Polyoxometalate/Ionic Liquid Desulfurization System for Hydrogen Sulfide Removal from High-Temperature Gas Stream

The temperature of industrial gas containing harmful H2S can reach hundreds of degrees. However, few processes can be used directly for H2S removal from industrial high-temperature gas. In this work, three polyoxometalates with different central atoms ((n-Bu4N)3VMo12O40, (n-Bu4N)3PMo12O40, and (n-Bu4N)4[α-SiMo12O40]) were synthesized and dissolved in four ionic liquids (Bmim]Cl, [Bmim]HCO3, [Bmim]Mes, or [Bmim]OAc) for H2S removal from high-temperature (90–180 °C) gases. The result showed that (n-Bu4N)3VMo12O40/[Bmim]OAc exhibited the optimal desulfurization performance, maintaining more than 98.6% desulfurization efficiency within 10 h. The reacted desulfurization solution can be regenerated by blowing air. FT-IR and XPS results show that both the central atom V and the coordination atom Mo of the polyoxometalate are involved in the oxidation of H2S; after the regeneration by introducing air, V(+IV) and Mo(+IV) recovered to V(+V) and Mo(+VI), respectively. Our research shows that (n-Bu4N)3VMo12O40/[Bmim]OAc is an efficient, easy-to-regenerate, and suitable high-temperature gas desulfurization solution.


Introduction
Hydrogen sulfide (H 2 S) is a toxic and harmful gas widely present in natural gas, refinery waste gas, coke oven gas, and syngas [1,2]. It not only poses a threat to human health, but also causes corrosion of production equipment and pipelines and poisoning of catalysts [3,4]. Therefore, it must be removed and transformed. The process of H 2 S removal is divided into the dry process and the wet process. The dry process mainly includes the molecular sieve method, activated carbon method, zinc oxide method, iron oxide method, etc. [2]. These desulfurization processes are characterized by high desulfurization precision, simple equipment, and convenient operation, but low desulfurization load and difficult regeneration. The wet process has the advantage of a high desulfurization load. Among them, the alcohol amine solution method is the most widely used wet desulfurization process [5]. However, this organic solvent and other solvents (usually water) have a very low applicable temperature and are difficult to meet the requirements of high-temperature desulfurization [4]. The temperature of some industrial waste gas can reach hundreds of degrees, and it must be cooled before entering the traditional wet desulfurization process, which increases the investment cost [2]. Therefore, it is of great significance to develop a desulfurization system with high-temperature resistance, easy regeneration, and high efficiency for the H 2 S removal from high-temperature gas.
Ionic liquids are salts that are liquid at low temperature (<100 • C) which represent a class of solvents with nonmolecular, ionic character [6]. Wang et al. [7,8] reported a novel supported ionic liquid phase Cs-based system for the conversion of 1,1,2-Trichloroethane to 1,1-Dichloroethylene. The results showed that the addition of ionic liquids could significantly improve the dispersion of Cs species in the preparation and performance evaluation of the catalyst, thereby significantly improving catalytic activity. In addition, ionic liquids change the adsorption behavior of the product, thereby improving the selectivity of the product 1,1-Dichloroethylene. Ionic liquid has extremely low vapor pressure, a wide liquid phase range, and good stability at high temperature [9]. These characteristics make ionic liquids an ideal solvent for high-temperature wet desulfurization. At present, most of the research on ionic liquids in the field of H 2 S absorption is concentrated under high-pressure static conditions [10][11][12]. There are few studies on H 2 S removal under dynamic conditions of atmospheric pressure and high temperature. The H 2 S removal by using ionic liquid mainly relies on the physical absorption or alkalinity of ionic liquid, and the desulfurization performance is very limited [12,13]. Polyoxometalates are a general term for a class of oxygen-containing polybasic acids with elements such as P, Si, Ge, and S as the central atom and elements such as Mo, W, and V as the coordination atom, bridged with oxygen atoms. Polyoxometalates have not only unique redox and catalytic properties, but also good thermal stability. Combining polyoxometalates and ionic liquids may meet the needs of high-temperature wet oxidative desulfurization. The redox properties of polyoxometalates are crucial for the desulfurization performance of polyoxometalate/ionic liquid desulfurization systems [14]. The central atoms, coordination atoms, and substitution atoms of polyoxometalates all influence the redox properties of polyoxometalates. Ma et al. [15]  However, the effect of the central atom species on the H 2 S removal performance is unknown.
