Electrosynthesis of Titanium Alloys from Spent SCR Catalysts

Abstract

Spent selective catalytic reduction (SCR) catalysts are hazardous wastes containing many valuable metals whose improper disposal can cause environmental pollution and resource waste. Therefore, it is significant to recover valuable metals from spent SCR catalysts. In this study, the molten salt electrolytic method was employed to treat the SCR catalyst to direct electrosynthesis titanium alloys, which is more environmentally friendly and economical to obtain metal or alloy from secondary resources. A systematic investigation was carried out via experimental analysis and thermodynamic calculation. The results show that high-temperature pretreatment induces the aggregation of W and the formation of CaWO₄. Through molten salt electrolysis, titanium alloys containing Ti(W) and Ti₅Si₃ were formed, with a metal recovery rate of 80–87%. The electrolytic process and the reaction mechanism were also investigated. It is suggested that the molten salt electrolytic method is an effective way to recover valuable metals from spent SCR catalysts.

1. Introduction

A large amount of NO_x is emitted from thermal power plants and steel works, which leads to generating acid rain, the greenhouse effect, and photochemical smog. To reduce NO_x emissions, selective catalytic reduction, as an effective technology, has been commonly employed in industry. Generally, the SCR catalyst of V₂O₅-WO₃/TiO₂ plays a key role in affecting the reduction [1]. However, the complex flue gas environment would cause SCR catalyst deactivation, such as catalyst blockage, catalyst wear, high-temperature sintering, and chemical poisoning [2]. The service life of the V₂O₅-WO₃/TiO₂ catalyst is usually three to five years. According to the statistics, 250,000–300,000 m³ of spent SCR catalysts are generated annually in China [3,4]. Spent V₂O₅-WO₃/TiO₂ catalysts are hazardous solid waste. If discarded randomly, it not only leads to serious environmental problems but also brings resource waste. Therefore, it is significant to recycle spent V₂O₅-WO₃/TiO₂ catalysts.

Spent V₂O₅-WO₃/TiO₂ catalysts contain many valuable metals, such as 80–85% TiO₂, 5–7% WO₃, and 0.7–1.5% V₂O₅, which can be a second source to recover [5,6,7]. So far, many recycling methods have been proposed for spent SCR catalysts, mainly divided into the hydrometallurgy method and the pyrometallurgy method. Hydrometallurgy methods have been commonly used to recycle valuable metals from spent SCR catalysts, including acid leaching, alkaline leaching, sodium roasting-leaching, and calcification roasting-leaching [3,7,8,9,10,11,12,13]. For acid leaching, inorganic acids and organic acid reagents were used to dispose of spent SCR catalysts, which makes the valuable metals transformed into acidic solutions. Wang et al. [9] investigated the acid system of H₂SO₄-ascorbic to recover V₂O₅ from spent SCR catalysts. The addition of ascorbic acid can raise the leaching rate of the vanadium in H₂SO₄. The effects of leaching temperature, acid concentration, liquid-solid (L/S) ratio, and leaching time were investigated. Alkaline leaching is usually employed as an alkali reagent to leach V and W, which has several advantages, such as higher leaching efficiency and lower costs. Liu et al. [8] utilized alkali pressure leaching to extract V and W from spent SCR catalysts. The leaching efficiency of V and W can reach 98.83% and 100% by adjusting reaction time, leaching temperature, NaOH concentration, and stirring speed. Also, anatase TiO₂ can be reconstructed and recovered by the NaOH hydrothermal method [3]. The sodium roasting-leaching and calcification roasting-leaching methods mainly extract target metals by roasting the waste with Na₂CO₃, NaOH, CaO, or CaCO₃ at high temperatures to be more effectively leaching. The effect of Na₂CO₃ addition, roasting time, and roasting temperature on the extraction efficiency of W and V was explored by Choi et al. [14]. It is shown that the addition of Na₂CO₃ plays a vital role. The increasing dosage of Na₂CO₃ can significantly increase the extraction efficiency of W. Yao et al. [15] employed a simple method of successive calcination-H₂C₂O₄ leaching to recover the valuable metal W from spent V₂O₅-WO₃/TiO₂ catalysts. The leaching efficiency reaches 87%, and the purity of WO₃ is 91%. Furthermore, the separation and purification of V and W from the leaching solution of spent V₂O₅-WO₃/TiO₂ catalysts are treated through chemical precipitation and ion exchange methods. Xiao et al. [16] utilized the primary amine N1923 to separate W and V from spent SCR catalyst solutions by anion exchange mechanism. Although the hydrometallurgy method is a common way to treat SCR catalysts, some issues still exist, such as the long process flow and the great environmental risks.

