Highly efficient and selective removal of vanadium from tungstate solutions by microbubble floating-extraction

Abstract

Selective separation of dissolved tungsten and vanadium is of great significance for the utilization of the secondary resources of these elements. In this work, selective removal of vanadium from tungstate solutions via microbubble floating-extraction was systematically investigated. The results indicated that vanadium can be more easily mineralized over tungsten from tungstate solutions using methyl trioctyl ammonium chloride as mineralization reagent under weak alkaline conditions. Owing to the higher bubble and interface mass transfer rates, high-efficiency enrichment and deep separation of vanadium could be achieved easily. Additionally, the deep recovery of tungsten and vanadium from the floated organic phase could be easily realized using a mixed solution of sodium hydroxide and sodium chloride as stripping agents. The separation mechanism mainly included the formation of hydrophobic complexes, their attachment on the surface of rising bubbles, and their mass transfer at the oil–water interface. Under the optimal conditions, the removal efficiency of vanadium reached 98.5% with tungsten loss below 8% after two-stage microbubble floating-extraction. Therefore, the microbubble floating-extraction could be an efficient approach for separating selectively vanadium from tungstate solutions, exhibiting outstanding advantages of high separation efficiency and low consumption of organic solvents.

Key words： tungsten; vanadium; selective separation; reagent mineralization; microbubble floating-extraction

1 Introduction

As indispensable strategic metals, tungsten and vanadium have been extensively applied in various fields including catalyst, armament, atomic energy, and aerospace due to their remarkable physicochemical properties and excellent performance [1–3]. With the depletion of high-quality mineral resources, ensuring resource safety through the recovery of tungsten and vanadium from secondary resources is of great significance [4]. In particular, spent selective catalytic reduction (SCR) catalysts have attracted attention of numerous researchers due to its huge amount discarded and rich in tungsten and vanadium [5,6]. Owing to the similar chemical properties, tungsten usually coexists with vanadium in leaching solutions of spent SCR catalyst. However, limited works are focused on the effective separation of tungsten and vanadium from the leaching solutions, which will influence the grade of subsequent recycled products [7,8]. Therefore, the highly efficient and selective separation of tungsten and vanadium has become a technical challenge in the resource utilization of spent SCR catalyst.

The selective separation of tungsten and vanadium is mainly based on the differences in physicochemical properties between the elements, including the ability of tungsten and vanadium to polymerize and the ability of vanadium to form cations in acidic solutions [8]. Based on the difference, some technologies such as chemical precipitation [9,10], ion exchange [11,12], and solvent extraction [13,14] have been performed to separate tungsten and vanadium from different leaching solutions. However, the common problems for chemical precipitation and ion exchange are the low separation efficiency, the relatively high loss of tungsten, and the low-efficiency desorption, resulting in their limited application [15–17]. Solvent extraction is usually used to separate similar metals from dilute aqueous solutions [18]. Nevertheless, the main drawbacks of this technology include the dispersion of oil–water and high consumption of the organic phase, resulting in slower interphase mass transfer and higher operating costs [19–21]. In addition, the solution pH needs to be multiple adjusted in the process of cascade extraction, which leads to the discharge of a large amount of wastewater. There is also a contradiction between the theoretical yield and the number of extraction stages in low-concentration solutions [22]. Therefore, it is urgent to develop a high-efficiency method for selective separation of tungsten and vanadium.

Several attempts were made to solve these problems by developing bubbling organic liquid membrane extraction, operated using a bubbling extraction tower for liquid–liquid solvent extraction. This technology improves the mass transfer efficiency and enrichment ratio by dispersing the extractant on the surface of the bubbles to form an organic liquid membrane [23–25]. On the basis of previous studies, the microbubble floating-extraction technology was inspired by the advantages of ion flotation and solvent extraction technology [26–28]. In addition, our previous work demonstrated that this technology presented great application prospects in the field of separation of similar metals [29]. In general, microbubble floating-extraction mainly includes the formation of hydrophobic complexes via the specific binding of oppositely charged surfactants and dissolved ions (reagent mineralization), and the enrichment of hydrophobic complexes in the organic layer on the top of the aqueous phase under the action of rising microbubbles (floating-extraction) [30,31]. This technology could have significant technical advantages over solvent extraction. Fig.1 provided a simple comparison of solvent extraction and microbubble floating-extraction. As depicted in Fig.1(a), metal ions and surfactants can be fully contacted in a homogeneous aqueous solution during microbubble floating-extraction, which significantly reduces the consumption of surfactants. As displayed in Fig.1(b), a bubble mass transfer process was added during microbubble floating-extraction, the mass transfer coefficient and the mass transfer balance were greatly improved by the microbubbles, resulting in a much higher interface mass transfer rate than that of solvent extraction [29]. Furthermore, it has been indicated that flotation efficiency is slightly dependent on the volume of the organic phase [32], which can reduce significantly the consumption of the organic phase and achieve high enrichment efficiency. Therefore, microbubble floating-extraction is a promising and feasible method for the selective separation of tungsten and vanadium.

Fig.1 Comparison of solvent extraction and microbubble floating-extraction: (a) reaction process, and (b) mass transfer process (k₁ and k₂ are the mass transfer coefficients between the oil–water interface during solvent extraction; K_OW′ and K_OW are the separation equilibrium constants of microbubble floating-extraction and solvent extraction, respectively).

Full size|PPT slide

The objective of the present investigation was to selectively remove vanadium from tungstate solutions via theoretical analysis and microbubble floating-extraction. Based on the present study, we used methyl trioctyl ammonium chloride (Aliquat336) as a mineralization reagent to form the Aliquat336-V hydrophobic complexes by the specific binding of all species of vanadium with Aliquat336, and enriched them into the top organic phase by rising bubbles. In current work, the effects of mineralization and floating-extraction behaviors on the performance of microbubble floating-extraction process were systematically studied, and the separation mechanism was discussed and analyzed. Based on the present study, a new method for the highly efficient and selective separation of tungsten and vanadium was proposed.