The purpose of this work is to develop a desulfurization system that is highly efficient, easy to regenerate, and suitable for high-temperature gas. Polyoxometalates with different central atoms were synthesized and dissolved in different ionic liquids to obtain polyoxometalates/ionic liquid desulfurization solution. The effects of various factors on the desulfurization performance were investigated under normal pressure and high-temperature conditions (90-180 • C). After the reaction, the air was pumped into the desulfurization system to realize the regeneration of the desulfurization solution. The desulfurization solution and products were analyzed by FT-IR, TGA-DSC, XRD, and XPS. The results show that the desulfurization products are elemental sulfur and sulfate, and the desulfurization system can realize multiple cycles of desulfurization at high temperature.

Polyoxometalate and Ionic Liquid Characterization
The FT-IR spectrum of (n-Bu 4 N) 3 VMo 12 O 40 is displayed in Figure 1a. The peak at 957 cm −1 is the stretching vibration peak of the Mo=O d bond, the peak at 887 cm −1 is attributed to the stretching vibration peak of V-O bond, and the peak at 775 cm −1 is the stretching vibration of Mo-O c -Mo bond peak [16][17][18]. Figure 1b is the XRD pattern of (n-Bu 4 N) 3 VMo 12 O 40 . The diffraction peaks at 11.7 • , 15.6 • , 23.8 • , and 31.7 • are the characteristic peaks of Keggin polyoxometalates [19][20][21]. The FT-IR and XRD spectra revealed that (n-Bu 4 N) 3 VMo 12 O 40 was successfully synthesized. The structure diagram of polyoxometalate anion XMo 12 O 40 (X = V, P, or Si) and the molecular structure diagram of ionic liquids are shown in Figures 2 and 3. cm is the stretching vibration peak of the Mo=Od bond, the peak at 887 cm is attributed to the stretching vibration peak of V-O bond, and the peak at 775 cm −1 is the stretching vibration of Mo-Oc-Mo bond peak [16][17][18]. Figure 1b is the XRD pattern of (n-Bu4N)3VMo12O40. The diffraction peaks at 11.7°, 15.6°, 23.8°, and 31.7° are the characteristic peaks of Keggin polyoxometalates [19][20][21]. The FT-IR and XRD spectra revealed that (n-Bu4N)3VMo12O40 was successfully synthesized. The structure diagram of polyoxometalate anion XMo12O40 (X = V, P, or Si) and the molecular structure diagram of ionic liquids are shown in Figures 2 and 3.   cm −1 is the stretching vibration peak of the Mo=Od bond, the peak at 887 cm −1 is attributed to the stretching vibration peak of V-O bond, and the peak at 775 cm −1 is the stretching vibration of Mo-Oc-Mo bond peak [16][17][18]. Figure 1b is the XRD pattern of (n-Bu4N)3VMo12O40. The diffraction peaks at 11.7°, 15.6°, 23.8°, and 31.7° are the characteristic peaks of Keggin polyoxometalates [19][20][21]. The FT-IR and XRD spectra revealed that (n-Bu4N)3VMo12O40 was successfully synthesized. The structure diagram of polyoxometalate anion XMo12O40 (X = V, P, or Si) and the molecular structure diagram of ionic liquids are shown in Figures 2 and 3.     