**Reaction Mechanism of Simultaneous Removal of H2S and PH<sup>3</sup> Using Modified Manganese Slag Slurry**

**Jiacheng Bao <sup>1</sup> , Xialing Wang <sup>1</sup> , Kai Li <sup>1</sup> , Fei Wang <sup>1</sup> , Chi Wang <sup>2</sup> , Xin Song <sup>1</sup> , Xin Sun 1,\* and Ping Ning 1,\***


Received: 25 October 2020; Accepted: 26 November 2020; Published: 27 November 2020

**Abstract:** The presence of phosphine (PH3) and hydrogen sulfide (H2S) in industrial tail gas results in the difficulty of secondary utilization. Using waste solid as a wet absorbent to purify the H2S and PH<sup>3</sup> is an attractive strategy with the achievement of "waste controlled by waste". In this study, the reaction mechanism of simultaneously removing H2S and PH<sup>3</sup> by modified manganese slag slurry was investigated. Through the acid leaching method for raw manganese slag and the solid–liquid separation subsequently, the liquid-phase part has a critical influence on removing H2S and PH3. Furthermore, simulation experiments using metal ions for modified manganese slag slurry were carried out to investigate the effect of varied metal ions on the removal of H2S and PH3. The results showed that Cu2<sup>+</sup> and Al3<sup>+</sup> have a promoting effect on H2S and PH<sup>3</sup> conversion. In addition, the Cu2<sup>+</sup> has liquid-phase catalytic oxidation for H2S and PH<sup>3</sup> through the conversion of Cu(II) to Cu(I).

**Keywords:** phosphine; hydrogen sulfide; manganese slag; metal ions; reaction mechanism

## **1. Introduction**

The presence of phosphine (PH3) and hydrogen sulfide (H2S) in industrial gases can result in reduced feed gas quality, excessive equipment corrosion, and catalyst deactivation and poisoning, limiting industrial gas recovery and utilization [1], especially yellow phosphorus off-gas. In addition, H2S, as a highly toxic, corrosive gas, can not only cause air pollution but also eye irritation and breathing problems. Even exposure to small amounts of H2S will pose a serious threat to humans [2]. Additionally, PH<sup>3</sup> may cause immediate death if one is exposed to a concentration level of 50 ppm, according to the National Institute for Occupational Safety and Health (NIOSH) [3]. Therefore, it is desirable to remove them from the point of view of the highly efficient use of industrial gases and human health.

Currently, the wet process has more advantages for simultaneously removing H2S and PH<sup>3</sup> compared to the dry process, due to lower cost and easier preparation process [4,5]. However, the wet process for the removal of H2S and PH<sup>3</sup> could rapidly consume oxidant, thus leading to a decrease in removal efficiency [4]. Hence, there is an urgent need to modify the traditional wet method to meet stricter environmental laws. In recent years, the use of metal ore, tailings, and metal smelting slag to remove industrial waste gas has attracted the attention of researchers [6–8]. Smelting slag contains a large number of transition metals such as Fe, Mn, and basic oxides, which could have been favorable for absorption of PH<sup>3</sup> and H2S in the wet process due to the liquid phase catalytic oxidation ability of transition metals and higher alkalinity of basic oxides such as CaO, MgO, etc. The liquid catalytic

oxidation by transition metals can be favorable for the oxidation of gas pollutants, thus being conducive to absorbing H2S or/and PH<sup>3</sup> [9–11]. At present, China has become the largest manganese producer, consumer, and exporter, and has emitted more than 2 <sup>×</sup> <sup>10</sup><sup>6</sup> tons of manganese slag (MS) every year, which mainly originate from the manganese metallurgy process. Piling manganese slag waste has caused color stain and pollution in the composition of soil and surface water and contaminated groundwater and rivers due to surface runoff [12]. Therefore, there is a pressing need to dispose of MS in a green and highly efficient manner. The chemical compositions of MS are rich in Mn-based oxides and contain a number of basic oxides, which make it possible to become a highly efficient absorbent for the removal of H2S and PH3.

According to our previous studies [13], modified manganese slag (MS) slurry was used to remove H2S and PH<sup>3</sup> and proved a great absorbent. However, a detailed investigation of the reaction mechanism was still lacking. Thus, our research extends the knowledge into the reaction mechanism of H2S and PH<sup>3</sup> using modified MS slurry. In this paper, the emphasis was laid on the role of modified MS slurry in removing H2S and PH3. In addition, to understand the liquid phase catalytic oxidation ability of modified MS slurry, we simulated the composition of MS slurry in the form of different metal salt solutions based on the actual composition of electrolytic MS. In addition, XPS (X-ray photoelectron spectrometer), IC (ion chromatography), and XRF (X-ray fluorescence) and XRD (X-ray diffraction) techniques were used to investigate the samples' surface information. Thus, a mechanism of simultaneous removal of PH<sup>3</sup> and H2S using modified MS slurry was proposed.

