Catalytic combustion of dimethyl disulfide on bimetallic supported catalysts prepared by the wet-impregnation method

: In this paper, the catalytic combustion of DMDS (dimethyl disulfide, CH 3 SSCH 3 ) over bimetallic supported catalysts were investigated. It was confirmed that Cu/γ-Al 2 O 3 -CeO 2 showed best catalytic performance among the five single-metal catalysts. Furthermore, six different metals were separately added into Cu/γ-Al 2 O 3 -CeO 2 to investigate the promoting effect. The experiments revealed Pt as the most effective promoter and the the best catalytic performance was achieved as the adding amount of 0.3 wt%. The characterization results indicated that high activity and resistance to sulfur poisoning of Cu-Pt/γ-Al 2 O 3 -CeO 2 could be attributed to the synergistic effect between Cu and Pt.

can have an impact on the metal-support interaction.
In this study, the support of catalyst was screened firstly. Then the most active of the several γ-Al2O3-CeO2 supported single-metal oxides was determined as the principal catalyst. A variety of metal oxides were separately added to this catalyst to investigate the promoting effect, and the most effective one was selected as a promoter in later investigation. The physicochemical properties of the catalyst and its structure-activity relationship were studied by using characterization techniques. Finally, the effects of catalyst preparation conditions on the catalytic activity of Cu-Pt/γ-Al2O3-CeO2 catalystic combustion DMDS were investigated, and long-term stability experiments were performed.

Effect of the supports
In order to study the effect of different supports on the activity of supported catalysts， Cu/(γ-Al2O3, CeO2, γ-Al2O3-CeO2) and Cu-Pt/(γ-Al2O3, CeO2, γ-Al2O3-CeO2) catalysts were prepared, the performance of catalytic combustion DMDS were investigated，and the results are shown in Fig. 1. Under the experimental conditions, the DMDS catalytic ignition temperatures of Cu/γ-Al2O3-CeO2, Cu/γ-Al2O3 and Cu/CeO2 catalysts are 200 o C-300 o C. The catalytic activity of DMDS is compared between three Cu-based catalysts and the order of activity: Cu/γ-Al2O3-CeO2>Cu/γ-Al2O3>Cu/CeO2. Moreover, it can be seen that the activity of Pt-Cu bimetallic supported catalyst is consistent with that of Cu single-metal supported catalyst. The catalytic activity of Cu-Pt/γ-Al2O3-CeO2 catalyst is significantly higher than that of Cu-Pt/γ-Al2O3 and Cu-Pt/CeO2. Compared with γ-Al2O3, γ-Al2O3-CeO2 has great advantages in catalytic combustion of DMDS for the addition of CeO2. Because CeO2 has a very strong oxygen storage capacity, and the transfer of charge between the active species and the CeO2 support is beneficial to enhance the reactivity of the catalyst. Therefore, γ-Al2O3-CeO2 was chosen as the catalyst support in this study. Cu-Pt/(γ-Al2O3, CeO2, γ-Al2O3-CeO2) catalysts. Al is γ-Al2O3, Ce is CeO2, and Ce-Al is γ-Al2O3-CeO2, which is the same as the following figure.
peak of CuO on the Cu/CeO2 catalyst can be clearly observed, indicating that the Cu phase on the surface of the CeO2 support is in the form of larger CuO, which is more unfavorable for the catalytic combustion reaction. Therefore, CeO2 supported single-metal catalysts have the worst activity. With no diffraction peaks of CuO appearing in the Cu-Pt/CeO2 catalyst, it can be inferred that the addition of Pt causes CuO to form smaller particle grains and improve the distribution of CuO. In addition, comparing with the Cu/γ-Al2O3-CeO2 and Cu/γ-Al2O3 XRD patterns, it was found that the addition of CeO2 enhanced the regularity of the Al2O3 micropores and facilitated the dispersion of CuO. The XRD patterns of Cu-Pt/(γ-Al2O3, CeO2, γ-Al2O3-CeO2) do not show the diffraction peaks of Cu phase or Pt phase. It can be inferred that Cu phase or Pt phase in support are present in the form of oxides of highly dispersed small particles, which is advantageous for the reaction [33].

