*2.4. XPS Analysis*

X-ray photoelectron spectroscopy (XPS) analysis was carried out to study the chemical state and surface ratio of MoS2, NiS2, and Ni-Mo sulfide/Al2O3 catalysts with different Ni/Mo molar ratios. Figure 4 shows the Mo 3d and Ni 2p spectra of the various catalysts. For Mo element, as shown in Figure 4a, the observed binding energy (BE) of Mo 3d5/2 was about 229.0 eV, indicating that the Mo species were Mo4<sup>+</sup> [22]. In Figure 4b, for the case of Ni, the main peaks at the BE of about 855.0 eV can be attributed to the Ni 2p3/2 peaks of Ni2<sup>+</sup> [22]. However, as the Ni/Mo molar ratio increased, the peak position of Mo 3d5/2 gradually shifted toward the peak position of the lower BE, accompanied by the BE shift of Ni 2p3/2. This phenomenon indicates the increased electron density in Mo 3d5/2, resulting from the electron donating property of Ni 2p3/2. Therefore, a strong electron interaction between Ni and Mo occurs on the catalyst surface, wherein electrons likely transfer from the Ni species to the Mo species in the Ni-Mo sulfide/Al2O3 catalysts.

**Figure 4.** XPS spectra of the Ni-Mo sulfide/Al2O3 catalysts with different Ni/Mo molar ratios: (**a**) Mo 3d, (**b**) Ni 2p.

Table 2 shows the surface and total Ni/Mo ratio of Ni-Mo sulfide/Al2O3 catalysts with different Ni/Mo molar ratios. As presented, the total Ni/Mo molar ratio was consistent with the theoretical ratio. However, these total Ni/Mo ratios (0.31 ˙ I2.97) were higher than the surface Ni/Mo ratio (0.18-2.41). This is because Ni2<sup>+</sup> ions were intercalated into the gap position of the MoS2 lattice, and a large number of Mo vacancies could be generated. Therefore, the surface of the Ni-Mo sulfide/Al2O3 catalysts became slightly Ni-depleted.


**Table 2.** The surface and total Ni/Mo atomic ratios of the Ni-Mo sulfide/Al2O3 catalysts with different Ni/Mo molar ratios.

> 1 By XPS. 2 By ICP.

#### *2.5. Catalytic Evaluation for the One-Step Conversion of CO2 and H2S to Syngas*

The catalytic performances of the various Ni-Mo sulfide/Al2O3 catalysts were evaluated through converting CO2 and H2S into syngas in non-thermal plasma. For comparison, the performances of NiS2/Al2O3 and MoS2/Al2O3 were also investigated. As seen in Figure 5a,b, all the Ni-Mo sulfide/Al2O3 catalysts possessed better activities in CO2 and H2S conversion than NiS2/Al2O3 and MoS2/Al2O3 catalysts, and the CO2 and H2S conversions could reach high levels. The experimental results show that the Ni/Mo molar ratio had a grea<sup>t</sup> influence on the conversion of CO2 and H2S. As the Ni/Mo molar ratio increased, the catalytic activity presented a primary enhancement followed by a decline. The CO2 and H2S conversions were strongly dependent on the SEI (Specific energy input). At SEI of 60.0 kJ/L, CO2 conversions were 25.1%, 45.0%, 46.2%, 46.9%, 47.7%, 56.3%, and 49.0%, and H2S conversions were 87.8%, 93.7%, 94.8%, 95.7%, 96.4%, 98.9%, and 97.3% when NiS2/Al2O3, MoS2/Al2O3, 2Ni-6Mo/Al2O3, 3Ni-5Mo/Al2O3, 4Ni-4Mo/Al2O3, 5Ni-3Mo/Al2O3, and 6Ni-2Mo/Al2O3 were filled in the gap, respectively. Especially, the 5Ni-3Mo/Al2O3 catalyst exhibited the best catalytic performance and achieved relatively high CO2 and H2S conversions with the lowest SEI.