Figure 4a shows the TGA-DSC curve of (n-Bu4N)3VMo12O40. There is only a 0.32% mass loss between 25-250 °C , which is caused by the evaporation of a small amount of water. Between 300-400 °C , the thermogravimetric curve decreased rapidly, the mass loss of the sample was 31.64%, and the DSC curve showed an obvious endothermic   Figure 4a shows the TGA-DSC curve of (n-Bu 4 N) 3 VMo 12 O 40 . There is only a 0.32% mass loss between 25-250 • C, which is caused by the evaporation of a small amount of water. Between 300-400 • C, the thermogravimetric curve decreased rapidly, the mass loss of the sample was 31.64%, and the DSC curve showed an obvious endothermic peak, indicating that (n-Bu 4 N) 3 VMo 12 O 40 decomposed in this temperature range. However, (n-Bu 4 N) 3 VMo 12 O 40 is completely stable at the highest operating temperature of 180 • C in this experiment. Figure 4b shows the TGA-DSC curve of ionic liquid [Bmim]OAc. As can be seen from the figure, the weight loss of the sample is 5.4% between 25-180 • C, which is mainly due to the evaporation of water. Between 200-250 • C, the TGA curve decreased rapidly, the sample weight loss was 67.5% in this range, and the DSC curve showed an endothermic peak in this range, indicating that [Bmim]OAc was decomposed in this temperature range. The TGA-DSC curve shows that the thermal stability of [Bmim]OAc also meets the requirements of high-temperature (<180 • C) gas desulfurization.  Figure 4a shows the TGA-DSC curve of (n-Bu4N)3VMo12O40. There is only a 0.32% mass loss between 25-250 °C , which is caused by the evaporation of a small amount of water. Between 300-400 °C , the thermogravimetric curve decreased rapidly, the mass loss of the sample was 31.64%, and the DSC curve showed an obvious endothermic peak, indicating that (n-Bu4N)3VMo12O40 decomposed in this temperature range. However, (n-Bu4N)3VMo12O40 is completely stable at the highest operating temperature of 180 °C in this experiment. Figure 4b shows the TGA-DSC curve of ionic liquid [Bmim]OAc. As can be seen from the figure, the weight loss of the sample is 5.4% between 25-180 °C , which is mainly due to the evaporation of water. Between 200-250 °C , the TGA curve decreased rapidly, the sample weight loss was 67.5% in this range, and the DSC curve showed an endothermic peak in this range, indicating that [Bmim]OAc was decomposed in this temperature range. The TGA-DSC curve shows that the thermal stability of [Bmim]OAc also meets the requirements of high-temperature (< 180 °C ) gas desulfurization.

Desulphurization Performance
The central atom of the polyoxometalate (i.e., heteroatom, which can be S 6+ , V 5+ , P 5+ , As 5+ , Ge 4+ , Si 4+ , B 3+ , etc.) affects its redox properties by changing the charge number of the heteropoly anion. As the number of charges increases, the oxidation potential of polyoxometalates decreases, for example, SMo 12 [22,23]. As shown in Figure 5, we used [Bmim]Cl as a solvent to investigate the effect of central atom species on desulfurization performance. It can be seen that the central atom has a great influence on the desulfurization performance of polyoxometalates. The desulfurization properties of polyoxometalates with different central atoms are V > P > Si, which matches the theoretical oxidation potential polyoxometalates (VMo 12  Cl solution can maintain a desulfurization rate of more than 92% within 300 min. Therefore, (n-Bu 4 N) 3 VMo 12 O 40 was selected as the solute for the next experiment.