Pyrometallurgy is another method for the recovery of valuable metals from SCR catalysts, which is usually by adding reducing agents such as carbon, Mg, H₂, and Al, and realized by thermal reduction [17,18,19]. Wang et al. [17] proposed an effective way to prepare V- and W-containing TiAl alloys by adding Al in spent SCR catalysts to reduce at 1600 °C. The influence factors of the reaction time, slag composition, and amount of reductant added to the recovery efficiency were explored. The maximum recovery efficiency is 98.14%, 94.16%, and 81.85% for V, Ti, and W, respectively. Li et al. [18,20] prepared Si–Ti alloy by reducing spent SCR catalyst with diamond-wire sawing silicon waste at 1500 °C. The pyrometallurgy method is indeed an environmentally friendly technology, but it has high energy consumption. So far, there is still no appropriate method to meet the demand. The new recycling method should be explored to recycle spent SCR catalysts.

Molten salt electrolysis is an effective and environmentally friendly method to treat solid waste. Compared with other methods, molten salt electrolysis was generally performed at low temperatures with low energy consumption, and no significant pollution was caused in the short process. The hydrometallurgy method generally has a long process flow and generates waste liquids that result in environmental problems, such as waste acid and alkali waste liquor. So far, many studies have been devoted to recycling valuable metals from solid waste using molten salt electrolysis. Lu et al. [21] present a solid-state electrolysis method in molten salts to effectively recycle aluminum scrap with low energy consumption. The electrochemical behavior of titanium-bearing slag in its molten state was investigated, and the Ir-Ti-Si alloy could be achieved by electrolysis [22]. Zhou et al. [23] treated high titanium slag and C through the molten salt electrolytic method to prepare TiC/SiC. Meanwhile, Ti₅Si₃, Ti₅Si₃/Ti₃SiC₂, and Ti₅Si₃/TiC can also be electro-synthesized from the Ti-bearing blast furnace slag in molten CaCl₂ [24]. It is evident that molten salt electrolysis is an effective method to recycle titania-containing waste. Herein, molten salt electrolysis was employed to treat spent SCR catalyst to recover valuable metals. The present paper successfully uses molten salt electrolysis to prepare Ti(W) and Ti₅Si₃ alloys directly from spent SCR catalyst in molten CaCl₂-NaCl electrolyte. The effects of high-temperature pretreatment and electrolyzing time were investigated. The electrolytic products at different conditions were characterized. Meanwhile, the electroreduction process and reaction mechanism were discussed based on the theoretical analysis and the experimental results. It is obvious that molten salt electrolysis provides a new direction for recycling spent SCR catalysts.

2. Experimental

2.1. Raw Materials and Characterization

The raw materials of spent commercial honeycomb SCR catalysts were collected from a thermal power plant in Hebei province, China. The spent SCR catalyst was first washed to remove the fly ash, then mashed by ball milling to prepare subsequent experiments. CaCl₂ and NaCl with analytical grades were provided by the Beijing Chemical Industry.

The chemical compositions were characterized by an X-ray fluorescence spectrometer (XRF). The phase compositions were investigated by X-ray diffraction (XRD). The scanning electron microscope with X-ray energy dispersive spectroscopy (EDS) was performed to investigate the morphology and surface element distribution of samples. The experimental voltage was set to 3.2 V, and the current was tested by a HP34401A digital multimeter.

