electrodes while separating it from the Ta2O5 mixed in the electrode. Volatilized Ir oxide gas is absorbed by CaO and recovered as an Ir–Ca composite oxide. The acid solubility of Ir is improved by forming a composite oxide that is soluble in oxidant-free acids.6–9) Research has indicated7,9) that hydrochloric acid could be used for dissolving Ir–Ca composite oxides, which is expected to reduce hazards and environmental impacts.
Nomura et al. conducted research on recovering metallic PGMs by vaporizing them at high temperatures and absorbing them into perovskite-type oxides, such as LaScO3, CaMnO3, and (La0.7Sr0.2Ba0.1)ScO3−δ (δ is oxygen vacancy), through the gas phase.10–13) These authors also studied a PGM separation method using perovskite-type oxides.14,15) However, as such methods require adjustment of the perovskite-type oxide, challenges occurred such as mixing of multiple metals in the solution. If Ir could be recovered using a simple oxide as absorbent, there would be no need to synthesize the absorbent. Moreover, the separation operation would be easier, as, when dissolved, only Ir and one other element would be present in the solution.
Accordingly, we investigated the volatile separation of Ir from the catalyst layer of an oxygen evolution electrodes; its recovery as an Ir–Ca composite oxide using CaO, a simple oxide, as absorbent; and we determined the optimal process conditions.
Figure 1 shows a schematic of the volatile separation and recovery of Ir. The oxygen evolution electrodes and CaO were arranged in a non-contact manner and heated to high temperatures. Only the IrO2 in the catalytic layer of the electrode was oxidized and volatilized, subsequently absorbed into CaO and recovered as a complex oxide. The low cost of CaO and the ease of separating and purifying Ir from solution are favorable factors for choosing CaO as absorbent material.
Schematic diagram of Ir separation and recovery in this experiment.
Assuming that CaIrO3, Ca2IrO4, or Ca4IrO6, which reportedly are Ir–Ca composite oxides,16) were formed, the standard Gibbs free energy of formation (ΔG°) of these oxides was used to investigate whether the volatilized IrO3 could be recovered.16) When CaO is used as the absorbent for IrO3, composite oxides are expected to form according to the reactions derived in eqs. (1), (2), and (3).
| \begin{equation} \text{CaO(s)} + \text{IrO$_{3}$(g)} \to \text{CaIrO$_{3}$(s)} + \text{1/2O$_{2}$(g)} \end{equation} | (1) |
| \begin{equation} \text{2CaO(s)} + \text{IrO$_{3}$(g)} \to \text{Ca$_{2}$IrO$_{4}$(s)} + \text{1/2O$_{2}$(g)} \end{equation} | (2) |
| \begin{equation} \text{4CaO(s)} + \text{IrO$_{3}$(g)} \to \text{Ca$_{4}$IrO$_{6}$(s)} + \text{1/2O$_{2}$(g)} \end{equation} | (3) |
IrO2(g) and IrO3(g) are known gas species of Ir oxide; however, in this study, we assumed that the vapor pressure of IrO3 was higher and that Ir was oxidized and volatilized by the reactions shown in eqs. (4) and (5).
| \begin{equation} \text{Ir(s)} + \text{3/2O$_{2}$(g)} \to \text{IrO$_{3}$(g)} \end{equation} | (4) |
| \begin{equation} \text{IrO$_{2}$(s)} + \text{1/2O$_{2}$(g)} \to \text{IrO$_{3}$(g)} \end{equation} | (5) |
To examine which oxidation reaction eqs. (4) or (5) is dominant, the decomposition of IrO2 under atmospheric pressure was investigated assuming that the oxygen partial pressure was PO2 = 0.21 atm. The reaction calculation for the formation of IrO2 from pure Ir is shown in eq. (6).
| \begin{equation} \text{Ir(s)} + \text{1/2O$_{2}$(g)} \to \text{IrO$_{2}$(s)} \end{equation} | (6) |
The equilibrium constant for the reaction in eq. (6) is expressed in eq. (7).
| \begin{equation} \text{K} = \text{a}_{\text{IrO2}}/(\text{a}_{\text{IrO2}} \times (P_{\text{O2}}/P^{\circ})^{1/2}) \end{equation} | (7) |
where K is the equilibrium constant, a is activity, and P is partial pressure (atm).