The absorption of H 2 S by ionic liquids may affect the oxidative removal of H 2 S by polyoxometalates. For dynamic removal of H 2 S, the better the absorption of H 2 S by the ionic liquid, the more reaction time the polyoxometalates dissolved in the ionic liquid have to oxidize the dissolved H 2 S. Factors such as the type of anion and cation, density, alkalinity, and hydrogen bond of the ionic liquid can affect the absorption performance of H 2 S by the ionic liquid [24]. Jalili et al. [13,24,25] investigated the solubility of H 2 S in ionic liquids formed by the cation [Hoemin] and the anions Tf 2 N − , OTF − , PF 6 − , and BF 4 − . They found that the order of the solubility of H 2 S in these four ionic liquids was opposite to the order of the density of the ionic liquids. As the density of the ionic liquid increases, the volume of the ionic liquid (or its free volume) decreases, thus reducing the potential for acidic gases such as H 2 S to come into contact with the ionic liquid, and so the gas solubility decreases. Huang et al. [26] studied the H 2 S absorption in imidazole ionic liquids with carboxylate as an anion, such as lactate (Lac), acetate (Ac), and propionate (Pro). The carboxylate is somewhat basic and forms a strong acid-base and hydrogen bond with H atoms of H 2 S, so this ionic liquid has a high absorption capacity for H 2 S. Under the same conditions, the order of absorption of H 2 S by ionic liquids with anions is Lac < Ac < Pro, which is consistent with the basic order of the anions. In the present work, we investigated the absorption performance of four ionic liquids by themselves (without polyoxometalates) for H 2 S, as shown in Figure 6a. The order of absorption of OAc. The pKa values for acetic acid (HOAc), morpholinoethanesulfonic acid (Mes), and carbonic acid (H 2 CO 3 ) are 4.74, 6.1, and 6.37, respectively. According to the conjugate acid-base theory, the order of alkalinity of these three anions is OAc − < Mes − < HCO 3 − . It does not appear that the hydrogen bonding theory can be fully used to explain the absorption of H 2 S, as the H 2 S absorption performance does not coincide with the order of alkalinity of the anions. The underlying mechanisms affecting the absorption performance of these four ionic liquids will need to be investigated in future work. Figure 6b shows the H 2 S removal performance of (n-Bu 4 N) 3  OAc solution exhibited the best desulfurization performance, maintaining more than 98% desulfurization performance within 420 min. Comparing Figure 6a,b, it is found that the H 2 S removal performance of the polyoxometalate/ionic liquid is significantly improved after adding the polyoxometalate, indicating that the polyoxometalate (n-Bu 4 N) 3 VMo 12 O 40 plays the main role in desulfurization. At the same time, it can also be found that the order of the H 2 S removal performance of the (n-Bu 4 N) 3 VMo 12 O 40 /ionic liquid is very similar to that of the single ionic liquid. The difference in the H 2 S removal performance of the polyoxometalate dissolved in different ionic liquids may be caused by the difference in the solubility of H 2 S in ionic liquids.
[Bmim]OAc was selected as the solvent for the polyoxometalate for the next experiment. The effect of polyoxometalate concentration on H 2 S removal performance is shown in Figure 6c. When the concentration of (n-Bu 4 N) 3 VMo 12 O 40 was 0.005 mol/L, the desulfurization solution could maintain a desulfurization efficiency of more than 99% within 390 min, and the desulfurization efficiency gradually decreased after 390 min. The desulfurization efficiency increases gradually with the increase in the concentration of polyoxometalates. Comparing the desulfurization curves of 0 and 0.005 mol/L polyoxometalates, it can be seen that the polyoxometalate plays a major role in the desulfurization process. In the following experiments, the concentration of the polyoxometalate was chosen to be 0.005 mol/L. Figure 6d shows the effect of H 2 S concentration on desulfurization performance. It can be seen from the figure that the removal efficiency of H 2 S decreases with the increase in H 2 S concentration. Within 180 min, the desulfurization solution can maintain 100% removal efficiency under different H 2 S concentrations because the amount of polyoxometalate is sufficient to oxidize H 2 S within this period. After 180 min, the desulfurization efficiency began to decrease. The higher the H 2 S concentration, the faster the desulfurization efficiency decreased. This is because the higher the amount of H 2 S introduced, the faster the