2.2. Experimental Methods

The spent SCR catalyst was pretreated at 1100 °C for 2 h. Then, the sintered sample was wrapped with iron mesh and bundled with stainless steel rods to serve as the cathode. The graphite electrode was employed as the anode during electrolysis. In addition, anhydrous CaCl₂ and NaCl in a molar ratio of 0.52:0.48 was used as the molten salt electrolyte, and the electrolysis cell was an alumina crucible. The schematic diagram of the electrolytic device is shown in Figure 1. Before electrolysis, the furnace temperature reached 500 °C under a high-purity argon atmosphere to remove any water. Then, the electrolysis experiment was performed at 800 °C with 3.2 V. Subsequently, the electrolytic products were washed with distilled water to remove any residual electrolyte. Finally, the samples were dried at 120 °C for 2 h for subsequent tests. The electrolyzing conditions of 800 °C and 3.2 V were selected according to the previous literature: high temperature and high cell voltage can enhance the electroreduction-driven forces [25].

3. Results and Discussion

3.1. Characterization of the Spent SCR Catalyst

The chemical compositions of the raw materials were characterized by XRF, as shown in

Table 1. Chemical composition of spent SCR catalysts analyzed by XRF (wt%).

Component	TiO2	WO3	SiO2	CaO	V2O5	SO3	F	Others
Mass fraction	85.4	5.1	4.4	1.4	0.9	0.6	0.6	1.6

. Table 1 presents that spent SCR catalyst mainly contains TiO₂, WO₃, SiO₂, and CaO with higher weight ratios of 85.4%, 5.1%, 4.4%, and 1.4%, respectively. Further, the weight ratio of V₂O₅ is less than 1% with 0.9%, typical of an SCR catalyst. It also indicates that there are many valuable metals to recover. The morphology, elemental distribution, and phase composition were analyzed by SEM and XRD, as exhibited in Figure 2. From Figure 2a, it is obvious that the spent SCR catalyst powder presents flocculent with small particle size. Figure 2b–f show the elemental distribution of the spent SCR catalyst. The elemental mapping results reveal that the elements of Ti, O, and W display homogeneous distribution, while Ca and Si elements overlap, suggesting some Ca-Si-O compounds are generated. The EDS analysis result is shown in Figure 2g. It also can be seen that the main composition elements of the spent SCR catalyst are Ti, W, Si, and Ca, which agrees well with the XRF results. The main phase composition of the spent SCR catalyst is anatase-type TiO₂, and there is no other obvious phase observed, as shown in Figure 2h, which may be owing to the low content and the dispersed distribution of W-O and Ca-Si-O compounds.

3.2. Effects of High-Temperature Pretreatment

High-temperature pretreatment is an effective way to remove some impurities and improve electron transfer ability, which is beneficial for subsequent electrolysis.

Table 2. Chemical composition of spent SCR catalyst after high-temperature treatment (wt%).

Component	TiO2	WO3	SiO2	CaO	V2O5	SO3	F	others
Mass fraction	88.2	5.3	3.5	1.0	0.9	0.0	0.0	1.1

shows the chemical composition of the spent SCR catalyst after high-temperature treatment for 2 h at 1100 °C, characterized by XRF. Compared with the raw materials of the spent SCR catalyst, it is evident that F and SO₃ nearly disappeared after high-temperature treatment, indicating some impurities are indeed removed and reduce the effect on the subsequent electrolysis. The phase composition was also investigated by XRD, as exhibited in Figure 3. It is obvious that the main phase transforms from anatase-type TiO₂ to rutile-type, and CaWO₄ and SiO₂ appear.

The morphology and the elemental distribution of the spent SCR catalyst after high-temperature treatment were investigated by SEM and EDS, as shown in Figure 4. After high-temperature treatment, the agglomeration and growth of particles are displayed, and the porous structure is increased, as exhibited in Figure 4a. From Figure 4b–f, it is obvious that Ti and O still distribute homogeneously, while W presents an aggregation with an apparent change, which overlaps with the elemental distribution of Ca. This agrees well with the XRD result of CaWO₄ appearance. According to the enlarged SEM image and EDS spot analysis, it can be seen that TiO₂ acts as matrix material with large and smooth morphology, where small particles of Ca(W, Ti)O₄ are attached, which is consistent with the experimental results of Bhagat. et al. [26].