Because Ir and IrO2 are individuals, K = 4.76 when a = 1. The standard Gibbs formula for energy of formation as a function of the equilibrium constant is expressed in eq. (8).
| \begin{equation} \Delta G^{\circ} = -RT \ln \text{K} \end{equation} | (8) |
Where ΔG° is standard Gibbs free energy of formation (J/mol), R is gas constant (J/mol/K) and T is Temperature (K).
According to the standard Gibbs energy of formation (ΔG°) reported16) for IrO2, the decomposition temperature of IrO2 was calculated as 1292 K using eq. (8), assuming an oxygen partial pressure of PO2 = 0.21 atm and atmospheric pressure. At temperatures above 1292 K, IrO2 decomposes into pure Ir, which is oxidized and volatilized by the reaction shown in eq. (4).
Figure 2 shows the vapor pressure of IrO3 in eqs. (4) and (5), which are the reaction equations for the oxidation and volatilization of Ir. The figure also shows the vapor pressure of IrO3 in eqs. (1), (2), and (3), which are the reaction equations for the formation of composite oxides under the assumption that the conditions are at atmospheric pressure (PO2 = 0.21 atm). Above 1292 K, Ir is believed to undergo oxidation volatilization by the reaction in eq. (4); the vapor pressure of IrO3 in equilibrium with CaIrO3 is lower than that in equilibrium with IrO2 up to 1442 K; the vapor pressure of IrO3 in equilibrium with Ca2IrO4 is lower up to 1475 K; and the vapor pressure of IrO3 in equilibrium with Ca4IrO6 is lower up to 1510 K. Evaporated IrO3 is estimated to be recoverable as a complex oxide up to a temperature of approximately 1510 K, using CaO as an absorbent.
Vapor pressure curves of IrO3 gas in equilibrium with the metal or its oxide and with the composite oxide at atmospheric pressure.
A simulated sample of the IrO2–Ta2O5 oxygen evolution electrodes catalyst layer was prepared based on an existing patent.17) The starting materials were 10 g hydrogen hexachloroiridate n-hydrate (H2IrCl·nH2O), 3.2 g tantalum pentachloride (TaCl5), 100 mL butanol (CH3(CH2)CH2OH), and 10 mL concentrated hydrochloric acid. The materials were mixed to prepare a precursor solution, which was applied to a dish and dried at 373 K until completely dry. After drying, the precursor solution was placed in an alumina crucible, heated at 723 K for 20 min, and crushed using an agate mortar and pestle.
To determine the Ir and Ta contents in the simulated sample, alkali fusion was performed on 0.1 g of the sample, using a method described by Yamamoto et al.18) The sample and sodium peroxide (Na2O2) were placed in a nickel crucible at a mass ratio of [simulated sample:Na2O2] = [1:50] and heated at 1073 K for 30 min. The fusion product was dissolved in aqua regia (200 mL). The Ir concentration in the solution was analyzed using inductively coupled plasma optical emission spectroscopy (ICP-OES), and the Ir content was calculated using eq. (9).
| \begin{equation} C^{\circ}{}_{\text{Ir}} = \rho_{\text{Ir}} \times V \times r \times 10^{-3} \end{equation} | (9) |
where, C°Ir is Ir content in simulated sample (g), ρIr is Ir concentration obtained by ICP-OES (mg/L), V is volume of flask (L) and r is dilution rate.
The alkali fusion result indicated that the amount of Ir in the simulated sample was 0.0543 g per 0.1 g of sample.
3.2 Separation and recovery using oxidized volatilization of IrWe conducted experiments to separate Ir from the prepared simulated sample (section 3.1) and to recover it as a composite oxide with CaO. In this experiment, CaCO3 was used and thermally decomposed to obtain CaO.