#### **2. Results and Discussion**

## *2.1. E*ff*ect of Di*ff*erent Component Slurry after Acid Leaching on Simultaneous Removal of H2S and PH<sup>3</sup>*

In order to investigate the mechanism of simultaneous removal of H2S and PH<sup>3</sup> using modified MS slurry, the acid leaching method was used to treat the raw MS. Thus, the majority of soluble metal ions can be leached and transferred from the solid phase of raw MS to the aqueous solution. It can be seen from Figure 1 that XRD results show that the main mineral phases of raw MS were 3CaO·SiO<sup>2</sup> (C3A), 2CaO·SiO<sup>2</sup> (C2A), CaO, Al2O3, and merwinite. These components were acid-soluble, which can be decomposed by the acid solution. Table 1 also shows that the main elements in raw MS were Si, Ca, and Mn. Thus, through acid leaching and then filtering, the active components mainly including metal *Catalysts* **2020** ions could be removed from the raw MS. , *10*, x FOR PEER REVIEW 3 of 12

**Figure 1.** XRD pattern of raw manganese slag. **Figure 1.** XRD pattern of raw manganese slag.

**Figure 2.** Effects of different component slurry after acid leaching on simultaneous removal of (**a**) H2S and (**b**) PH3 by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction

The metal ions leached from MS during the reaction period played a leading role in removing

H2S and PH3. Thus, to gain more insight, the modified MS slurry was simulated in the form of metal salts based on the real component composition. The various simulated metal salts species (including metal nitrates, metal chlorides, and metal sulfates) were investigated and the simulated contents were listed in Table 2. The H2S and PH3 removal efficiency by simulated modified MS slurry is shown in Figure 3. The results showed that the three groups obtained a 100% H2S conversion efficiency; whereas Group 3 (metal chlorides) obtained the higher PH3 removal efficiency relative to those of the other groups, with the order being metal chlorides > metal nitrate > metal sulfates. In order to achieve a better understanding of the variation in PH3 and H2S conversion by different metal salts, the reaction products in aqueous solutions were detected by the IC method. The IC results, as shown in Figure 4, showed that the generated PO3− and SO42− concentrations in Group 3 increased with the reaction proceeding, which indicated that the H2S and PH3 could be converted to PO3− and SO42−, respectively. Thus, the addition of metal sulfates inhibited higher H2S and PH3 conversion efficiency because the generation of SO42− was accelerated and inhibited the oxidation of PH3 when the copper concentration was at relatively high concentration, according to our previous study [13]. In other words, more sulfates existing in the solution led to lower PH3 conversion efficiency. Hence, the metal

temperature = 35 °C; Oxygen content = 1 vol %; stirring rate = 800 r/min.

chloride was chosen for the following experiments.

*2.2. Effect of Simulated Modified MS Slurry on Simultaneous Removal of H2S and PH3*



As shown in Figure 2, the 100% H2S removal efficiency of modified MS slurry (MS + CuSO<sup>4</sup> group) can maintain 7 h while the 65% PH<sup>3</sup> removal efficiency of modified MS slurry can maintain 5 h. In addition, the acid leaching residues and CuSO<sup>4</sup> slurry can only obtain around 7% H2S removal efficiency and approximately 9% PH<sup>3</sup> removal efficiency. Compared with the simultaneous removal of H2S and PH<sup>3</sup> by modified MS slurry, the acid leaching residue and CuSO<sup>4</sup> slurry obtained a poorer removal efficiency due to the consumption of active components after the acid leaching method. Additionally, the raw MS and Cu2<sup>+</sup> have a synergistic effect on the removal of H2S and PH3. Thus, leached metal ions in slurry have a leading role in removing H2S and PH3. The mixture of MS and CuSO<sup>4</sup> slurry obtained the best PH<sup>3</sup> conversion efficiency while the effect of CuSO<sup>4</sup> solution was not obvious. According to our previous report [13], the increase in PH<sup>3</sup> conversion efficiency can be ascribed to the oxidation ability of Cu2<sup>+</sup> and Mn2<sup>+</sup> from MS, thus inhibiting the formation of CuS/Cu2S. **Figure 1.** XRD pattern of raw manganese slag.

**Figure 2.** Effects of different component slurry after acid leaching on simultaneous removal of (**a**) H2S and (**b**) PH3 by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction temperature = 35 °C; Oxygen content = 1 vol %; stirring rate = 800 r/min. **Figure 2.** Effects of different component slurry after acid leaching on simultaneous removal of (**a**) H2S and (**b**) PH<sup>3</sup> by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH<sup>3</sup> concentration = 400 ppm; gas flow rate = 110 mL min−<sup>1</sup> ; reaction temperature = 35 ◦C; Oxygen content = 1 vol %; stirring rate = 800 r/min.

#### *2.2. Effect of Simulated Modified MS Slurry on Simultaneous Removal of H2S and PH3 2.2. E*ff*ect of Simulated Modified MS Slurry on Simultaneous Removal of H2S and PH<sup>3</sup>*