consumption of the polyoxometalate. In short, the lower the H 2 S concentration, the better the H 2 S removal efficiency. The effect of reaction temperature on H 2 S removal performance is shown in Figure 6e. When the reaction temperature increased from 90 • C to 150 • C, the desulfurization efficiency increased accordingly. When the reaction temperature continued to increase to 180 • C, the desulfurization efficiency decreased. The temperature has two effects on H 2 S removal. On the one hand, increasing the temperature is beneficial to reducing the viscosity of the ionic liquid and thus speeding up the mass transfer rate, and at the same time, the high temperature is also beneficial to speeding up the reaction rate. On the other hand, since the reaction between the polyoxometalate and H 2 S is exothermic, high temperature is not conducive to the progress of the reaction [5]. The two influences jointly determine the H 2 S removal efficiency. In this experiment, the desulfurization solution showed the best desulfurization performance at 150 • C. Therefore, 150 • C was chosen as the optimal reaction temperature. oxometalates decreases, for example, SMo12O40 2− > PMo12O40 3− > SiMo12O40 4− , PW12O40 3− > SiW12O40 4− ≈ GeW12O40 4− > BW12O40 5− [22,23]. As shown in Figure 5, we used [Bmim]Cl as a solvent to investigate the effect of central atom species on desulfurization performance. It can be seen that the central atom has a great influence on the desulfurization performance of polyoxometalates. The desulfurization properties of polyoxometalates with different central atoms are V > P > Si, which matches the theoretical oxidation potential polyoxometalates (VMo12O40 3− > PMo12O40 3− > SiMo12O40 4− ). The (n-Bu4N)3VMo12O40/[Bmim]Cl solution can maintain a desulfurization rate of more than 92% within 300 min. Therefore, (n-Bu4N)3VMo12O40 was selected as the solute for the next experiment. The absorption of H2S by ionic liquids may affect the oxidative removal of H2S by polyoxometalates. For dynamic removal of H2S, the better the absorption of H2S by the ionic liquid, the more reaction time the polyoxometalates dissolved in the ionic liquid have to oxidize the dissolved H2S. Factors such as the type of anion and cation, density, alkalinity, and hydrogen bond of the ionic liquid can affect the absorption performance of H2S by the ionic liquid [24]. Jalili et al. [13,24,25] investigated the solubility of H2S in ionic liquids formed by the cation [Hoemin] and the anions Tf2N − , OTF − , PF6 − , and BF4 − . They found that the order of the solubility of H2S in these four ionic liquids was opposite to the order of the density of the ionic liquids. As the density of the ionic liquid increases, the volume of the ionic liquid (or its free volume) decreases, thus reducing the potential for acidic gases such as H2S to come into contact with the ionic liquid, and so the gas solubility decreases. Huang et al. [26] studied the H2S absorption in imidazole ionic liq-

Regeneration Performance
The easy regeneration of the desulfurization solution is very important for the desulfurization performance and economic cost of the whole desulfurization process. In this work, the regeneration of the desulfurization solution can be realized only by introducing air into the desulfurization solution after the desulfurization is completed. The effects of regeneration temperature and multiple desulfurization cycles on regeneration performance were investigated. The effect of regeneration temperature on the regeneration performance of the desulfurization system is shown in Figure 7a. The vertical axis represents the percentage of the amount of H 2 S removed to the total amount of H 2 S introduced within 10 h. It can be seen from the figure that the regeneration temperature has little effect on the regeneration performance. At 30 • C, the desulfurization solution can be easily regenerated, and the desulfurization efficiency after regeneration can reach 97.8%. When the regeneration temperature is 60 • C and 120 • C, the regeneration performance reaches the maximum value of 98.4%. From the perspective of energy saving, the lower the regeneration temperature, the better the energy-saving effect. Therefore, 60 • C was chosen as the regeneration temperature in this experiment. The multiple regeneration performance of (n-Bu 4 N) 3 Figure 7b. After four cycles of use, the desulfurization efficiency within 10 h can be maintained above 98.6%. It can be seen that (n-Bu 4 N) 3

VMo 12 O 40 /[Bmim]OAc is an efficient and easily regenerated desulfurization solution.