2 Experimental

2.1 Materials

The mixed solutions of tungsten and vanadium were prepared from sodium tungstate dihydrate and sodium vanadate (both were purchased from Shanghai McLean Biochemical Technology Co., Ltd.). Based on the composition of the spent SCR catalysts (concentration ranges for leaching solutions were listed in Table S1, cf. Electronic Supplementary Material, ESM), the concentrations of tungsten and vanadium in the simulated aqueous solutions were selected to be 10 and 1.0 g∙L^‒1, respectively. Aliquat336 (Aladdin Chemical Reagent Co., Ltd., China) was used as a surfactant to mineralize the vanadate ions in the solutions, and the organic phase employed in the flotation consisted of 2-octanol (Aladdin Chemical Reagent Co., Ltd., China) and sulfonated kerosene. Sodium hydroxide (Shanghai Aladdin Biochemical Technology Co., Ltd. China), sodium chloride, sodium sulfate, sodium carbonate, and sodium bicarbonate were selected for the stripping experiments. Sodium hydroxide and hydrochloric acid (Sinopharm Chemical Reagent Co., Ltd., China) were employed to adjust pH. All other reagents adopted in the experiments were of analytical grade without further purification, and deionized water was used throughout the experiments.

2.2 Experimental procedures

Fig.2 displayed the schematic diagram of the experimental procedure for separating vanadium from tungstate solutions by microbubble floating-extraction. As depicted in Fig.2, the experimental procedure mainly included four stages. In stage 1, the mixed solutions of tungsten and vanadium with initial concentrations of 10 and 1.0 g∙L^‒1 were prepared. In stage 2, 50 mL of a homogeneous mixed solution was prepared in a conical flask, and its pH was adjusted by sodium hydroxide and hydrochloric acid. The samples were then mineralized to form Aliquat336-V hydrophobic complexes by adding a certain dosage of Aliquat336 and stirring for a predetermined time. The effects of pH, mineralization time, Aliquat336 dosage, and initial concentrations on mineralization efficiency were systematically investigated. In stage 3, the mineralized solution was transferred to the flotation column, a certain amount of organic phase was added to the top of the flotation column, and bubbles were injected at a certain rate. After the floating-extraction process, the organic phase and raffinate were separated for the next experiment and analysis, respectively. The effects of organic phase composition, flotation time, and flow rate on separation efficiency were systematically investigated. In stage 4, a certain volume of the organic phase was mixed with the stripping agent solution in a conical flask and then placed in a constant-temperature oscillator with a certain speed to shake for a predetermined time. It was then transferred to a separatory funnel to separate the phases. Unless otherwise mentioned, all experiments were conducted at room temperature.

Fig.2 Schematic diagram for separating vanadium from tungstate solutions by microbubble floating-extraction.

Full size|PPT slide

2.3 Analysis and characterizations

The pH was measured by pH meter (Shanghai Mettler Instruments Factory, China). The Fourier transform infrared spectroscopy (FTIR, Nicolet Nexus 670, Thermo Electron Corporation, America) was conducted to characterize the functional groups of Aliquat336 before and after mineralization. The tungsten and vanadium concentrations were determined by inductively coupled plasma-optical emission spectrometry (Thermo Scientific ICAP PRO X, America). The ion concentrations in the organic phase were calculated by the subtraction method based on the mass balance. The key indexes of the separation process were summarized in Tab.1 [29]:

Tab.1 Key indexes of the microbubble floating-extraction process

Index	Formula	Nomenclature
Mineralization efficiency	M=1-{\displaystyle \frac{C_{\mathrm{M}}}{C_{\mathrm{F}}}}\times 100\mathrm{\%}	C_{\mathrm{F}}, C_{\mathrm{M}}, C_{\mathrm{R}}, C_{\mathrm{S}}, and C_{\mathrm{F}\mathrm{L}} represent the concentrations of tungsten and vanadium in the feed, supernatant, raffinate, stripping solution, and floated organic phases; V_{\mathrm{F}\mathrm{L}} and V_{\mathrm{S}} are the volumes of the floated organic phase and stripping solution, respectively; DV and DW are the distribution coefficient of V and W, respectively.
Flotation efficiency	F=\left(1-{\displaystyle \frac{C_{\mathrm{R}}}{C_{\mathrm{F}}}}\right)\times 100\mathrm{\%}
Stripping efficiency	ST={\displaystyle \frac{C_{\mathrm{S}}\times V_{\mathrm{S}}}{C_{\mathrm{F}\mathrm{L}}\times V_{\mathrm{F}\mathrm{L}}}}\times 100\mathrm{\%}
Distribution coefficient	D={\displaystyle \frac{C_{\mathrm{F}\mathrm{L}}}{C_{\mathrm{R}}}}
Separation factor	{\beta }_{\mathrm{V},\mathrm{W}}={\displaystyle \frac{D_{\mathrm{V}}}{D_{\mathrm{W}}}}

In addition, mineralization efficiency refers to the efficiency of the formation of hydrophobic tungsten or vanadium complexes through reagent mineralization; flotation efficiency refers to the efficiency of tungsten or vanadium enrichment into the organic phase through rising bubbles.

2.4 Data acquisition and error analysis

All experiments were conducted in triplicate to ensure accuracy under identical conditions and the average values were reported. Statistical analysis including the analysis of variance was treated with Origin 9.0 software. Additionally, the simulation software of Visual MINTEQ and MATLAB were adopted to calculate the species distributions.

3 Results and discussion

3.1 Theoretical fundamentals for selective separation of vanadium

During the microbubble floating-extraction process, the separation behavior of tungsten and vanadium strongly depends on the species of tungsten and vanadium. Based on previous reports [8,13,14,33], the common ion species of tungsten and vanadium in aqueous solutions were investigated by Visual MINTEQ software [34]. The molar ratio of tungsten and vanadium species as a function of pH in the aqueous solution of single tungsten (W-H₂O) and aqueous solution of single vanadium (V-H₂O) systems were presented in Fig. S1 (cf. ESM). It was found that tungsten and vanadium mainly existed as WO₄^2‒, W₇O₂₄^6‒, HW₇O₂₄^5‒, HVO₄^2‒, HV₂O₇^3‒ and V₄O₁₂^4‒ species in the weak alkaline aqueous solution containing 10 g∙L^‒1 tungsten, and 1 g∙L^‒1 vanadium. On the basis of the study of tungsten and vanadium species in the W-H₂O and V-H₂O systems, the thermodynamic data of relevant species in the tungsten and vanadium mixed solution (W-V-H₂O) system were summarized in Tab.2 [8,35]. According to the principle of simultaneous equilibrium and conservation of matter, Eqs. (11) and (12) can be obtained in the W-V-H₂O system. The concentration of each species can be obtained by respectively specifying the total concentration of tungsten and vanadium in the system and the pH of the solution (to simplify the calculations, only the species of 10 g∙L^‒1 tungsten and 1 g∙L^‒1 vanadium under weak alkaline conditions were considered). All the collected thermodynamic data for the W-V-H₂O system were provided in Table S2 (cf. ESM). Based on the above analysis, the molar ratio of tungsten and vanadium in the pH range of 8.0–9.5 were depicted in Fig.3. HVO₄^2‒ was gradually converted to V₄O₁₂^4‒ as the pH decreased from 9.5 to 8.3, whereas tungsten mainly existed as WO₄^2‒. With further decreased in pH, tungstovanadic heteropoly acid of V₂W₄O₁₉^4‒ gradually formed. These results indicated that vanadium has a stronger ability to form polymeric ions than tungsten in the pH range of 8.0–9.0, which further provides a theoretical basis for the selective separation of tungsten and vanadium:

Tab.2 Equilibrium constants and calculation formulas of relevant ion species

Equation No.	Relevant reaction	lg k	Formula	Ref.
(1)	WO42‒ + H+ = HWO24‒	3.5	[HWO4‒] = 103.5[WO42‒][H+]	[8]
(2)	7WO42‒ + 8H+ = W7O246‒ + 4H2O	65.19	[W7O246‒] = 1065.19[WO42‒]7[H+]8	[35]
(3)	7WO42‒ + 9H+ = HW7O245‒ + 4H2O	69.96	[HW7O245‒] = 1069.96[WO42‒]7[H+]9	[35]
(4)	12WO42‒ + 14H+ = H2W12O4210‒ + 6H2O	115.38	[H2W12O4210‒] = 10115.38[WO42‒]12[H+]14	[8]
(5)	VO43‒ + H+ = HVO42‒	13.36	[HVO42‒] = 1013.36[VO43‒][H+]	[8]
(6)	2VO43‒ + 3H+ = HV2O73‒ + H2O	37.17	[HV2O73‒] = 1037.17[VO43‒]2[H+]3	[8]
(7)	4VO43‒ + 8H+ = V4O124‒ + H2O	95.11	[V4O124‒] = 1095.11[VO43‒]4[H+]8	[8]
(8)	VO43‒ + 4H+ = VO2+ + 2H2O	28.23	[VO2+] = 1028.23[VO43‒][H+]4	[8]
(9)	2VO43‒ + 4WO42‒ + 10H+ = V2W4O194‒ + 5H2O	99.29	[V2W4O194‒] = 1099.29[VO43‒]2[WO42‒]4[H+]10	[8]
(10)	3VO43‒ + 3WO42‒+ 10H+ = V3W3O195‒ + 5H2O	105.49	[V3W3O195‒] = 10105.49[VO43‒]3[WO42‒]3[H+]10	[8]

Fig.3 Molar ratio of tungsten and vanadium species as a function of pH in W-V-H₂O system (C_W = 10 g∙L^‒1, C_V = 1.0 g∙L^‒1; dash line: tungsten, solid line: vanadium, 25 °C).

Full size|PPT slide

(11)\begin{array}{rl}[\mathrm{W}{]}_{\mathrm{T}}=& [{\mathrm{W}{\mathrm{O}}_4}^{2-}]+[{\mathrm{H}\mathrm{W}{\mathrm{O}}_4}^{-}]+7[{\mathrm{W}}_7{{\mathrm{O}}_{24}}^{6-}]\\ & +7[\mathrm{H}{\mathrm{W}}_7{{\mathrm{O}}_{24}}^{5-}]+12[{\mathrm{H}}_2{\mathrm{W}}_{12}{{\mathrm{O}}_{42}}^{10-}]\\ & +4[{\mathrm{V}}_2{\mathrm{W}}_4{{\mathrm{O}}_{19}}^{4-}]+3[{\mathrm{V}}_3{\mathrm{W}}_3{{\mathrm{O}}_{19}}^{5-}],\end{array}

(12)\begin{array}{rl}[\mathrm{V}{]}_{\mathrm{T}}=& [{\mathrm{V}{\mathrm{O}}_4}^{3-}]+[{\mathrm{H}\mathrm{V}{\mathrm{O}}_4}^{2-}]+2[\mathrm{H}{\mathrm{V}}_2{{\mathrm{O}}_7}^{3-}]+4[{\mathrm{V}}_4{{\mathrm{O}}_{12}}^{4-}]\\ & +[{\mathrm{V}{\mathrm{O}}_2}^{+}]+2[{\mathrm{V}}_2{\mathrm{W}}_4{{\mathrm{O}}_{19}}^{4-}]+3[{\mathrm{V}}_3{\mathrm{W}}_3{{\mathrm{O}}_{19}}^{5-}].\end{array}

3.2 Mineralization behaviors of tungsten and vanadium

Similar to foam mineralization during mineral flotation, reagent mineralization is crucial for the formation of Aliquat336-V hydrophobic complexes which is a key step for separating selectively tungsten and vanadium [36]. As shown in Fig.4, the surfactant Aliquat336 was combined with vanadium to form an Aliquat336-V hydrophobic complex in a homogeneous aqueous solution, which is considered the reagent mineralization process.

Fig.4 Schematic diagram of flotation foam mineralization and reagent mineralization processes.

Full size|PPT slide

Prior to floating-extraction, the effects of pH, mineralization time, Aliquat336 dosage, and initial concentrations on the mineralization behaviors were investigated systematically, and the results were depicted in Fig.5. As shown in Fig.5(a), the initial pH is a key factor in the selective separation of tungsten and vanadium. The mineralization efficiency of vanadium decreased rapidly and the loss of tungsten increased significantly as the initial pH decreased from 8.5 to 8.1, it could be attributed to the formation of tungstovanadic heteropoly acid. In addition, when the pH was higher than 8.5, the mineralization efficiency of vanadium gradually decreased due to the species of vanadate changing from polynuclear ions to mononuclear ions, and the mineralization efficiency of tungsten varied between 2% and 3%. Owing to the sufficient contact between metal ions and surfactants in a homogeneous aqueous solution, the reagent mineralization process was quite fast. As shown in Fig.5(b), more than 85% of vanadium was mineralized after only 1 min, the separation efficiency reached maximum after mineralization for 20 min, and the mineralization efficiencies of tungsten and vanadium of 4.53% and 91.92%, respectively.