The metal ions leached from MS during the reaction period played a leading role in removing H2S and PH3. Thus, to gain more insight, the modified MS slurry was simulated in the form of metal salts based on the real component composition. The various simulated metal salts species (including metal nitrates, metal chlorides, and metal sulfates) were investigated and the simulated contents were listed in Table 2. The H2S and PH3 removal efficiency by simulated modified MS slurry is shown in Figure 3. The results showed that the three groups obtained a 100% H2S conversion efficiency; whereas Group 3 (metal chlorides) obtained the higher PH3 removal efficiency relative to those of the other groups, with the order being metal chlorides > metal nitrate > metal sulfates. In order to achieve a better understanding of the variation in PH3 and H2S conversion by different metal salts, the reaction products in aqueous solutions were detected by the IC method. The IC results, as shown in Figure 4, showed that the generated PO3− and SO42− concentrations in Group 3 increased with the reaction proceeding, which indicated that the H2S and PH3 could be converted to PO3− and SO42−, respectively. Thus, the addition of metal sulfates inhibited higher H2S and PH3 conversion efficiency because the generation of SO42− was accelerated and inhibited the oxidation of PH3 when the copper The metal ions leached from MS during the reaction period played a leading role in removing H2S and PH3. Thus, to gain more insight, the modified MS slurry was simulated in the form of metal salts based on the real component composition. The various simulated metal salts species (including metal nitrates, metal chlorides, and metal sulfates) were investigated and the simulated contents were listed in Table 2. The H2S and PH<sup>3</sup> removal efficiency by simulated modified MS slurry is shown in Figure 3. The results showed that the three groups obtained a 100% H2S conversion efficiency; whereas Group 3 (metal chlorides) obtained the higher PH<sup>3</sup> removal efficiency relative to those of the other groups, with the order being metal chlorides > metal nitrate > metal sulfates. In order to achieve a better understanding of the variation in PH<sup>3</sup> and H2S conversion by different metal salts, the reaction products in aqueous solutions were detected by the IC method. The IC results, as shown in Figure 4, showed that the generated PO<sup>3</sup> <sup>−</sup> and SO<sup>4</sup> <sup>2</sup><sup>−</sup> concentrations in Group 3 increased with the reaction proceeding, which indicated that the H2S and PH<sup>3</sup> could be converted to PO<sup>3</sup> <sup>−</sup> and SO<sup>4</sup> <sup>2</sup>−, respectively. Thus, the addition of metal sulfates inhibited higher H2S and PH<sup>3</sup> conversion efficiency because the generation of SO<sup>4</sup> <sup>2</sup><sup>−</sup> was accelerated and inhibited the oxidation of PH<sup>3</sup> when the copper concentration

was at relatively high concentration, according to our previous study [13]. In other words, more sulfates existing in the solution led to lower PH<sup>3</sup> conversion efficiency. Hence, the metal chloride was chosen for the following experiments. *Catalysts* **2020**, *10*, x FOR PEER REVIEW 4 of 12 **Table 2.** Composition of simulated modified manganese slag slurry. *Catalysts* **2020**, *10*, x FOR PEER REVIEW 4 of 12


**Table 2.** Composition of simulated modified manganese slag slurry. **Table 2.** Composition of simulated modified manganese slag slurry.

**Figure 3.** Effect of different simulated modified manganese slag (MS) slurry on simultaneous removal of H2S and PH3 by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction **Figure 3.** Effect of different simulated modified manganese slag (MS) slurry on simultaneous removal of H2S and PH<sup>3</sup> by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH<sup>3</sup> concentration = 400 ppm; gas flow rate = 110 mL min−<sup>1</sup> ; reaction temperature = 35 ◦C; Oxygen content = 1 vol %; stirring rate = 800 r/min. **Figure 3.** Effect of different simulated modified manganese slag (MS) slurry on simultaneous removal of H2S and PH3 by manganese slag before and after acid leaching. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction temperature = 35 °C; Oxygen content = 1 vol %; stirring rate = 800 r/min.

**Figure 4.** Ion concentration in Group 3 (metal chloride) as a function of time (50× dilution). **Figure 4.** Ion concentration in Group 3 (metal chloride) as a function of time (50× dilution).

5 that all groups can achieve 100% H2S removal efficiency. The highest PH3 conversion efficiency of Cu2+ + Al3+, Cu2+ + Ca2+, Cu2+ + Mg2+, and Cu2+ + Mn2+ were 95.52%, 89.45%, 87.43%, and 81.63%, respectively. The Al3+ + Cu2+ group obtained higher PH3 conversion efficiency relative to that of Cu2+ alone, which indicated that the Al3+ and Cu2+ have a synergistic effect on PH3 removal. The addition of Mg2+, Ca2+, and Mn2+ slightly reduced the PH3 conversion efficiency compared to those of the Cu2+ group and Al3+ + Cu2+ group. Thus, the analysis emphasis was laid on the Al3+ + Cu2+ group, and the

experiments to examine the effect of metal ions on H2S and PH3 removal. It can be seen from Figure 5 that all groups can achieve 100% H2S removal efficiency. The highest PH3 conversion efficiency of Cu2+ + Al3+, Cu2+ + Ca2+, Cu2+ + Mg2+, and Cu2+ + Mn2+ were 95.52%, 89.45%, 87.43%, and 81.63%, respectively. The Al3+ + Cu2+ group obtained higher PH3 conversion efficiency relative to that of Cu2+ alone, which indicated that the Al3+ and Cu2+ have a synergistic effect on PH3 removal. The addition of Mg2+, Ca2+, and Mn2+ slightly reduced the PH3 conversion efficiency compared to those of the Cu2+ group and Al3+ + Cu2+ group. Thus, the analysis emphasis was laid on the Al3+ + Cu2+ group, and the

reaction products in long-term experiments are analyzed in the next section (Section 2.4).

reaction products in long-term experiments are analyzed in the next section (Section 2.4).

To further explore the role of metal ions in removing H2S and PH3, we provide a series of

**Figure 4.** Ion concentration in Group 3 (metal chloride) as a function of time (50× dilution).

*2.3. Effect of Single and Multi-Metal Ions on Simultaneous Removal of H2S and PH3*

decreased dramatically.