Molecules 2022, 27, 6723 7 of 14 efficiency. In this experiment, the desulfurization solution showed the best desulfurization performance at 150 °C . Therefore, 150 °C was chosen as the optimal reaction temperature.  easily regenerated, and the desulfurization efficiency after regeneration can reach 97.8%. When the regeneration temperature is 60 °C and 120 °C , the regeneration performance reaches the maximum value of 98.4%. From the perspective of energy saving, the lower the regeneration temperature, the better the energy-saving effect. Therefore, 60 °C was chosen as the regeneration temperature in this experiment. The multiple regeneration performance of (n-Bu4N)3VMo12O40/[Bmim]OAc is shown in Figure 7b. After four cycles of use, the desulfurization efficiency within 10 h can be maintained above 98.6%. It can be seen that (n-Bu4N)3VMo12O40/[Bmim]OAc is an efficient and easily regenerated desulfurization solution.

Desulphurisation Product
As shown in Figure 8, the XRD pattern of the obtained desulfurization product is in good agreement with the sulfur standard card (PDF#08-0247), indicating that the obtained desulfurization product is elemental sulfur. The characteristic peaks at 23

Desulphurisation Product
As shown in Figure 8, the XRD pattern of the obtained desulfurization product is in good agreement with the sulfur standard card (PDF#08-0247), indicating that the obtained desulfurization product is elemental sulfur. The characteristic peaks at 23

Mechanism
To explore the mechanism of desulfurization and regeneration, FT-IR and XPS were used to characterize the samples before and after the reaction and regeneration (Figure 9). The FT-IR of the samples before H2S absorption and before and after air regeneration are shown in Figure 9a. The characteristic peaks exist stably before and after the reaction and regeneration, indicating that the polyoxometalate also maintains the structure stability at high temperature. The V-O bond at 1116 cm −1 shifted after absorption; after regeneration, the V-O bond recovered to that before absorption [21]. This indicates that the central atom V is involved in the oxidation of H2S by the polyoxometalate. The Mo=Od bond at 951 cm −1 and the Mo-Ob-Mo bond at 846 cm −1 did not shift [16][17][18]. Figure 9b shows the XPS spectra of Mo 3d orbitals of the samples before absorption and before and after regeneration. The binding energy of Mo 3d5/2 of the sample before absorption is 232.73 eV, indicating that Mo atoms exist in the compound with a valence of +6 [27]. However, after absorbing H2S, the binding energy of Mo 3d5/2 becomes 229.52 eV, indicating that the Mo atom becomes +4 valence. After air regeneration, the binding energy changed to 233.12 eV, indicating that Mo atoms recovered from +4 to +6 valence [28]. Therefore, the characterization results clearly show that Mo(+VI) in the polyoxometalate is reduced to Mo(+IV) by H2S, and then Mo(+IV) is oxidized to Mo(+VI) by oxygen. The XPS spectra of the V 2p orbitals of the samples before absorption and before and after regeneration are shown in Figure 9c. The binding energy of V 2p3/2 of the sample before absorption was 517.79 eV, indicating that the V atom exists in the compound with a valence of +5 [29]. After absorbing H2S, the binding energy of V 2p3/2 becomes 516.08 eV, indicating that the V atom becomes +4 valence [30]. After air regeneration, the absorption peak of V 2p3/2 was not observed, which may be caused by the weak absorption peak intensity of V 2p3/2. Although the absorption peak of V 2p3/2 after regeneration was not observed, the FT-IR Figure 8. XRD spectra of desulphurization product and sulfur standard card.