Fig.5 Effect of solution chemical conditions on the mineralization efficiency of tungsten and vanadium: (a) effect of initial pH (Aliquat336 = 1%, mineralization time = 20 min), (b) effect of mineralization time (Aliquat336 = 1%; pH = 8.67), and (c)–(f) effects of Aliquat336 dosage and initial concentrations (pH = 8.67, mineralization time = 20 min).

Full size|PPT slide

As shown in Fig.5(c), the Aliquat336 dosage significantly affected the mineralization efficiencies of tungsten and vanadium. With increasing Aliquat336 dosage from 1% to 10%, the mineralization efficiency of vanadium and tungsten increased from 90.25% to 99.29% and from 4.70% to 61.73%, respectively. Furthermore, the mineralization efficiencies of tungsten and vanadium with different initial concentrations and Aliquat336 dosage were investigated. As depicted in Fig.5(d)–Fig.5(f), the initial solutions with different ratios of tungsten and vanadium exhibited higher separation efficiency at the Aliquat336 dosage of 1%, and the mineralization efficiency of tungsten was greatly affected by the concentration of vanadium in the solutions. In conclusion, the optimal mineralization conditions for vanadium (Aliquat336 = 1%, time = 20 min and pH = 8.5) were chosen for subsequent experiments. In addition, the effect of impurity ions on the mineralization of tungsten and vanadium was considered, and the results were shown in Fig. S2 (cf. ESM).

3.3 Floating-extraction behaviors of tungsten and vanadium

Based on the mineralization behaviors, Aliquat336-V hydrophobic complexes can be easily attached to rising bubbles and enriched in the organic phase by floating-extraction due to its strong hydrophobicity. Fig.6 displayed the effect of different flotation conditions on the flotation behavior of tungsten and vanadium. As shown in Fig.6(a), among the five diluents, the flotation efficiency of vanadium varied within 1%–2% for different diluents, and the maximum flotation efficiency of 93.14% was obtained when sulfonated kerosene was adopted as a diluent. Based on the flotation behaviors and economic considerations, the diluent sulfonated kerosene was chosen as a suitable diluent for this study. To prevent the formation of the third phase, 2-octanol was added to the organic phase as a phase modifier. As displayed in Fig.6(b), when the content of 2-octanol exceeded 10%, the flotation efficiency of vanadium decreased slightly as the 2-octanol content continued to increase, which was attributed to the decrease of the relative content of sulfonated kerosene with the increase of 2-octanol content. Therefore, the organic phase consists of sulfonated kerosene and 10% of 2-octanol was considered optimal.

Fig.6 Effect of flotation conditions (O/A = 1/5) on flotation efficiency of tungsten and vanadium: (a) effect of different organic solvents (flow rate = 30 mL∙min^‒1, time = 30 min, 2-octanol = 10%), (b) effect of 2-octanol concentration (flow rate = 30 mL∙min^‒1, time = 30 min), (c) effect of flow rate (time = 30 min, 2-octanol = 10%), and (d) effect of flotation time (flow rate = 30 mL∙min^‒1, 2-octanol = 10%).

Full size|PPT slide

Floating-extraction experiments for the separation of tungsten and vanadium were performed at different flow rates ranging from 10 to 90 mL∙min^‒1. The results in Fig.6(c) indicated that above 50 mL∙min^‒1, the flow rate had little effect on the flotation efficiency of tungsten and vanadium. Furthermore, due to the higher bubble mass transfer rate, the flotation rate of Aliquat336-V hydrophobic complexes is fast. As presented in Fig.6(d), the flotation efficiency of vanadium was as high as 85.88% at the flotation time of 1 min, and the flotation equilibrium can be reached within 30 min.

3.4 Stripping of tungsten and vanadium from floated organic phase

Prior to stripping, a scrubbing experiment of the co-floating tungsten in the organic phase was performed. However, it was difficult to wash out tungsten alone to recover high-purity vanadium products by a simple scrubbing process. This may be caused by tungsten mainly existing as tungstovanadicheteropoly acid in the floated organic phase, and the scrubbing experiment results were shown in Fig. S3 (cf. ESM). To realize the comprehensive utilization of tungsten and vanadium from the floated organic phase, the co-stripping study of tungsten and vanadium from the organic phase was performed, which can be used for the fabrication of tungsten-vanadium alloys in subsequent experiments.

The effects of different conditions on the stripping process of tungsten and vanadium were considered and the results were shown in Fig.7. As illustrated in Fig.7(a), among these stripping agents, the combined stripping agent of 0.5 mol∙L^‒1 sodium hydroxide + 0.5 mol∙L^‒1 sodium chloride was considered the best stripping agent. The stripping efficiency of vanadium was as high as 94.60%, while that of tungsten was still 57.75%. Therefore, the effects of the concentration and proportion of sodium hydroxide and sodium chloride on the stripping efficiency of tungsten and vanadium were investigated. As displayed in Fig.7(b) and Fig.7(c), the stripping efficiency of tungsten was greatly affected by the concentration of the reagents, and the stripping efficiency of vanadium increased from 57.75% to 87.54% with the concentration of sodium hydroxide and sodium chloride increasing from 0.5 to 1.5 mol∙L^‒1. However, the stripping efficiency of tungsten and vanadium was weakly affected by sodium chloride content. Thus, the mixed solution of 1.5 mol∙L^‒1 sodium hydroxide + 0.25 mol∙L^‒1 sodium chloride was selected as the stripping agent, and Fig.7(d) displayed the stripping efficiency of tungsten and vanadium throughout four-stage stripping. The stripping efficiency of vanadium increased from 95.55% to 98.06%, and that of tungsten increased from 88.42% to 95.10%.

Fig.7 Effect of different conditions on the stripping process of tungsten and vanadium (O/A = 1): (a) effect of stripping agents, (b) effect of concentrations of sodium hydroxide and sodium chloride, (c) effect of the proportion of sodium hydroxide (A) and sodium chloride (B), and (d) multi-stage stripping efficiency using 1.5 mol∙L^‒1 sodium hydroxide + 0.25 mol∙L^‒1 sodium chloride.