To clarify the reaction mechanism of the spent SCR catalysts in high-temperature pretreatment, the thermodynamic data was calculated, as exhibited in Figure 5. The standard Gibbs free energies (ΔG^θ) of possible reactions of the spent SCR catalyst as a function of temperature are presented. A negative value of ΔG^θ indicates the corresponding reaction may occur. The lower the value of ΔG^θ is, the more energetically advantageous for the reaction to occur. All values of ΔG^θ present negative above 600 °C, indicating CaTiO₃, CaSiO₃, CaV₂O₆, and CaWO₄ may be formed at high temperatures. Among them, the ΔG^θ of CaWO₄ shows the lowest value at 1100 °C, indicating CaWO₄ is first formed in theory. Meanwhile, considering the amount of CaO in the spent SCR catalyst, CaO mainly reacts with WO₃ in high-temperature pretreatment, and no residual CaO reacts with SiO₂ and TiO₂. Therefore, the main phase compositions are TiO₂, SiO₂, and CaWO₄ after high-temperature treatment. This is consistent with the XRD results, in which the phase compositions are TiO₂, SiO₂, and CaWO₄.

3.3. Electrolytic Products

The electrolytic product was obtained by the spent SCR catalyst electrolysis at 800 °C, 3.2 V, for 12 h. The phase compositions were analyzed by XRD, as presented in Figure 6. α-Ti and Ti₅Si₃ were formed after electrolysis, indicating the spent SCR catalyst totally transformed to titanium alloys via the electro-deoxidization method in molten salt of CaCl₂-NaCl. Figure 7 exhibits the SEM image and element mapping images of the electrolytic product. The electrolytic product presents a network structure owing to the localized melting of the alloy, as shown in Figure 7a. From Figure 7b–e, the elements of Ti, W, and V have a uniform distribution, while Si shows aggregation. This reveals that W and V partially substitute Ti to form W- and V-containing α-Ti alloy, and the Ti₅Si₃ alloy is generated, which conforms to the XRD results. Meanwhile, the metal recovery rate was calculated, which is in the range of 80–87%. The electrolytic products of Ti(W) and Ti₅Si₃ can be used as raw materials to prepare titanium alloys. Meanwhile, the particle size of electrolytic products can be controlled by optimum electrolyzing conditions, which has a potential application in 3D printing. Furthermore, the electrolytic product can be carbonized to prepare high-temperature structural material or be used as reinforcement to improve the mechanical properties of metal matrix composites.

3.4. Electroreduction Process

To investigate the process of electrochemical reduction, the electrochemical reaction in different stages was explored, as shown in Figure 8. Figure 8 exhibits the time-current plot of spent SCR catalyst electrolysis. The current decreases rapidly within 10 min and changes slightly in the time range of 10–40 min. After 40 min, the current decreases slowly. According to the changing trend of the curve, it can be divided into five stages. The electrochemical reduction at different stages was investigated by choosing different electrolytic durations of 10 min, 0.5 h, 1 h, 2 h, 4 h, 8 h, and 12 h, respectively.

Figure 9 presents the XRD patterns of electrolytic products of spent SCR catalysts with different durations, which can be employed to analyze the electrochemical reduction. After the spent SCR catalysts are immersed in molten salt at 800 °C for 10 min, the main phase remains rutile-type TiO₂, as presented in Figure 9a. The secondary phases of CaWO₄ and SiO₂ show a smaller peak compared with raw materials. This demonstrates that the molten salt has little effect on the spent catalyst without electrolysis.