Single-metal supported catalyst
In order to select the principal catalyst, activity of single-metal supported catalyst was evaluated, and the characteristic temperature diagram of catalytic combustion of DMDS of (Cu, Fe, Zn, Mo, V)/γ-Al2O3-CeO2 catalyst are showed in Fig.3. The complete conversion temperature of Cu/γ-Al2O3-CeO2 is about 308 o C, which is obviously superior to other catalysts. The complete conversion temperature of V/γ-Al2O3-CeO2 and Zn/γ-Al2O3-CeO2 catalysts is about 345 o C, and the catalytic effect is not obvious. At the same time, T50 (temperature at which DMDS conversion rate is 50%) is the main evaluation condition, supplemented by T10 (temperature when DMDS conversion rate is 10%) and T90 (temperature when DMDS conversion rate is 90%). The DMDS catalytic activity of a transition metal catalyst is obtained in order of activity from high to low: (Cu>Fe>Mo>V>Zn)/γ-Al2O3-CeO2. Based on the above analysis, the catalytic activity of Cu/γ-Al2O3-CeO2 catalyst for oxidizing DMDS is the highest in the transition metal supported γ-Al2O3-CeO2 supported catalyst and was served as the principal catalyst for further investigation. Fig.4 shows the XRD patterns of (Cu, Fe, Zn, Mo, V)/γ-Al2O3-CeO2 catalyst, in which we can know the effect of the dispersibility of the active component on the catalytic activity. It can be found that after metal supported such as Cu, Fe, Zn, Mo and V, the diffraction peak of the support spectrum is clearly visible, and the position of the peak does not change, indicating that the support retains the mesoporous structure intact after metal ion impregnation [34,35]. Only ZnO is present in the form of large particles on the support, which diffraction peaks are clearly visible, so its catalytic combustion activity is lowest. The intensity of diffraction peaks of (Fe, Zn, Mo, V)/γ-Al2O3-CeO2 catalysts is slightly decreased, and the intensity of Cu/γ-Al2O3-CeO2 diffraction peaks is slightly increased, indicating that the Fe, Zn, Mo, and V metal loadings reduce the regularity of the mesopores, while the Cu loading increases the regularity of the mesopores [36].
Intensity/a.u. attributed to CuO reduction [37]. By comparison, it was found that the reduction peak of Cu/γ-Al2O3-CeO2 catalyst had the lowest temperature, the highest peak intensity and the best reduction. The redox capacities of the catalysts are ranked: (Cu>Fe>Mo>V>Zn) /γ-Al2O3-CeO2, which is highly consistent with the catalytic combustion activity of the transition metal catalyst DMDS. It is inferred that the stronger the redox ability of the catalyst, the stronger the catalytic combustion activity of DMDS.