Mechanism
To explore the mechanism of desulfurization and regeneration, FT-IR and XPS were used to characterize the samples before and after the reaction and regeneration (Figure 9). The FT-IR of the samples before H 2 S absorption and before and after air regeneration are shown in Figure 9a. The characteristic peaks exist stably before and after the reaction and regeneration, indicating that the polyoxometalate also maintains the structure stability at high temperature. The V-O bond at 1116 cm −1 shifted after absorption; after regeneration, the V-O bond recovered to that before absorption [21]. This indicates that the central atom V is involved in the oxidation of H 2 S by the polyoxometalate. The Mo=O d bond at 951 cm −1 and the Mo-O b -Mo bond at 846 cm −1 did not shift [16][17][18]. Figure 9b shows the XPS spectra of Mo 3d orbitals of the samples before absorption and before and after regeneration. The binding energy of Mo 3d 5/2 of the sample before absorption is 232.73 eV, indicating that Mo atoms exist in the compound with a valence of +6 [27]. However, after absorbing H 2 S, the binding energy of Mo 3d 5/2 becomes 229.52 eV, indicating that the Mo atom becomes +4 valence. After air regeneration, the binding energy changed to 233.12 eV, indicating that Mo atoms recovered from +4 to +6 valence [28]. Therefore, the characterization results clearly show that Mo(+VI) in the polyoxometalate is reduced to Mo(+IV) by H 2 S, and then Mo(+IV) is oxidized to Mo(+VI) by oxygen. The XPS spectra of the V 2p orbitals of the samples before absorption and before and after regeneration are shown in Figure 9c. The binding energy of V 2p 3/2 of the sample before absorption was 517.79 eV, indicating that the V atom exists in the compound with a valence of +5 [29]. After absorbing H 2 S, the binding energy of V 2p 3/2 becomes 516.08 eV, indicating that the V atom becomes +4 valence [30]. After air regeneration, the absorption peak of V 2p 3/2 was not observed, which may be caused by the weak absorption peak intensity of V 2p 3/2 . Although the absorption peak of V 2p 3/2 after regeneration was not observed, the FT-IR characterization results (Figure 9a) clearly showed that after air regeneration, the V-O bond returned to the state before absorption. Therefore, according to the characterization results of XPS and FT-IR, it can be proved that V(+V) in the polyoxometalate is reduced to V(+IV) by H 2 S, and then V(+IV) is oxidized to V(+V) by oxygen. The XPS spectrum of the S 2p orbital of the sample after H 2 S absorption is shown in Figure 9d. After absorbing H 2 S, the absorption peak at the binding energy of the S 2p orbital in the sample is 161.3 eV, which is assigned to S(-II) [31]. The binding energy absorption peaks at 163.3 eV and 168.15 eV belong to elemental sulfur (S 0 ) and S(+VI), respectively [31][32][33]. This characterization suggests that H 2 S is oxidized to elemental sulfur and sulfate by the polyoxometalate.  Figure 9d. After absorbing H2S, the absorption peak at the binding energy of the S 2p orbital in the sample is 161.3 eV, which is assigned to S(-II) [31]. The binding energy absorption peaks at 163.3 eV and 168.15 eV belong to elemental sulfur (S 0 ) and S(+VI), respectively [31][32][33]. This characterization suggests that H2S is oxidized to elemental sulfur and sulfate by the polyoxometalate.    China). The reagents were all analytically pure. All chemicals were used as received.

Synthesis of Polyoxometalates and Ionic Liquids
(n-Bu 4 N) 3 VMo 12 O 40 was synthesized as described previously [18]. A total of 6.1 g of Na 2 MoO 4 ·2H 2 O was dissolved in 170 mL of distilled water, and 0.3 g of NH 4 VO 3 was dissolved in 26 mL of concentrated hydrochloric acid. A total of 300 mL of acetonitrile was added to the mixed solution, and the solution changed from pale yellow to orange. After stirring at room temperature for 30 min, 5 g of n-Bu 4 NBr was added to produce a yellow precipitate, which was filtered, washed with water and ethanol, and air-dried to obtain (n-Bu 4 N) 3 VMo 12 O 40 .