At the initial stage of electrolysis, within 10 min, the main phase composition of the sample is CaTiO₃, and some minor phases appear, such as Ti₂O₃, TiO, W, and Ti₅Si₃, as shown in Figure 9b. To uncover the reaction mechanism, the theoretical decomposition voltage (∆E^θ) of the possible electrochemical reduction was calculated, as depicted in Figure 10. A larger value of ∆E^θ indicates the corresponding reaction is easier to occur. According to the value of the theoretical decomposition voltage, the order of theoretical electrolytic reduction is WO₃, CaWO₄, SiO₂, CaSiO₃, TiO₂, CaTiO₃, and CaO at 800 °C. it is obvious that CaWO₄, CaSiO₃, and CaTiO₃ require much more negative potentials compared with WO₃, SiO₂, and TiO₂, respectively. This demonstrates the formation of complex oxides generates the barriers for the electroreduction. Combined with the XRD results of Figure 9a, the electroreduction sequence is CaWO₄, SiO₂, and TiO₂. First, CaWO₄ was completely reduced to convert into W metal in the electrolytic process, which corresponds to the appearance of W(Ti) in Figure 9b. Then, SiO₂ is completely electrochemically reduced to form Si. Moreover, there are three ways for TiO₂ at this initial stage. Most TiO₂ chemically reacts with O²⁻ and Ca²⁺ in the molten salt and converts to CaTiO₃, which presents a block with a layered stacking structure, as shown in Figure 11a. The oxide ions mainly come from the rapid electroreduction of WO₃ and SiO₂. Some TiO₂ is not completely electrochemically reduced and converted to Ti₂O₃ and TiO, while some other TiO₂ is completely electroreduced to Ti but subsequently reacts with Si to form Ti₅Si₃ alloy, which is in agreement with the results of Li. et al. [24]. According to the phase composition in Figure 9b, it can be seen that TiO₂ is the most difficult to electroreduce, which is consistent well with the theoretical analysis.

Figure 9c shows the phase compositions at the second stage after 0.5 h of electrochemical reduction. With electrolysis, Ti₂O comes to emerge, while TiO and Ti₂O₃ disappear. Also, the relative content of CaTiO₃ presents a decrease. So, there are two main transformation behaviors that occur: TiO and Ti₂O₃ are totally converted to Ti₂O with lower valence, and a part of CaTiO₃ starts to be electrolyzed and generate Ti₂O in one step.

In the electrolytic reduction time range of 0.5 h to 4 h, the main reaction is the electrochemical reduction of CaTiO₃. As shown in Figure 9d–f, the XRD peak intensity of CaTiO₃ presents an obvious decrease, while Ti₂O has an increase. After electrolysis for 4 h, as shown in Figure 9f, there is almost no CaTiO₃ detected. This suggests that CaTiO₃ is electrochemically reduced to form Ti₂O. The micro-morphology image of the electrolytic product is exhibited in Figure 11b. Compared with the sample of the first stage, it is obvious that the particle size of CaTiO₃ appears to decrease and has a rough surface. This is also verified by the electrochemical reduction of CaTiO₃. The phase transitions are in agreement with the previous study [25]. Ti₂O can be obtained from CaTiO₃ directly, and there are no observed intermediates of Ti₂O₃ and TiO. However, the intermediates of CaTiO₃, Ti₂O₃, TiO, and Ti₃O₅ were detected in the electrolytic process of TiO₂. This result is also corresponding to the thermodynamic calculation. CaTiO₃ has more negative potentials, which may be lower than the intermediates of TiO and Ti₂O₃.