## *2.3. E*ff*ect of Single and Multi-Metal Ions on Simultaneous Removal of H2S and PH<sup>3</sup>*

To further explore the role of metal ions in removing H2S and PH3, we provide a series of experiments to examine the effect of metal ions on H2S and PH<sup>3</sup> removal. It can be seen from Figure 5 that all groups can achieve 100% H2S removal efficiency. The highest PH<sup>3</sup> conversion efficiency of Cu2<sup>+</sup> + Al3+, Cu2<sup>+</sup> + Ca2+, Cu2<sup>+</sup> + Mg2+, and Cu2<sup>+</sup> + Mn2<sup>+</sup> were 95.52%, 89.45%, 87.43%, and 81.63%, respectively. The Al3<sup>+</sup> + Cu2<sup>+</sup> group obtained higher PH<sup>3</sup> conversion efficiency relative to that of Cu2<sup>+</sup> alone, which indicated that the Al3<sup>+</sup> and Cu2<sup>+</sup> have a synergistic effect on PH<sup>3</sup> removal. The addition of Mg2+, Ca2+, and Mn2<sup>+</sup> slightly reduced the PH<sup>3</sup> conversion efficiency compared to those of the Cu2<sup>+</sup> group and Al3<sup>+</sup> + Cu2<sup>+</sup> group. Thus, the analysis emphasis was laid on the Al3<sup>+</sup> + Cu2<sup>+</sup> group, and the reaction products in long-term experiments are analyzed in the next section (Section 2.4). *Catalysts* **2020**, *10*, x FOR PEER REVIEW 5 of 12

**Figure 5.** Effect of different single and multi-metal ions (metal chlorides) on simultaneous removal of H2S and PH3. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction temperature = 35 °C; Oxygen content = 1 vol %; stirring rate = **Figure 5.** Effect of different single and multi-metal ions (metal chlorides) on simultaneous removal of H2S and PH<sup>3</sup> . Experimental conditions: H2S concentration = 800 ppm; PH<sup>3</sup> concentration = 400 ppm; gas flow rate = 110 mL min−<sup>1</sup> ; reaction temperature = 35 ◦C; Oxygen content = 1 vol %; stirring rate = 800 r/min.

#### 800 r/min. *2.4. Reaction Mechanism of Metal Ions to Simultaneous Removal of H2S and PH<sup>3</sup>*

*2.4. Reaction Mechanism of Metal Ions to Simultaneous Removal of H2S and PH3* To understand the effect of Al3+ combined with Cu2+ on simultaneous removal of H2S and PH3, the long-term experiments were carried out and the variation in the pH value of solution, solid or aqueous products is analyzed in this section. For the Cu2+ group, Figure 6a shows that the PH3 conversion efficiency increased initially and then decreased with the reaction proceeding, accompanied by a gradual decrease in the pH value. The XRD results in Figure 6b indicate that the main reaction product was CuS, Cu8S5, and CuCl, which resulted from the reaction between H2S and Cu2+ (Equation (1)). Meanwhile, the H+ ion was increased, thereby leading to a decrease in pH value. The Al3+ + Cu2+ group showed a similar conversion trend for H2S and PH3. However, the reaction products of the Al3+ + Cu2+ group became different. It can be seen from Figure 6d that the Al2(SO4)3, To understand the effect of Al3<sup>+</sup> combined with Cu2<sup>+</sup> on simultaneous removal of H2S and PH3, the long-term experiments were carried out and the variation in the pH value of solution, solid or aqueous products is analyzed in this section. For the Cu2<sup>+</sup> group, Figure 6a shows that the PH<sup>3</sup> conversion efficiency increased initially and then decreased with the reaction proceeding, accompanied by a gradual decrease in the pH value. The XRD results in Figure 6b indicate that the main reaction product was CuS, Cu8S5, and CuCl, which resulted from the reaction between H2S and Cu2<sup>+</sup> (Equation (1)). Meanwhile, the H<sup>+</sup> ion was increased, thereby leading to a decrease in pH value. The Al3<sup>+</sup> + Cu2<sup>+</sup> group showed a similar conversion trend for H2S and PH3. However, the reaction products of the Al3<sup>+</sup> + Cu2<sup>+</sup> group became different. It can be seen from Figure 6d that the Al2(SO4)3, CuSO4, Cu4(SO4)(OH)6·2H2O, and CuS were the main reaction products for removing H2S while AlPO3, AlP, and Cu5O2(PO4)<sup>2</sup> were the main reaction products for removing PH3. When Al3<sup>+</sup> was added into the solution, the pH value slowly decreased during 5 h of initial reaction time, which indicated that Al3<sup>+</sup> could inhibit the rapid decline of pH value, although the pH value subsequently decreased dramatically.

$$\text{Cu}^{2+} + \text{H}\_2\text{S} \rightarrow \text{CuS} + 2\text{H}^+ \tag{1}$$

added into the solution, the pH value slowly decreased during 5 h of initial reaction time, which indicated that Al3+ could inhibit the rapid decline of pH value, although the pH value subsequently

of reaction.

valence state of Cu2+ to Cu+.

800 r/min.