Full size|PPT slide

3.5 Separation mechanism analysis

3.5.1 Mineralization mechanism

Based on the above results, it can be speculated that reagent mineralization is a key step for the selective separation of tungsten and vanadium, and vanadium can be preferentially mineralized by Aliquat336 according to the polymerization ability differences of tungsten and vanadium. Additionally, controlled-variables experiment, FTIR analysis, molecular dynamics (MD) simulations, and density functional theory (DFT) calculations were implemented to elucidate the interaction mechanisms and the species of hydrophobic complexes. According to the experiment results for the controlled-variables (results were listed in Table S3, cf. ESM), the flotation efficiency of vanadium slightly decreased with increasing W concentration. Therefore, it is speculated that the two metals exhibit a competitive mineralization relationship, and the combining capacity of vanadium to Aliquat336 is much greater than that of tungsten.

The FTIR spectra of Aliquat336 were conducted before and after mineralization of different solutions at pH 8.5 in order to confirm the existence of chemical interactions in the mineralization process, and the results were displayed in Fig.8. The broadband in the area of 3100–3500 cm^‒1 was ascribed to N–H vibrations, the peaks from 3000 to 2800 cm^‒1 were belonged to C–H stretching vibration and the peaks at 1200–1500 cm^‒1 corresponded to the C–N bond frequency [37,38]. Compared to the FTIR spectrum of Aliquat336, new peaks at 919, 835, 825, 786, and 641 cm^‒1 appeared after the mineralization of tungsten and vanadium in the ionic liquid phase. According to previous reports, the band of Aliquat336 mineralized tungsten alone appeared at 825 cm^‒1 was attributed to W–O [39,40], and the band of Aliquat336 mineralized vanadium alone appeared at 919 cm^‒1 was caused by the V=O stretching vibrations corresponding to HVO₄^2‒ [3,41], the bands at 835 and 641 cm^‒1 were attributed to the symmetric and asymmetric V–O–V stretching present in isopolyvanadate [42,43], which corresponding to V₄O₁₂^4‒. In addition, the band of Aliquat336 mineralized tungsten and vanadium mixed solution appeared at 786 cm^‒1 was attributed to V–O–W stretch vibrations, which corresponding to V₂W₄O₁₉^4‒ [44,45]. Comparing the infrared spectra of Aliquat336 mineralized vanadium alone and Aliquat336 mineralized tungsten and vanadium mixed solution, only one peak’s position has changed. Hence, we infer that the mineralization reactions in the two systems were basically consistent at pH 8.5, and trace species of V₂W₄O₁₉^4‒ may be formed during the acidification of the solutions.

Fig.8 FTIR analysis of Aliquat336 before and after mineralization (black line: Aliquat336; red line: Aliquat336 mineralized tungsten and vanadium mixed solution; green line: Aliquat336 mineralized vanadium alone; blue line: Aliquat336 mineralized tungsten alone).

Full size|PPT slide

Based on the above results, we speculate that the possible mineralized species of vanadium mainly include V₄O₁₂^4‒, V₂W₄O₁₉^4‒ and HVO₄^‒, and the Aliquat336-V hydrophobic complexes were formed by anion exchange reaction during the reagent mineralization process [46]. DFT calculations and MD simulations were used to further reveal the relationship between Aliquat336 and different species (computation methods and details were described in ESM). The electrostatic potential maps of WO₄^2‒, HVO₄^‒, V₄O₁₂^4‒, and V₂W₄O₁₉^4‒ were displayed in Fig. S4 (cf. ESM), and the Mulliken charges and atomic number of them were shown in Table S4 and Fig. S5 (cf. ESM), respectively. The physicochemical properties of polymeric ions and mononuclear ions exhibited considerable differences, which provided the possibility for the selective separation of tungsten and vanadium. In addition, the interaction energy (ΔE) was used to indicate the bonding strength between Aliquat336 and different species, and the increasingly negative ΔE values indicate more favorable interactions between Aliquat336 and species [47]. As displayed in Fig. S6 (cf. ESM), the negative ΔE followed the order as Aliquat336-WO₄^2‒ < Aliquat336-HVO₄^‒ < Aliquat336-V₄O₁₂^4‒ < Aliquat336-V₂W₄O₁₉^4‒. Thus, V₂W₄O₁₉^4‒ is preferentially mineralized, followed by V₄O₁₂^4‒, while the mononuclear ions HVO₄^2‒ and WO₄^2‒ form weaker bonds with Aliquat336 during mineralization process. Therefore, we infer that vanadium reacted with Aliquat336 mainly in the form of V₄O₁₂^4‒ by ion exchange during the mineralization process, and the mineralization reaction equation was expressed in Eq. (13) by calculating the stoichiometric ratio of extractant to vanadium (details were described in ESM),

(13)4\overline{({\mathrm{R}}_4\mathrm{N})\mathrm{C}\mathrm{l}}+\mathrm{N}{\mathrm{a}}_4{\mathrm{V}}_4{\mathrm{O}}_{12}=\overline{{({\mathrm{R}}_4\mathrm{N})}_4{\mathrm{V}}_4{\mathrm{O}}_{12}}+4\mathrm{N}\mathrm{a}\mathrm{C}\mathrm{l}

3.5.2 Floating-extraction mechanism

Generally, microbubble flotation-extraction is established on the basis of the corresponding solvent extraction system, but unlike the mass transfer process at the oil−water interface during the solvent extraction process. Fig.9 displayed the mechanism of floating-extraction mechanism. As depicted in Fig.9(a), we infer that the separation process of flotation-extraction consists of two individual processes: (1) a mass transfer process of bubbles and water droplets, and (2) a diffusive transport at the oil–water interface driven by a concentration gradient [48]. During the first process, Aliquat336-V hydrophobic complexes can be easily attached to the surface of rising microbubbles due to its strong hydrophobicity, and then enriched in the organic phase at the top. In addition, we infer that the small bubbles cannot cross the water–organic interface immediately, and they should coalesce into large ones to overcome the interfacial tension [48]. Furthermore, the morphology of the hydrophobic complexes was kept constant before and after the collapse of bubbles, only the hydrophobic chain was transferred from the gas phase to the organic phase. Therefore, we infer that the hydrophobic complexes merge into a small water droplet due to the collapse of the bubbles when reaching the top of the organic phase. Fig.9(b) and Fig.9(c) displayed the process of a bubble crossing the water–organic interface and the bubble bursting at the oil–gas interface, respectively. Compared with the traditional solvent extraction, the concentration of the target component on the bubble surface is higher than the water–organic interface. Furthermore, after the bubble burst, the target components converged into small droplets, and the total area of the interface rapidly decreased. Therefore, bubbles effectively increase the target component concentration of the mass transfer interface.