As shown in Figure 9g, the main phase composition becomes α-Ti within 8 h of electrochemical reduction at the V stage. CaTiO₃ and low-valence titanium oxides have disappeared. This demonstrates that CaTiO₃ and low-valence titanium oxides have been converted to α-Ti by electrolysis. The morphology and elemental distribution of the electrolytic product were also investigated, as shown in Figure 12. From Figure 12a, it can be seen that the electrolytic product presents a sponge-like agglomerated structure, and lots of fine particles are attached to the surface. Figure 12b–f show the elemental distribution. The element of Ti exhibits a homogenous distribution, while a small amount of Ca and O are still observed, suggesting that the electrochemical reduction is not complete, although there are no phase compositions of CaTiO₃ and low-valence titanium oxides observed in XRD patterns in Figure 9g. Also, the elemental of W and Si distributes unevenly, which is closely related to the distribution of fine particles. The more tiny particles are located, the more elements of W and Si are distributed. In order to verify this result, a local enlarged zone was investigated, as shown in Figure 12g–f. It is obvious that the large particle with a smooth surface is α-Ti, and the fine particles mainly contain W(Ti) solid solution and Ti₅Si₃ alloy.

The final electrolytic product was obtained by 12 h of electrochemical reduction. The main phase compositions are Ti₅Si₃ and α-Ti, as shown in Figure 9h. There is no W(Ti) phase observed. Meanwhile, it can be seen that the number of fine particles has a significant decrease from the view of Figure 7. This indicates that W(Ti) particles attached to the α-Ti surface have totally diffused into the α-Ti and formed an α-Ti(W) solid solution. This result is consistent with the experimental conclusion of Bhagat et al., which is that a small amount of W(Ti) could be dissolved into α-Ti at this temperature [26]. To summarize, the main changes at the V stage are the electrochemical reduction of Ti₂O and the dissolution of W into Ti.

Combined with the time-current curve and phase composition analysis, the electrochemical reduction process is illustrated in Figure 13. At stage I, a part of TiO₂, Ca(W, Ti)O₄, and SiO₂ occur in electrochemical reduction. Among them, Ca(W, Ti)O₄ is reduced to form a W(Ti) solid solution, and a part of TiO₂ and SiO₂ are converted to Ti₅Si₃. Also, some TiO₂ is reduced to lower valence. Meanwhile, the free O²⁻ ions, Ca²⁺, and TiO₂ have chemical reactions together to generate CaTiO₃. At stage II, the main transformation is that Ti₂O₃ and TiO are reduced to form Ti₂O. At stages III and IV, CaTiO₃ is electrochemically reduced to Ti₂O. This reduction process takes a relatively long time. The reason for this is the formation of the CaTiO₃ increases the difficulty of deoxygenation. This result is consistent with the conclusion of Hu et al. [27]. At stage V, the Ti₂O is further reduced to α-Ti, and W(Ti) is dissolved into α-Ti to form a homogeneous α-Ti(W) solid solution. As an overview of the electrochemical reduction process, the reduction sequence of TiO₂ was: TiO₂ → Ti₂O₃ → TiO → Ti₂O → Ti, and the reduction sequence of CaTiO₃ was: CaTiO₃ → Ti₂O → Ti. These results agree well with the conclusion of Hu et al. [27].

4. Conclusions

In summary, the titanium alloys of Ti(W) and Ti₅Si₃ were successfully electrosynthesized from the spent SCR catalyst by using the molten salt electrolysis method at 800 °C and 3.2 V within 12 h in molten CaCl₂-NaCl. W presents a uniform distribution, suggesting the solid solution of Ti(W) was formed. The metal recovery rate is in the range of 80–87%. In the electrolytic reduction process, W and Si were first electrolyzed to generate W (Ti) and Ti₅Si₃ with low theoretical decomposition voltage. CaTiO₃ was rapidly formed as an intermediate product in the electrolytic reduction of TiO₂. The deoxidation of CaTiO₃ and titanium suboxides are the main limiting steps. Titanium alloys containing Ti(W) and Ti₅Si₃ were formed as final electrolytic products. This result is consistent with the thermal dynamical calculation, in which the electroreduced subsequence is CaWO₄, SiO₂, TiO₂, and CaTiO₃. The present study paves the way for the comprehensive utilization of complex solid waste. Using the molten salt electrolytic method to treat the SCR catalyst provides a new direction for the disposal of spent SCR catalyst, in which the electrolytic product of titanium alloys shows a high additional value. In the next work, one objective is to investigate the effects of molten salt compositions or additives on the process efficiency.