**Figure 5.** The conversions of CO2 (**a**) and H2S (**b**) as a function of SEI on the Ni-Mo sulfide/Al2O3 catalysts with various Ni/Mo molar ratios. Reaction conditions: feed: H2S/CO2 molar ratio = 20:15; flow rate: 35 mL/min; catalyst bed volume: 15.0 mL.

In addition, as seen in Figure 6a,b, the major products were CO and H2 in the CO2 and H2S conversion. CO and H2 concentrations were in line with SEI, which indicates that the behavior for CO2 and H2S conversion had relatively stronger dependence on the energy input. An increase of SEI could generate more active H species and obviously promote CO2 activation and CO production, together with the decrease in H2 yields. Meanwhile, very small amounts of light hydrocarbons (others: CH4, C2H4, and C2H6) were also generated. The selectivity to light hydrocarbons was very low (<2%) during the reaction. Furthermore, there were not any C3+ hydrocarbons. Therefore, this novel method may produce clean syngas. Additionally, it was also found that SEI strongly affected the H2/CO ratio. In Figure 7, when SEI was changed from 20 to 110 kJ/L, the H2/CO ratio considerably decreased from about 4.5 to 1.0, which illustrates that the H2/CO ratio strongly depends on the energy input. An increase of SEI could induce the decrease in the H2/CO ratio. Hence, the H2/CO ratio can be controllably adjusted on a large scale through varying SEI by this method.

**Figure 6.** CO concentration (**a**) and H2 concentration (**b**) in gas product as a function of SEI in the plasma-induced CO2 and H2S conversion on the Ni-Mo sulfide/Al2O3 catalysts with various Ni/Mo molar ratios. Reaction conditions: feed: H2S/CO2 molar ratio = 20:15; flow rate: 35 mL/min; catalyst bed volume: 15.0 mL.

**Figure 7.** H2/CO molar ratio as a function of SEI in the plasma-induced CO2 and H2S conversion on the Ni-Mo sulfide/Al2O3 catalysts with various Ni/Mo molar ratios. Reaction conditions: feed: H2S/CO2 molar ratio = 20:15; flow rate: 35 mL/min; catalyst bed volume: 15.0 mL.

A series of characterizations of the Ni-Mo sulfide/Al2O3 catalysts displays that the Ni/Mo molar ratio had a significant effect on the physical and chemical properties of the catalyst. We have reported that the synergistic effects of semiconductor catalyst and non-thermal plasma in the H2S decomposition [12]. In the present work, the Ni-Mo sulfide/Al2O3 catalyst in non-thermal plasma

can be excited by both the strong electric field and UV-visible light irradiation, and thus generate highly active hole–electron pairs. The hole–electron pairs will react with the adsorbed surface species, thereby accelerating the conversion of CO2 and H2S. Hence, since the generated hole–electron pairs are su fficiently reactive to convert CO2 and H2S to H2 and CO, the rate of CO2 and H2S conversion depends on the number of electron–hole pairs generated on the surface of the Ni-Mo sulfide/Al2O3 catalyst. A higher number of hole–electron pairs may be linked to the relatively higher behavior on CO2 and H2S conversion. From the results of UV-vis spectra (shown in Figure 2 and Table 1), the change in the Ni/Mo molar ratio a ffects the optical properties of the Ni-Mo sulfide/Al2O3 catalyst. With increasing the Ni/Mo molar ratio, a monotonous variation in the absorption in visible light region and band gap of Ni-Mo sulfide could be clearly found. For a semiconductor catalyst with a narrower band gap, less energy for electrons is required to jump from valence band (VB) to conduction band (CB). Therefore, a decrease in band gap can lead to the increase in the amount of hole–electron pairs. Moreover, the other optical properties of semiconductor catalyst, such as conduction band position and valence band position, are also related to its chemical compositions. According to the XRD and TEM results, the Ni-Mo sulfide possessed the layer structure, the Ni2<sup>+</sup> ions can replace the position of Mo ions or enter the gap position of MoS2 to form Ni-Mo-Sx phase. Hence, the suitable impurity energy level could be provided through a proper doping amount of Ni2<sup>+</sup> ions into MoS2. The presence of impurity levels leads to the easy injection of the excited electrons from VB to CB of MoS2.