A schematic of the heating process is shown in Fig. 3. The simulated sample and CaCO3 (0.1 g and 1 g, respectively) were placed on two alumina plates (50 mm × 50 mm, 50 mm × 25 mm) in a non-contact state within an alumina box. The alumina box was covered with a lid, heated in an atmospheric environment in an electric furnace and maintained at a constant temperature.
Schematic diagram of inside the furnace during heating.
To investigate the dissolution characteristics of the Ir-containing complex oxide in hydrochloric acid, all samples from the CaCO3 side were placed in a beaker, and 20 mL hydrochloric acid was added. The mixture was heated on a hot plate at 353 K for 3 h.
After dissolution, the solution was filtered, and the volume was adjusted to 100 mL using pure water.
Subsequently, the solution was diluted appropriately and the Ir and Ta concentrations were measured using ICP-OES. Using the obtained Ir concentration, the dissolution rate for the Ir content in the simulated sample was calculated according to eq. (10).
| \begin{equation} S_{\text{Ir}} = (\rho_{\text{Ir}} \times V \times r \times 10^{-3}/C^{\circ}{}_{\text{Ir}}) \times 100 \end{equation} | (10) |
where, ρIr is Ir concentration in solution (mg/L) and SIr is Ir solubility (mass%).
The experiments in sections 3.2 were conducted for the volatile separation and recovery of Ir by heating and holding for 24 h at 1373 K,19) which is considered the most suitable temperature for the oxidation and volatilization of pure Ir.
The images before and after heating are shown in Fig. 4. After heating, the simulated sample turned white, whereas the CaO sample turned black. These results are ascribed to the IrO2 in the simulated sample oxidizing to IrO3 and evaporating, leaving only Ta2O5, which turned white. Meanwhile, the CaO absorbed the evaporated IrO3 and turned black.
Simulated (left) and CaCO3-side (right) samples before (a) and after (b) heating at 1373 K.
Figure 5 shows the X-ray diffraction analysis (XRD) profiles of the CaO-side samples after heating. A composite oxide of CaO, namely Ca2IrO4, and unknown peaks were identified. The results demonstrated that evaporated IrO3 could be recovered as a Ca–Ir composite oxide by absorption into CaO.
XRD profiles of CaCO3 side after heating.
Table 1 shows the Ir dissolution rate when the CaO-side sample obtained by the Ir separation experiment was dissolved in hydrochloric acid in the presence or absence of residue. The table also shows the Ir dissolution rate when the untreated simulated sample, which was not subjected to the separation and recovery experiment, was dissolved in hydrochloric acid. The untreated simulated sample exhibited a low Ir dissolution rate of 3.5%, which made it difficult to dissolve the Ir. However, when the separation experiment was conducted using CaCO3, the Ir dissolution rate increased significantly to 91.1 mass% compared with that of the untreated simulated sample. No residue was found in the dissolution test and Ta dissolution was not detected by ICP-OES analysis. Accordingly, these results demonstrated the effectiveness of volatilizing and separating Ir and recovering it as a composite oxide with CaO in the gas phase.
We amended the experimental conditions (temperature and time period) to determine the optimal conditions for the volatile separation of Ir and its recovery as Ca–Ir composite oxide.
Figure 6 shows the results of XRD analysis after 24 h of heating at 1073–1573 K, changing only the experimental temperature conditions. At 1273–1373 K, Ca2IrO4 formed, whereas Ca4IrO6 formed at 1423 and 1473 K. The formation reactions of these composite oxides are shown in eqs. (2) and (3). Unknown diffraction lines were observed between 1273 and 1473 K. These diffraction lines were strong at 1323 and 1373 K, where the dissolution rate was high, suggesting that they derived from a type of Ca–Ir composite oxide.
XRD profiles of heating temperatures (a) 1073 K, (b) 1173 K, (c) 1273 K, (d) 1323 K, (e) 1373 K, (f) 1423 K, (g) 1473 K, and (h) 1573 K at 24 h heating time.