**Figure 5.** Effect of different single and multi-metal ions (metal chlorides) on simultaneous removal of H2S and PH3. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction temperature = 35 °C; Oxygen content = 1 vol %; stirring rate =

To understand the effect of Al3+ combined with Cu2+ on simultaneous removal of H2S and PH3, the long-term experiments were carried out and the variation in the pH value of solution, solid or aqueous products is analyzed in this section. For the Cu2+ group, Figure 6a shows that the PH3 conversion efficiency increased initially and then decreased with the reaction proceeding, accompanied by a gradual decrease in the pH value. The XRD results in Figure 6b indicate that the main reaction product was CuS, Cu8S5, and CuCl, which resulted from the reaction between H2S and Cu2+ (Equation (1)). Meanwhile, the H+ ion was increased, thereby leading to a decrease in pH value. The Al3+ + Cu2+ group showed a similar conversion trend for H2S and PH3. However, the reaction products of the Al3+ + Cu2+ group became different. It can be seen from Figure 6d that the Al2(SO4)3, CuSO4, Cu4(SO4)(OH)6·2H2O, and CuS were the main reaction products for removing H2S while AlPO3, AlP, and Cu5O2(PO4)2 were the main reaction products for removing PH3. When Al3+ was added into the solution, the pH value slowly decreased during 5 h of initial reaction time, which indicated that Al3+ could inhibit the rapid decline of pH value, although the pH value subsequently

Cuଶା + HଶS → CuS + 2Hା (1)

*2.4. Reaction Mechanism of Metal Ions to Simultaneous Removal of H2S and PH3*

**Figure 6.** Variation in pH value of (**a**) Cu2+ group and (**c**) Al3+ + Cu2+ group as a function of time. XRD patterns of reaction process of (**b**) Cu2+ group and (**d**) Al3+ + Cu2+ group. Experimental conditions: H2S concentration = 800 ppm; PH3 concentration = 400 ppm; gas flow rate = 110 mL min<sup>−</sup>1; reaction temperature = 35 °C; Oxygen content = 1 vol %; stirring rate = 800 r/min. **Figure 6.** Variation in pH value of (**a**) Cu2<sup>+</sup> group and (**c**) Al3<sup>+</sup> + Cu2<sup>+</sup> group as a function of time. XRD patterns of reaction process of (**b**) Cu2<sup>+</sup> group and (**d**) Al3<sup>+</sup> + Cu2<sup>+</sup> group. Experimental conditions: H2S concentration = 800 ppm; PH<sup>3</sup> concentration = 400 ppm; gas flow rate = 110 mL min−<sup>1</sup> ; reaction temperature = 35 ◦C; Oxygen content = 1 vol %; stirring rate = 800 r/min.

To gain more insight, we conducted several XPS studies for the reaction products of the Cu2+ group and Al3+ + Cu2+ groups with the reaction proceeding. For the Al3+ + Cu2+ group, as shown in Figure 7a, the binding energy (B. E.) centered at about 159.4 eV for S 2p3/2 could be attributed to S2<sup>−</sup> [14], which indicated that CuS was generated in the Al3+ + Cu2+ group. In addition, the B. E. located at 164.2 eV may be ascribed to polysulfide species [15], which indicated that CuS/Cu8S5 was oxidized in the existence of oxygen. With the reaction further proceeding, the sulfate was formed as evidence that the B. E. of 169.8 eV appeared [16] and the sulfate content further increased from 7 h to complete reaction, which indicated that CuS was further oxidized to CuSO4 by oxygen in the presence of the water environment [17]. This was consistent with the XRD results as shown in Figure 6d. For the Cu2+ group, the variation in S valence state in the initial reaction period was similar to the Al3+ + Cu2+ group. However, with further increase of the reaction time, the content of the polysulfide species (53.2%, calculated by XPS data as shown in Table 3) was the same as the sulfate content (46.8%); whereas the sulfate content (91.4%) was dominant in the Al3+ + Cu2+ group. This can be explained by noting that the addition of Al3+ could effectively accelerate the CuS oxidation, thus generating more sulfate species. To gain more insight, we conducted several XPS studies for the reaction products of the Cu2<sup>+</sup> group and Al3<sup>+</sup> + Cu2<sup>+</sup> groups with the reaction proceeding. For the Al3<sup>+</sup> + Cu2<sup>+</sup> group, as shown in Figure 7a, the binding energy (B. E.) centered at about 159.4 eV for S 2p3/<sup>2</sup> could be attributed to S <sup>2</sup><sup>−</sup> [14], which indicated that CuS was generated in the Al3<sup>+</sup> + Cu2<sup>+</sup> group. In addition, the B. E. located at 164.2 eV may be ascribed to polysulfide species [15], which indicated that CuS/Cu8S<sup>5</sup> was oxidized in the existence of oxygen. With the reaction further proceeding, the sulfate was formed as evidence that the B. E. of 169.8 eV appeared [16] and the sulfate content further increased from 7 h to complete reaction, which indicated that CuS was further oxidized to CuSO<sup>4</sup> by oxygen in the presence of the water environment [17]. This was consistent with the XRD results as shown in Figure 6d. For the Cu2<sup>+</sup> group, the variation in S valence state in the initial reaction period was similar to theAl<sup>3</sup><sup>+</sup> <sup>+</sup> Cu2<sup>+</sup> group. However, with further increase of the reaction time, the content of the polysulfide species (53.2%, calculated by XPS data as shown in Table 3) was the same as the sulfate content (46.8%); whereas the sulfate content (91.4%) was dominant in the Al3<sup>+</sup> + Cu2<sup>+</sup> group. This can be explained by noting that the addition of Al3<sup>+</sup> could effectively accelerate the CuS oxidation, thus generating moresulfate species.