In addition, all the Ni-Mo sulfide/Al2O3 catalysts exhibited relatively high BET surface areas (shown in Table 1). The high surface area facilitates photon absorption, provides more active sites, and reduces the distance of generated carriers from the catalyst surface [23]. Moreover, the average particle size was around 10 nm. The small nanoparticles with low crystallinity are favorable for the fast electron transportation from bulk to surface, which prevent the recombination of the generated electrons and holes of the catalyst [24]. Therefore, the reduction in the particle size of the Ni-Mo sulfide/Al2O3 catalyst also contributes to the improvement of the catalytic activity.

Additionally, the Ni2<sup>+</sup> ions can be evenly incorporated into the MoS2 lattice to form Ni-Mo-Sx phase, which would bring about the Mo vacancies formation. Ideally, the incorporation of two Ni2<sup>+</sup> ions may generate one Mo vacancy. Therefore, the incorporation of Ni2<sup>+</sup> ions can produce a large amount of Mo vacancies. Mo vacancies favor the separation of the energy-induced electrons and holes, which induce the high catalytic performance in CO2 and H2S conversion, compared to the MoS2/Al2O3 catalyst. Nevertheless, the 6Ni-2Mo/Al2O3 catalyst with higher Ni content possessed relatively stronger visible light absorption capacity than other Ni-Mo sulfide/Al2O3 catalysts, but the CO2 and H2S conversions were lower. The reason for the low catalytic activity over the 6Ni-2Mo/Al2O3 catalyst may be that the excessive amount of Ni would result in the unevenly distributed Ni2<sup>+</sup> ions. Owing to the higher concentration of Ni, the probability of electron–hole recombination was regarded to become comparably high. Consequently, a superfluous increase in the Ni/Mo molar ratio not only encumbered the light absorption, but also o ffered more recombination sites for hole–electron pairs, so the catalytic activity was repressed. In particular, when the Ni/Mo molar ratio was 5/3, the Ni-Mo sulfide/Al2O3 catalyst exhibited the best catalytic activity for CO2 and H2S conversion with the most proper optical and structural properties.

Figure 8 presents the CO2 and H2S conversion, and H2/CO molar ratio variations over 5Ni-3Mo/Al2O3 during the long-term test. The results demonstrate that the catalytic activity did not exhibit loss in the runs. The XPS spectra and SEM images were taken before and after evaluation, as shown in Figures S1 and S2 (Supplementary Materials), respectively. There was no obvious di fference detected in the spent 5Ni-3Mo/Al2O3 catalyst after the reaction test. Moreover, Figure S3 (Supplementary Materials) shows a comparison between the fresh 5Ni-3Mo/Al2O3 catalyst and the spent one after reaction tests in the XRD patterns. According to the Scherrer equation, the average particle size of the 5Ni-3Mo/Al2O3 catalyst increased from 11.5 to 13.2 nm after the 50 h long-term test, proving high stability of the active phases on the 5Ni-3Mo/Al2O3 catalyst in the CO2 and H2S conversion process. Furthermore, the surface area decreased only from about 250 to 231 m<sup>2</sup>/g. The

active phases were highly dispersed in Al2O3 support, which could also prevent the agglomeration formation and inhibit the growth of the particles. Hence, it is clear that Ni-Mo sulfide underwent no variation in non-thermal plasma and can maintain stable structures in the plasma-induced CO2 and H2S conversion.

**Figure 8.** Variations of CO2 and H2S conversion and H2/CO molar ratio with time in the plasma-induced conversion over 5Ni-3Mo/Al2O3. Reaction conditions: H2S/CO2 molar ratio = 20:15; flow rate: 35 mL/min; catalyst bed volume: 15.0 mL; SEI: 50.0 kJ/L.

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