Figure 7 shows the Ir dissolution rate in hydrochloric acid of the CaO-side samples heated for 24 hours at 1073–1573 K. The dissolution rate of Ir at 1073 K was 0.1 mass%, but it started to increase from 1173 K, reaching a maximum of 100% at 1323 K. However, this dissolution rate started to decrease after 1373 K to 2.2 mass% at 1573 K. The decreasing trend in the Ir dissolution rate was ascribed to the following factors:
Relationship between experimental temperature at 24 h heating and Ir solubility.
Above 1323 K, when the Ir dissolution rate started decreasing, IrO2 decomposed and evaporated through the reaction shown in eq. (4). Assuming that the experimental condition was atmospheric (PO2 = 0.21 atm), Fig. 8 shows the vapor pressure of IrO3 in the reaction shown in eq. (4), and the vapor pressure of IrO3 in eqs. (2) and (3), which are the formation reaction equations for Ca2IrO4 and Ca4IrO6. The vapor pressure difference between IrO3 and Ca4IrO6 decreased along with the increasing temperature but reversed at 1511 K. The decrease in the vapor pressure difference was considered the cause for the decrease in the rate of CaO absorption by IrO3, which led to decreases in the number of complex oxides formed and the Ir dissolution rate. In addition, complex oxides ceased to form because the vapor pressure of the complex oxides exceeded that of IrO3 owing to the reversal of the vapor pressure.
Vapor pressure curves of IrO3 in equilibrium with IrO3 and composite oxides.
The XRD profiles after heating for 24 h at 1073–1573 K (Fig. 6) confirmed the formation of Ca2IrO4 at 1273–1373 K. The formation of Ca4IrO6 was confirmed at 1423 K and 1473 K. As the decomposition temperature of Ca4IrO6 is reportedly 1510 K,16) the Ir dissolution rate is considered to have decreased at temperatures over 1323 K. The decrease was ascribed to the complex oxide ceasing to form as it was approaching the decomposition temperature. Figure 9 shows appearance that simulated and CaCO3-side samples before and after heating at 1573 K. The simulated samples turned white when heated to 1373 K (Fig. 4) but remained mostly black at 1573 K, although some turned white. However, IrO2 remained on the simulated sample side without volatilization. The decomposition temperature of the composite oxide was reached at 1573 K; therefore, the volatilized IrO3 was not absorbed by CaO and was saturated in the alumina box. The oxidative volatilization of IrO2 from the simulated sample was suppressed, and IrO2 remained on the simulated sample side after heating.
Simulated (left) and CaCO3-side (right) samples before (a) and after (b) heating at 1573 K.
Figure 10 shows the Ir dissolution rate at an experimental temperature of 1323 K. The highest Ir dissolution rate was achieved at this temperature, at an experimental time of 6–24 h. The Ir dissolution rate increased linearly for up to 18 h and subsequently increased gradually, reaching a maximum of 100 mass% after 24 h. The Ir dissolution rate increased linearly up to 18 h, as the IrO2 near the surface oxidized and volatilized, and the IrO2 deeper in the surface appeared, oxidized, and volatilized again. However, the simulated sample used in this study contained Ta2O5 and, as the oxidation and volatilization of IrO2 progressed, only Ta2O5 remained near the surface, but the oxidation of IrO2 was suppressed. Therefore, the dissolution rate of Ir increased slowly after 18 h.
Relationship between experimental time and Ir solubility at 1323 K.
Utilizing the ability of Ir to oxidize and volatilize at high temperatures facilitated IrO2 recovery from the catalyst layer of an oxygen evolution electrodes and adsorption onto CaO to form a Ca–Ir composite oxide. This composite oxide dissolved in hydrochloric acid and exhibited a high Ir dissolution rate. In addition, no dissolution of Ta was observed and only Ir could be separated and recovered. This method is useful in terms of efficiency, safety, and cost because only Ir can be recovered from the spent electrode. Moreover, the process does not require a strong acid and uses CaCO3. Further, Ir is recovered from the gas phase, and it is expected that it could be recovered from products other than oxygen evolution electrodes at the same time.