It can be seen from Figure 7e that the B. E. located at around 934.9 and 932.6 eV in the Al3+ + Cu2+ group could be attributed to Cu(II) and Cu(I), respectively [18]. The relative Cu2+ content in the Al3+ + Cu2+ group was 83.5% for 1 h of reaction time and then decreased to 73.6% until the complete reaction, which indicated that part of Cu2+ was converted to Cu+ as evidenced by the formation of CuCl and Cu8S5 by the XRD technique. From Figure 7f, it can be seen that the ratio of Cu(II)/Cu(I) in the Cu2+ group decreased dramatically relative to the Al3+ + Cu2+ group, which indicated that Cu2+ ion had a liquid-phase catalytic oxidation effect on removing H2S and PH3 through variation in the

As shown in Figure 7c, when the Al3+ + Cu2+ group reacted for 1 h, phosphate was formed, since the peak at 134.6 eV could be attributed to PO43− while the B. E. of 131.0 eV may be ascribed to P3−. With an increase in reaction time, the PO43 content increased from 51.7% to 69.8%, up to 94.4% in the final. The variation of the P valence state in the Cu2+ group as shown in Figure 7d was the same as for the Al3+ + Cu2+ group; but the generation rate of phosphate in the Cu2+ group was slower than that of

*Catalysts* **2020**, *10*, x FOR PEER REVIEW 7 of 12

**Figure 7.** XPS spectra of Al3+ + Cu2+ group and Cu2+ group with different reaction time for surveys of (**a**,**b**) S 2p, (**c**,**d**) P 2p, and (**e**,**f**) Cu 2p. **Figure 7.** XPS spectra of Al3<sup>+</sup> + Cu2<sup>+</sup> group and Cu2<sup>+</sup> group with different reaction time for surveys of (**a**,**b**) S 2p, (**c**,**d**) P 2p, and (**e**,**f**) Cu 2p.


**Table 3.** XPS data of Al3<sup>+</sup> + Cu2<sup>+</sup> group and Cu2<sup>+</sup> group with different reaction time for surveys of S 2p; P 2p, and Cu 2p.

<sup>1</sup> S<sup>n</sup> refers to the polysulfide species.

As shown in Figure 7c, when the Al3<sup>+</sup> + Cu2<sup>+</sup> group reacted for 1 h, phosphate was formed, since the peak at 134.6 eV could be attributed to PO<sup>4</sup> <sup>3</sup><sup>−</sup> while the B. E. of 131.0 eV may be ascribed to P3−. With an increase in reaction time, the PO<sup>4</sup> 3 content increased from 51.7% to 69.8%, up to 94.4% in the final. The variation of the P valence state in the Cu2<sup>+</sup> group as shown in Figure 7d was the same as for the Al3<sup>+</sup> + Cu2<sup>+</sup> group; but the generation rate of phosphate in the Cu2<sup>+</sup> group was slower than that of the Al3<sup>+</sup> + Cu2<sup>+</sup> group. When the reaction time was for 1 h, the relative phosphate content in the Cu2<sup>+</sup> group was 39.9%, thus leading to a slower increased PH<sup>3</sup> conversion efficiency than the Cu2<sup>+</sup> group. The faster conversion of PH<sup>3</sup> to phosphate resulted in accelerated PH<sup>3</sup> absorption in the initial period of reaction.

It can be seen from Figure 7e that the B. E. located at around 934.9 and 932.6 eV in the Al3<sup>+</sup> + Cu2<sup>+</sup> group could be attributed to Cu(II) and Cu(I), respectively [18]. The relative Cu2<sup>+</sup> content in the Al3<sup>+</sup> + Cu2<sup>+</sup> group was 83.5% for 1 h of reaction time and then decreased to 73.6% until the complete reaction, which indicated that part of Cu2<sup>+</sup> was converted to Cu<sup>+</sup> as evidenced by the formation of CuCl and Cu8S<sup>5</sup> by the XRD technique. From Figure 7f, it can be seen that the ratio of Cu(II)/Cu(I) in the Cu2<sup>+</sup> group decreased dramatically relative to the Al3<sup>+</sup> + Cu2<sup>+</sup> group, which indicated that Cu2<sup>+</sup> ion had a liquid-phase catalytic oxidation effect on removing H2S and PH<sup>3</sup> through variation in the valence state of Cu2<sup>+</sup> to Cu+.

Based on the above analysis, the metal ions played a leading role in removing H2S and PH3. The reaction process can be summarized as follows (Equations (2)–(13)):

$$\text{H}\_2\text{S} \leftrightarrow \text{HS}^- + \text{H}^+ \leftrightarrow \text{S}^{2-} + 2\text{H}^+ \tag{2}$$

$$\text{Me}^{2+} + \text{H}\_2\text{S} + 2\text{H}\_2\text{O} \rightarrow \text{MeS} \downarrow + 2\text{H}^+ \quad \left(\text{Me}^{2+} = \text{Cu}^{2+}\right) \tag{3}$$

$$\text{CuS}(\text{s}) \rightarrow \text{Cu}^{2+}(\text{aq}) + \text{S}^{2-} \quad \left(\text{Me}^{2+} = \text{Cu}^{2+}\right) \tag{4}$$

$$8\text{Cu}^{2+} + \text{PH}\_3 + 4\text{H}\_2\text{O} \rightarrow 8\text{Cu}^+ + \text{PO}\_4^{3-} + 11\text{H}^+\tag{5}$$

$$4\text{Cu}^+ + \text{O}\_2 + 2\text{H}\_2\text{O} \to 4\text{Cu}^{2+} + 4\text{OH}^-\tag{6}$$

$$\text{Cu}^+ + \text{O}\_2 \rightarrow \text{Cu}^{2+} + \text{O}\_2^- \tag{7}$$

$$\text{O}\_2 + 2\text{H}\_2\text{S} \to 2\text{S} \downarrow + 2\text{H}\_2\text{O} \tag{8}$$