Fig.9 (a) Mass transfer process during the floating-extraction (C_O is the concentration of the target substance in the organic phase at a certain time, and C_W is the concentration of the target substance in the water phase at a certain time), (b) process of bubbles from oil phase to water phase, and (c) aqueous layer and droplets in floating-extraction.

Full size|PPT slide

3.6 Comprehensive evaluation of microbubble floating-extraction process

To further evaluate the advantages of microbubble floating-extraction technology, the separation of tungsten and vanadium were compared via microbubble floating-extraction and solvent extraction under optimal experimental conditions. Fig.10 displayed a comprehensive evaluation of floating-extraction mechanism. As displayed in Fig.10(a), the recovery efficiency of vanadium increased from 94.08% to 95.44%, while the loss efficiency of tungsten decreased from 9.68% to 6.11%. Therefore, the separation factor was greatly increased from 148 to 321. In addition, Tab.3 compared the present developed method with earlier reported in related literature on the separation of tungsten and vanadium. Microbubble-floating extraction offers significantly lower consumption of reagents and significantly higher separation efficiency of tungsten and vanadium compared to solvent extraction. In the same separation system, the consumption of Aliquat336 in microbubble floating-extraction is almost twice lower than that in solvent extraction, which greatly reduces the costs of the selective separation of tungsten and vanadium.

Tab.3 Comparison of the present developed method with earlier reported in related literature on the separation of tungsten and vanadium

Method	Feed solutions	Experimental condition	Separation efficiency/%	Ref.
Solvent extraction	1.0 g∙L‒1 tungsten0.1 g∙L‒1 vanadium	0.1 mol∙L‒1 LiX63-decyl alcohol-kerosene pH = 8.0	Tungsten extraction could be neglectedVanadium extraction of 70%	[5]
Solvent extraction	52.5 g∙L‒1 tungsten6.4 g∙L‒1 vanadium	20% Aliquat336-40% 2-octanol-sulfonated kerosene as the organic phase pH = 8.7, O/A = 2/1	Tungsten extraction of 12.34%Vanadium extraction of 97.13%	[49]
Solvent extraction	10 g∙L‒1 tungsten1.0 g∙L‒1 vanadium	10% Aliquat336-10% 2-octanol-sulfonated kerosene as the organic phase pH = 8.6, O/A = 1/6	Tungsten extraction of 7.74%Vanadium extraction of 92.02%	[50]
Microbubble floating-extraction	10 g∙L‒1 tungsten1.0 g∙L‒1 vanadium	1% Aliquat336 as the mineralization reagent, 10% 2-octanol-sulfonated kerosene as the organic phase, pH = 8.5, and O/A = 1/5	Tungsten flotation efficiency of 6.11%Vanadium flotation efficiency of 95.44%	This work

Fig.10 (a) Comparison of the separation efficiencies of tungsten and vanadium between microbubble floating-extraction and solvent extraction, and (b) effect of Aliquat336 dosage on two-stage microbubble floating-extraction.

Full size|PPT slide

In order to further improve the separation efficiency of tungsten and vanadium, we studied the two-stage microbubble floating-extraction. During the two-stage microbubble floating-extraction process, the raffinate after the first microbubble floating-extraction was collected, and the removal efficiency of trace vanadium with different dosages of Aliquat336 was studied. As shown in Fig.10(b), at the Aliquat336 dosage of 0.2%, the removal efficiency of vanadium in the two-stage process reached 98.5% with tungsten loss below 8%, and the separation factor was as high as 771, indicating excellent the selective separation of tungsten and vanadium. In conclusion, microbubble floating-extraction could be considered as an efficient separation process for similar metals.

4 Conclusions

Thermodynamic calculations of the W-V-H₂O system indicated that the polymerization ability of vanadium is much stronger than that of tungsten in the pH range of 8.0–9.0, which expanded the differences in affinity between the reagent and the metal ions. Therefore, vanadium can be selectively separated from tungstate solutions by floating-extraction using Aliquat336 as a surfactant.

Based on the separation behaviors of tungsten and vanadium, the selective separation of tungsten and vanadium by floating-extraction was conducted under the following conditions: pH 8.5, mineralization time of 20 min, 1.2% (v/v) Aliquat336 as a surfactant, and 10% (v/v) 2-octanol in sulfonated kerosene as an organic phase, flow rate of 40 mL∙min^‒1, and flotation time of 30 min. Followed by stripping via 1.5 mol∙L^‒1 sodium hydroxide + 0.25 mol∙L^‒1 sodium chloride, reaction temperature of 25 °C and 30 min. Ultimately, the removal efficiency of vanadium was as high as 98.5%, and the loss efficiency of tungsten was less than 8% in the two-stage microbubble floating-extraction.

The formation of the Aliquat336-V hydrophobic complexes is a key step for separating selectively tungsten and vanadium, and the mineralized vanadium species was identified mainly isopolyvanadate of V₄O₁₂^4‒ via experimental results, FTIR analysis, and simulation calculation. In addition, owing to the higher bubble and interface mass transfer rates, floating-extraction can achieve high-efficiency enrichment and deep separation of the Aliquat336-V₄O₁₂^4‒ hydrophobic complexes.

Microbubble floating-extraction technology combines the advantages of solvent extraction and ion flotation, solving the scientific problems of oil−water dispersion and high consumption of organic phase during solvent extraction. Compared with solvent extraction, the removal efficiency of vanadium was increased by 1.36%, and the loss rate of tungsten was decreased by 3.57% under optimal conditions. Therefore, microbubble floating-extraction can be considered a highly efficient separation process in the field of similar metals.

Acknowledgements

This work was financially supported by the Original Exploration Project of China (Grant No. 52150079), the National Natural Science Foundation of China (Grant Nos. U2004215, 51974280 and 51774252), the Educational Commission Fund of Henan Province of China (Grant Nos. 20HASTIT012, 18A450001 and 17A450001).