(n-Bu 4 N) 3 PMo 12 O 40 was synthesized by the following process [34]. Mix 120 mL of 1 mol·L −1 sodium molybdate solution, 18 mL 13 mol·L −1 HNO 3 solution, and 100 mL of 1,4-dioxane solution in a 500-milliliter beaker. A total of 10 mL of 1 mol·L −1 H 3 PO 4 solution was added to the mixture while stirring. A yellow precipitate was produced after adding 10 mL of a solution containing 10 g of n-Bu 4 NBr to the above-mixed solution. The resulting precipitate was filtered and treated with 100 mL of boiling water, then filtered again, and finally washed with distilled water, ethanol, and ether, respectively, and recrystallized in acetone. The resulting crystals were dried in the air, and the resulting product was (n-Bu 4 N) 3 PMo 12 O 40 .
(n-Bu 4 N) 4 [α-SiMo 12 O 40 ] was synthesized by the following process [34]. A total of 37 mL of 13 mol·L −1 concentrated nitric acid solution was added to 120 mL of 1 mol·L −1 sodium molybdate solution. Then, 50 mL of 0.2 mol·L −1 sodium silicate solution was added dropwise to the above solution; the solution turned yellow, and [β-SiMo 12  [Bmim]HCO 3 was synthesized by the following process [35].
[Bmim]Cl (0.3 mol, 52.4 g) was dissolved in 50 mL dichloromethane; then, NaHCO 3 (0.36 mol, 30.24 g) was added. The mixture was stirred and refluxed at 45 • C for 24 h. After cooling to room temperature, the precipitate was removed by filtration. After the filtrate was removed by rotary evaporation at 60 • C, dichloromethane was added again. After standing at 4 • C for 6 h, the salt was removed by suction filtration, and the filtrate was rotary evaporated. After several times of suction filtration to remove salt until no salt was analyzed, the obtained light-yellow liquid was [Bmim]HCO 3 .
[Bmim]Mes was synthesized by the following process [36]. NaOH (0.12 mol, 4.8 g), morpholineethylsulfonic acid (0.1 mol, 21.3 g), and [Bmim]Cl (0.1 mol, 17.45 g) were dissolved in 200 mL of water. After stirring for 1 h at room temperature, the solvent was evaporated and a yellow-green viscous liquid was produced. A total of 500 mL of dichloromethane was added to the solution and stirred well to produce a white solid and a yellow-green liquid. The white solid was removed by filtration. The filtrate was evaporated at 60 • C to remove the solvent and then dried under vacuum to obtain [Bmim]Mes.

H 2 S Removal, Desulfurization Solution Regeneration, and Desulfurization Product Recovery
H 2 S removal: This experiment was carried out under normal pressure. The desulfurization solution used was obtained by dissolving a certain amount of polyoxometalate in a certain volume (10 mL) of ionic liquid. The resulting desulfurized solution was transferred to a quartz reactor. A mixed gas containing H 2 S and N 2 was passed through the reactor at a flow rate of 100 mL/min. The reaction temperature was controlled by a constant temperature oil bath. The H 2 S concentration at the outlet of the reactor was measured by a TH-990 hydrogen sulfide analyzer. The exhaust gas was absorbed by supersaturated sodium hydroxide solution and discharged into air. The H 2 S removal efficiency was calculated by the following formula: where C 0 (H 2 S) and C(H 2 S) are the H 2 S concentration of inlet and outlet gases, respectively, (mg/m 3 ). Desulfurization solution regeneration: After the reaction, 800 mL/min of air was introduced into the desulfurization solution for 12 h, and the regeneration was completed. The desulfurization solution after regeneration continued to undergo desulfurization experiments.
Desulfurization product recovery: After the desulfurization reaction was completed, water was added to the desulfurization solution. After standing, the supernatant was discarded, and the ionic liquid was removed with the supernatant. Carbon disulfide (CS 2 ) was added to the remaining solution for extraction of the desulfurization product. The sample obtained after drying was used for XRD characterization.

Conclusions
In the present work, three polyoxometalates with different central atoms ((n-Bu 4 N) 3