$$\text{S} + 1.5\text{O}\_2 + 2\text{H}\_2\text{O} \rightarrow \text{SO}\_4^{2-} + 2\text{H}^+ \tag{9}$$

$$(\text{CuS})\_{\text{n}} + \text{O}\_{2} \rightarrow (\text{CuS})\_{\text{n-1}}(\text{CuS})^{\bullet+} + \text{O}\_{2} \text{•}^{\bullet-} \tag{10}$$

$$\text{Cu}^{2+} + \text{H}\_2\text{O} \rightarrow \text{Cu}^+ + \text{H}^+ + \text{OH}^\bullet \tag{11}$$

$$\text{2OH}^{\bullet} \rightarrow \text{H}\_2\text{O}\_2\tag{12}$$

$$4\text{H}\_2\text{O}\_2 + \text{S}^{2-} \rightarrow \text{SO}\_4^{2-} + 4\text{H}\_2\text{O} \tag{13}$$

#### **3. Materials and Methods**

#### *3.1. Materials*

The manganese slag samples were collected from Wenshan, China. The electrolytic manganese slag is firstly dried at 105 ◦C in the oven for 12 h, then mechanically ground by ball mill and sieved to 200 mesh (74 µm) for use. The standard gases include N<sup>2</sup> (≥99.99%), H2S/N<sup>2</sup> (1.00% H2S, *v*/*v*), PH3/N<sup>2</sup> (1.00% PH3, *v*/*v*), and O<sup>2</sup> (≥99.99%), all of which were purchased from Dalian Special Gases Co., Ltd., Dalian, China.

#### *3.2. Acid Leaching Procedure and Preparation of Modified MS Slurry*

First, the electrolytic manganese slag was weighed and transferred into a three-necked flask. Then, 2 mol L−<sup>1</sup> hydrochloric acid solution was taken into the three-necked flask with the connecting condensing device. The solid-to-liquid ratio of the slag to the acid solution was 1:6. The acid leaching temperature was set at 100 ◦C for 60 min. After acid leaching, the acid-leached solution was filtered. Then, filter residue and filtrate were obtained. The obtained filter residue was named as acid leaching residue. The acid leaching residue was then mixed with deionized water to obtain a 40 mL acid leaching residue slurry. Furthermore, the acid leaching residue was mixed with 0.01 mol CuSO4, named as acid leaching residue + CuSO4. The mixture of acid leaching residue and CuSO<sup>4</sup> was added into 40 mL deionized water to prepare the acid leaching residue + CuSO<sup>4</sup> slurry. The modified MS slurry was obtained by a mixture of 3 g MS, 40 mL deionized water, and 0.01 mol CuSO4, named MS + CuSO4. In addition, the 3 g of raw MS was added to 40 mL deionized water to prepare the raw MS slurry, named MS. The CuSO<sup>4</sup> solution also was prepared by adding a mixture of 40 mL deionized water and 0.01 mol CuSO4, named CuSO4.

#### *3.3. Analytical Method*

The solid samples were characterized by XPS technology to obtain the information of S 2p, P 2p, and Cu 2p that were measured by the ESCALAB 250 X-ray photoelectron spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) with a resolution of 0.45 eV (Ag) and 0.82 eV (PET). The sensitivity of the instrument is 180 kbps (200 µm, 0.5 eV). The anions in the solutions were detected by ICS-600 (Thermo Fisher Scientific, Waltham, MA, USA). The ion chromatography is equipped with an AS12A anion separation column consisting of sodium carbonate and sodium bicarbonate. The instrument is turned on for 1–2 h, and the background conductance is less than 30 µs to start measurement. XRD method was used to determine the phase structure of the solid samples prior and after H2S and PH<sup>3</sup> absorption experiments by a D/MAX-2200 X-ray diffractometer (Rigaku, Tokyo, Japan) with Cu Kα ray, at a voltage of 36 kV, at a current of 30 mA, with scanning range between 10 and 90◦ , and at a scanning speed of 5◦ /min.

#### *3.4. Catalytic Activity Test*

The experiment device for evaluating the activity of the remover is shown in Figure 8. The simulated flue gas consists of O2, H2S, PH3, and N2, all of which were supplied by gas cylinders (1a–1d), and their concentration were 1 vol %, 800 ppm (parts per million by volume), 400 ppm, respectively, with a total gas flow rate of 110 mL/min. The gas flow in each gas path is controlled by mass flow controllers (2) (Beijing Seven-Star Electronics Co., Ltd., Beijing, China) with a digital display (4) (Beijing Seven-star Electronics Co., Ltd., Beijing, China)., then all of the gases were mixed in the mixing tank (5) (J.K Fluid Technology, Jiaxing, China). The mixed gas will reach the desired concentration by adjusting the mass flow controllers, and be analyzed by gas chromatography (6) (FULI-9790II gas chromatography, FULI Instrument Co., Ltd., Taizhou, China). Then, the mixed gas passes through three-necked flask (10) combined with the constant temperature magnetic stirrer (DF-101S, INESA Scientific Instrument Co., Ltd., Shanghai, China), contacting the prepared modified manganese slag slurry in which mixed gas will react with the slurry. During the reaction process, the pH value of slurry was measured by pH meter (PHS-3C, INESA Scientific Instrument Co., Ltd., Shanghai, China). Moreover, the tail gas will first be analyzed by a gas chromatography and then be removed in the tail gas treatment system (9) (scrubbed by the 5 wt.% KMnO<sup>4</sup> solution with 5 wt.% H2SO4). *Catalysts* **2020**, *10*, x FOR PEER REVIEW 10 of 12 Moreover, the tail gas will first be analyzed by a gas chromatography and then be removed in the tail gas treatment system (9) (scrubbed by the 5 wt.% KMnO4 solution with 5 wt.% H2SO4). The calculation method for the conversion efficiency of PH3 and H2S is shown in Equation (14). PHଷ(HଶS) conversion efficiency (%) <sup>=</sup> PHଷ(HଶS)inlet െ PHଷ(HଶS)outlet PHଷ(HଶS)inlet ൈ 100 (14) where PH3 inlet and H2S inlet refer to the inlet concentrations of PH3 and H2S at the initial of the reactor, respectively, in mg/m3; and PH3 outlet and H2S outlet refer to the outlet concentrations of PH3 and H2S at the end of the reactor, respectively, in mg/m3.