Electronic Supplementary Material

Supplementary material is available in the online version of this article at https://dx.doi.org/10.1007/s11705-022-2235-2 and is accessible for authorized users.

References

1 Nguyen T H, Lee M S. Separation of vanadium and tungsten from sodium molybdate solution by solvent extraction. Industrial & Engineering Chemistry Research, 2014, 53(20): 8608–8614

2 Zhang W J, Chen Y Q, Che J Y, Wang C Y, Ma B Z. Green leaching of tungsten from synthetic scheelite with sulfuric acid-hydrogen peroxide solution to prepare tungstic acid. Separation and Purification Technology, 2020, 241: 116752

3 Liu Z S, Huang J, Zhang Y M, Liu T, Hu P C, Liu H, Zheng Q S. Separation and recovery of vanadium and iron from oxalic-acid-based shale leachate by coextraction and stepwise stripping. Separation and Purification Technology, 2020, 244: 116532

4 Truong H T, Nguyen T H, Lee M S. Separation of molybdenum(VI), rhenium(VII), tungsten(VI), and vanadium(V) by solvent extraction. Hydrometallurgy, 2017, 171: 298–305

5 Choi I H, Kim H R, Moon G, Jyothi R K, Lee J Y. Spent V₂O₅–WO₃/TiO₂ catalyst processing for valuable metals by soda roasting-water leaching. Hydrometallurgy, 2018, 175: 292–299

6 Zhang Q J, Wu Y F, Yuan H R. Recycling strategies of spent V₂O₅–WO₃/TiO₂ catalyst: a review. Resources, Conservation and Recycling, 2020, 161: 104983

7 Ferella F. A review on management and recycling of spent selective catalytic reduction catalysts. Journal of Cleaner Production, 2020, 246: 118990

8 Zhang J L, Zhao Z W. Thermodynamic analysis of tungsten–vanadium separation in W(VI)-V(V)-H₂O system. Chinese Journal of Nonferrous Metals, 2014, 24(6): 1656–1662 (in Chinese)

9 Luo L, Liu K J, Shibayama A, Yen W T, Fujita T, Shindo O, Katai T. Recovery of tungsten and vanadium from tungsten alloy scrap. Hydrometallurgy, 2004, 72(1-2): 1–8

10 Luo L, Miyazaki T, Shibayama A, Yen W T, Fujita T. Separation of vanadium and tungsten from a sodium tungstate solution. Canadian Metallurgical Quarterly, 2003, 42(4): 411–420

11 Zhu X Z, Huo G S, Ni J, Song Q. Removal of tungsten and vanadium from molybdate solutions using ion exchange resin. Transactions of Nonferrous Metals Society of China, 2017, 27(12): 2727–2732

12 Wu W C, Tsai T Y, Shen Y H. Tungsten recovery from spent SCR catalyst using alkaline leaching and ion exchange. Minerals, 2016, 6(4): 107–117

13 Wang L P, Zhang G Q, Guan W J, Zeng L, Zhou Q, Xia Y, Wang Q, Li Q G, Cao Z Y. Complete removal of trace vanadium from ammonium tungstate solutions by solvent extraction. Hydrometallurgy, 2018, 179: 268–273

14 Nguyen T H, Lee M S. A review on the separation of molybdenum, tungsten, and vanadium from leach liquors of diverse resources by solvent extraction. Geosystem Engineering, 2016, 19(5): 247–259

15 Zeng L, Yong C Y. A literature review of the recovery of molybdenum and vanadium from spent hydrodesulphurisation catalysts. Part II: separation and purification. Hydrometallurgy, 2009, 98(1-2): 10–20

16 Kim J W, Hwang I J. Separation of valuables from spent selective catalytic reduction catalyst leaching solution by fabricated anion extraction resins. Journal of Environmental Chemical Engineering, 2018, 6(1): 1100–1108

17 Wu J, Wei C, Li X B, Wang S F, Wang M S, Li C X. Selective extraction of Mo using Cyanex-272 and tributyl phosphate from low grade Ni−Mo ore leach liquor. Separation and Purification Technology, 2012, 99: 120–126

18 Wilson A M, Bailey P J, Tasker P A, Turkington J R, Grant R A, Love J B. Solvent extraction: the coordination chemistry behind extractive metallurgy. Chemical Society Reviews, 2014, 43(1): 123–134

19 Huang K, Liu J, Wu H Z, Liu H Z. A new bubbling extraction tower: toward liquid–liquid solvent extraction at large aqueous-to-oil phase ratios. AIChE Journal, 2015, 61(11): 3889–3897

20 Liu J, Huang K, Wu H Z, Liu H Z. A feasible strategy for calculating the required mass transfer height of a new bubbling organic liquid membrane extraction tower directly based upon the experimental kinetic data. Industrial & Engineering Chemistry Research, 2016, 55(16): 4426–4434

21 Rout P C, Sarangi K. A comparative study on extraction of Mo(VI) using both solvent extraction and hollow fiber membrane technique. Hydrometallurgy, 2013, 133: 149–155

22 Liao C S, Cheng F X, Wu S, Yan C H. Review and recent progresses on theory of countercurrent extraction. Journal of the Chinese Society of Rare Earths, 2017, 35: 1–8 (in Chinese)

23 Wang W T, Sang F N, Xu J H, Wang Y D, Luo G S. The enhancement of liquid–liquid extraction with high phase ratio by microfluidic-based hollow droplet. RSC Advances, 2015, 5(100): 82056–82064

24 Liu J, Huang K, Wu X H, Liu H Z. Enrichment of low concentration rare earths from leach solutions of ion-adsorption ores by bubbling organic liquid membrane extraction using N1923. ACS Sustainable Chemistry & Engineering, 2017, 5(9): 8070–8078

25 Liu J, Huang K, Wu X H, Liu W Q, Song W Y, Liu H Z. Extraction of rare earths using bubbling organic liquid membrane with un-saponified P507. Hydrometallurgy, 2018, 175: 340–347

26 Huang Y F, Shi K P, Liu B B, Su S P, Han G H. Research status and prospect of deep separation technology for dissolved molybdenum and vanadium. Conservation and Utilization of Mineral, 2021, 41(5): 65–72 (in Chinese)