**Figure 8.** Experimental device diagram of extractant activity evaluation. 1a, 99.99% O2 cylinder gas; 1b, 1 vol % H2S cylinder gas; 1c, 1 vol % PH3 cylinder gas; 1d, 99.99 vol % N2 cylinder gas; 2, mass flow meter; 3, on-off valve; 4, digital display; 5, gas mixing tank; 6, FULI-9790II gas chromatography; 7, data analysis; 8, constant temperature magnetic stirrer; 9, exhaust gas absorption bottle; 10, three-**Figure 8.** Experimental device diagram of extractant activity evaluation. 1a, 99.99% O<sup>2</sup> cylinder gas; 1b, 1 vol % H2S cylinder gas; 1c, 1 vol % PH<sup>3</sup> cylinder gas; 1d, 99.99 vol % N<sup>2</sup> cylinder gas; 2, mass flow meter; 3, on-off valve; 4, digital display; 5, gas mixing tank; 6, FULI-9790II gas chromatography; 7, data analysis; 8, constant temperature magnetic stirrer; 9, exhaust gas absorption bottle; 10, three-necked flask.

necked flask. The calculation method for the conversion efficiency of PH<sup>3</sup> and H2S is shown in Equation (14).

$$\text{PH}\_3(\text{H}\_2\text{S}) \text{ conversion efficiency} \,\%= \frac{\text{PH}\_3(\text{H}\_2\text{S})\_{\text{inlet}} - \text{PH}\_3(\text{H}\_2\text{S})\_{\text{outlet}}}{\text{PH}\_3(\text{H}\_2\text{S})\_{\text{inlet}}} \times 100\tag{14}$$

removing H2S and PH3. On the basis of different characterization techniques, we investigated the reaction mechanism of removing H2S and PH3 by modified MS slurry. The main findings of this study are as follows: (1) Through acid leaching experiments, the liquid-phase part after filtration has a leading role in where PH<sup>3</sup> inlet and H2S inlet refer to the inlet concentrations of PH<sup>3</sup> and H2S at the initial of the reactor, respectively, in mg/m<sup>3</sup> ; and PH<sup>3</sup> outlet and H2S outlet refer to the outlet concentrations of PH<sup>3</sup> and H2S at the end of the reactor, respectively, in mg/m<sup>3</sup> .

#### removing H2S and PH3. The highest PH3 conversion efficiency of leaching residue slurry + CuSO4 **4. Conclusions**

group can only obtain 15.14%, which indicated that the main active components were consumed by the acid leaching method. (2) By simulation of the modified MS slurry with metal ions based on real chemical composition, the catalytic activity for H2S and PH3 is relative to the types of metal salts, with the order being metal chlorides > metal nitrates > metal sulfates. All of the metal salts can obtain 100% H2S In this study, modified MS slurry was systematically carried out toward simultaneously removing H2S and PH3. On the basis of different characterization techniques, we investigated the reaction mechanism of removing H2S and PH<sup>3</sup> by modified MS slurry. The main findings of this study are as follows:

(3) H2S was oxidized to element S and sulfate due to the reaction between Cu2+ and H2S and part of the H2S oxidation by O2, while the PH3 was oxidized to PO43− by liquid-phase catalytic oxidation

(4) The best PH3 and H2S conversion efficiency was obtained by the modified MS slurry (MS + CuSO4), and the maximum removal efficiency of H2S and PH3 were 100% and ~78%, respectively. The simple modification process for raw MS through adding Cu2+ can effectively improve the

PH3 compared to Ca2+, Mg2+, and Mn2+ combined with Cu2+ groups.

of metal ions with the conversion of Cu2+ to Cu+.

removal efficiency. In addition, the metal chlorides can maintain above 70% PH3 conversion efficiency for 10.5 h and the highest PH3 conversion efficiency was 86.85%; whereas the highest


**Author Contributions:** J.B. and X.W. contributed equally. Conceptualization, X.S. (Xin Sun) and P.N.; methodology, X.S. (Xin Sun); software, C.W. and X.S. (Xin Song); validation, X.W.; formal analysis, J.B.; investigation, X.W.; resources, X.S. (Xin Sun), K.L., and P.N.; data curation, J.B.; writing—original draft preparation, J.B.; writing—review and editing, J.B..; visualization, J.B.; supervision, X.S. (Xin Sun); funding acquisition, X.S. (Xin Sun), K.L., F.W., and P.N. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Natural Science Foundation of China (grant 51708266, 51968034, 21667015, and 41807373) and the National Key R&D Program of China (grant 2018YFC0213400, 2018YFC1900305).

**Conflicts of Interest:** The authors declare no conflict of interest.

## **References**


**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Communication*