27 Han G H, Wang H Y, Su S P, Huang Y F, Liu B B. Research progress and discussion on selective separation technology of dissolved tungsten and vanadium. Chinese Journal of Nonferrous Metals, 2021, 31(11): 3380–3395 (in Chinese)

28 Han G H, Wang J W, Liu B B, Sun H, Huang Y F. Progress and prospect of cobalt recovery from cobalt slag produced by zinc hydrometallurgy. Journal of Guizhou University, 2022, 39(2): 1–6

29 Su S P, Huang Y F, Liu B B, Han G H, Xue Y B, Wang Y Z. A feasible strategy for deeply separating low concentrations of molybdenum from tungstate solutions with a high-efficiency microbubble floating-extraction concept. ACS Sustainable Chemistry & Engineering, 2021, 10(1): 146–158

30 Kang J H, Hu Y H, Sun W, Liu R Q, Gao Z Y, Guan Q J, Tang H H, Yin Z G. Utilisation of FGD gypsum for silicate removal from scheelite flotation wastewater. Chemical Engineering Journal, 2018, 341: 272–279

31 Chen J H. The interaction of flotation reagents with metal ions in mineral surfaces: a perspective from coordination chemistry. Minerals Engineering, 2021, 171: 107067

32 Valsaraj K T, Porter J L, Liljenfeldt E K, Springer C. Solvent sublation for the removal of hydrophobic chlorinated compounds from aqueous solutions. Water Research, 1986, 20(9): 1161–1175

33 Sola A B C, Parhi P K, Lee J Y, Kang H N, Jyothi R K. Environmentally friendly approach to recover vanadium and tungsten from spent SCR catalyst leach liquors using Aliquat 336. RSC Advances, 2020, 10(34): 19736–19746

34 Huang Y F, Zhang B, Liu B B, Su S P, Han G H, Guo H, Cao Y J. Clean and deep separation of molybdenum and rhenium from ultra-low concentration solutions via vapidly stepwise selective coagulation and flocculation precipitation. Separation and Purification Technology, 2021, 267: 118632

35 Zhang J L, Zhao Z W, Chen X Y, Liu X H. Thermodynamic analysis for separation of tungsten and molybdenum in W–Mo–H₂O system. Chinese Journal of Nonferrous Metals, 2013, 23(5): 1463–1470 (in Chinese)

36 Wu H Y, Wang W J, Huang Y F, Han G H, Yang S Z, Su S P, Sana H, Peng W J, Cao Y J, Liu J T. Comprehensive evaluation on a prospective precipitation-flotation process for metal-ions removal from wastewater simulants. Journal of Hazardous Materials, 2019, 371: 592–602

37 Jeon J H, Sola A B C, Lee J Y, Koduru J R, Jyothi R K. Separation of vanadium and tungsten from synthetic and spent catalyst leach solutions using an ion-exchange resin. RSC Advances, 2022, 12(6): 3635–3645

38 Xiong P, Zhang Y M, Huang J, Bao S X, Yang X, Shen C. High-efficient and selective extraction of vanadium(V) with N235-P507 synergistic extraction system. Chemical Engineering Research & Design, 2017, 120: 284–290

39 Kovács T N, Pokol G, Gáber F, Nagy D, Igricz T, Lukács I E, Fogarassy Z, Balázsi K, Szilágyi I M. Preparation of iron tungstate (FeWO₄) nanosheets by hydrothermal method. Materials Research Bulletin, 2017, 95: 563–569

40 Rakshit S, Sallman B, Davantes A, Lefevre G. Tungstate(VI) sorption on hematite: an in situ ATR-FTIR probe on the mechanism. Chemosphere, 2017, 168: 685–691

41 Jayadas S, Reddy M L. Solvent extraction separation of vanadium(V) from multivalent metal chloride solutions using 2-ethylhexyl phosphonic acid mono-2-ethylhexyl ester. Journal of Chemical Technology and Biotechnology, 2002, 77(10): 1149–1156

42 Mahandra H, Singh R, Gupta B. Recovery of vanadium(V) from synthetic and real leach solutions of spent catalyst by solvent extraction using Cyphos IL 104. Hydrometallurgy, 2020, 196: 105405

43 Bal Y, Bal K E, Cote G, Lallam A. Characterization of the solid third phases that precipitate from the organic solutions of Aliquat® 336 after extraction of molybdenum(VI) and vanadium(V). Hydrometallurgy, 2004, 75(1-4): 123–134

44 Paul S, Berrier E, França M C K, Eon J G. Oxidative dehydrogenation of propane under steady-state and transient regimes over alumina-supported catalysts prepared from mixed V₂W₄O₁₉^4‒ hexametalate precursors. Journal of Natural Gas Chemistry, 2010, 19(2): 123–133

45 Qi Y F, Wang E B, Li J, Li Y G. Two organic-inorganic poly (pseudo-rotaxane)-like composite solids constructed from polyoxovanadates and silver organonitrogen polymers. Journal of Solid State Chemistry, 2009, 182(10): 2640–2645

46 Nayl A A, Aly H F. Solvent extraction of V(V) and Cr(III) from acidic leach liquors of ilmenite using Aliquat 336. Transactions of Nonferrous Metals Society of China, 2015, 25(12): 4183–4191

47 Su S P, Wang W J, Liu B B, Huang Y F, Yang S Z, Wu H Y, Han G H, Cao Y J. Enhancing surface interactions between humic surfactants and cupric ion: DFT computations coupled with MD simulations study. Journal of Molecular Liquids, 2021, 324: 114781

48 Bi P Y, Dong H R, Dong J. The recent progress of solvent sublation. Journal of Chromatography A, 2010, 1217(16): 2716–2725

49 Li Q, Xiao L S, Zhang G Q, Cao Z Y, Li Q G, Zeng L, Guan W J. Vanadium separation from sodium tungstate solution by solvent extraction with quaternary ammonium salt of N263. Rare Metals and Cemented Carbides, 2017, 45(2): 20–27 (in Chinese)

50 WangH YHanG HHuangY FSuS P. Solvent extraction separation of tungsten and vanadium from simulated leaching solution of spent SCR catalyst. In: Ouchi T, Azimi G, Forsberg K, Kim H, Alam S, Neelameggham N, Baba A, Peng H, eds. Rare Metal Technology 2022. Berlin: Springer, 2022