Bimetallic catalyst
Cu/γ-Al2O3-CeO2 catalyst was determined as the principal catalyst, due to its low can be seen that the addition of V inhibits the catalytic activity of Cu/γ-Al2O3-CeO2.
When promoter was a noble metal, it was found that (Pt, Pd)/γ-Al2O3-CeO2 had good activity for catalytic combustion of DMDS and Pt-Cu/γ-Al2O3-CeO2 was more significant.
In order to evaluate the ability of catalyst to resist sulfur poisoning, the stability of Pt/γ-Al2O3-CeO2 is rapidly inactivated for poor resistance to sulfur poisoning. The reason for catalyst deactivation is that sulfur compounds bond tightly to the active site of the catalyst forming stable surface metal sulfides during catalytic combustion, which prevent adsorption of reactants on the surface [38][39][40]. It can be seen from Fig.10 that the Pt phsae in the Cu-Pt/CeO2-Al2O3 catalyst exists in the form of Pt 0 . The presence of Pt 0 can enhance the dispersibility, reduction and sulfur poisoning ability of CuO than PtO [41].
Combined with Fig.6, it is found that the catalytic activity of Cu-Pt/γ-Al2O3-CeO2 catalyst DMDS is the highest, and the good sulfur poisoning ability. Cu-Pt/γ-Al2O3-CeO2 catalyst is an ideal catalyst for catalytic combustion of DMDS. Cu-Pd supported on the γ-Al2O3-CeO2 support, the diffraction peak of the γ-Al2O3-CeO2 support spectrum is clearly visible, and the position of the peak does not change [42]. It indicates that the support retains its structure intact after metal ion impregnation.
However, after the addition of Fe, Zn, Mo, V, Pt and Pd, the intensity of the diffraction peak of the bimetallic catalyst is lower than that of the Cu/γ-Al2O3-CeO2 catalyst. It indicates that the addition of Fe, Zn, Mo, V, Pt and Pd changes the regularity of Cu/γ-Al2O3-CeO2 mesopores and reduces the crystallinity of γ-Al2O3-CeO2. Only the characteristic diffraction peaks of Cu-Zn/γ-Al2O3-CeO2 catalysts are observed, which catalytic activity and sulfur poisoning ability are not good, other metal oxide characteristic peaks are not found. This may be because the impregnation process and the calcination process are good for the catalyst treatment and the Cu, Fe, Mo, V, Pt and Pd phase have small particle size and are highly dispersed [43].
In order to obtain the information on the interaction between metal oxides or metal oxides and supports during the reduction process of supported metal catalysts, H2-TPR spectrum of (Cu, Cu-Mo, Cu-Fe, Cu-Zn, Cu-V, Cu-Pt, Cu-Pd)/γ-Al2O3-CeO2 catalysts are showed in Fig.9. It is found that the reduction peak of Cu-Pd/γ-Al2O3-CeO2 catalyst at Cu-Mo highly dispersed catalyst surface reduction [44]. The loading of Pt reduced the temperature of Cu/γ-Al2O3-CeO2 support reduction, enhanced the strength of the reduction peak and enhanced its reducibility. This indicates that the effect of Pt on the surface of Cu-Pt/γ-Al2O3-CeO2 on the reduction of CeO2 is affected by CeO2 hydrogen spillover effect [45]. The loading of Mo, Fe and Zn metals enhances the strength of the reduction peak of Cu-M/γ-Al2O3-CeO2 catalyst, and the loading of V greatly reduces the reducibility of Cu-V/γ-Al2O3-CeO2 catalyst.   Pt4f XPS spectrum of Cu-Pt/γ-Al2O3-CeO2 In order to obtain the chemical and electronic states of the medium metal of the bimetallic catalyst, Cu2p of (Cu-Pd, Cu-Pt, Cu-Mo, Cu-Fe, Cu-Zn, Cu-V, Cu)/γ-Al2O3-CeO2 catalyst are showed in Fig.10 (a) . By comparison, it can be found that all the catalysts have three peaks, the main peak at BE=934.5 eV, corresponding to the characteristic peak of Cu2p2/3 orbital. The satellite peaks at BE=943.5 eV and BE=954.1 eV belong to the characteristic peak of Cu2p1/2, and there is no peak shift. The above information indicates that Cu phase on the surface of (Cu-Pd, Cu-Pt, Cu-Mo, Cu-Fe, Cu-Zn, Cu-V, Cu) /γ-Al2O3-CeO2 catalysts mainly exist in the form of CuO [46]. Fig.10 (b) shows XPS spectrum of a Cu-Pt/γ-Al2O3-CeO2 catalyst. It can be found that the peaks of Cu-Pt/γ-Al2O3-CeO2 catalyst Pt 4f at BE=71.2 eV and BE=74.65 eV are Pt 4f7/2 and Pt 4f5/2, respectively, which can be attributed to Pt 0 [47]. The shoulder peak at BE =72.8 eV can be attributed to PtO on the surface of the catalyst.
The right amount of loading is critical to the impact of the supported catalyst, so of the catalyst carrier itself is fixed. At the beginning of increasing the loading of Cu, more active sites can be introduced, which is beneficial to increase the catalytic combustion activity of DMDS. However, when the dispersibility of Cu species reaches the threshold, increasing the loading of Cu, it is impossible to introduce more active sites, and the excessive Cu loading will cause the clustering effect of the metal, and the catalytic combustion activity of DMDS cannot be further improved. It can be seen from Fig.11 (b) that as the Pt loading increases from 0% to 0.5%, the DMDS catalytic combustion activity of the catalyst gradually increases. However, as the Pt loading was from 0.3% to 0.5%, there was almost no change in catalyst activity growth. In summary, when the loading amount of Cu is 5% and the loading amount of Pt is 0.3%, it is economical and efficient.
Since the above results indicate that the Cu-Pt/γ-Al2O3-CeO2 catalyst performance is ideal, we conducted long-term stability experiments, and the stability diagram of catalytic combustion of DMDS for 1000 h is showed in Fig.12. It can be found that the The conversion of DMDS, and yields of SO2 are defined in the following way: where Ci is the initial feed concentration of DMDS (ppm), C0 is the outlet concentration of DMDS (ppm) and [SO2] is the concentrations of the corresponding compounds in mol▪L −1 .

Catalyst characterization
The crystal structure analysis of the supported catalyst in this study was characterized by D8FOCUS X-ray diffractometer (XRD, Bruker, Germany). The diffractometer is equipped with CuKα rays, and the setting parameters are as follows: 2θ In this study, the oxidation/reduction performance of the catalyst was characterized by a TPR/D/O1100 catalyst fully automatic analyzer(Thermo Fisher Scientific). The instrument is equipped with a highly sensitive TCD detector. The test procedure is as follows. First, 0.05 g of the catalyst sample is charged into the catalyst fully automatic analyzer for pretreatment. Pretreatment conditions: maintained at 200 o C for 1 h under a nitrogen atmosphere. Then cooled to room temperature and subjected to H2 temperature programmed reduction. Test conditions: 5% (Vol.) H2/N2 flow rate 20 ml/min, temperature from 50 o C to 700 o C, heating rate 10 o C/min.

Conclusion
γ-Al2O3-CeO2 was observed to be the most optimal support and Cu was determined as the principal active phase for catalytic combustion of DMDS. Among the six different types of biimetallic supported catalysts, the Cu-Pt/γ-Al2O3-CeO2 catalyst exhibits the highest activity and sulfur poisoning ability for the DMDS combustion based on the conversion. According to XRD，H2-TPR and XPS results, there are close interaction between Pt and Cu (intermetallic Pt-Cu) in the nano sized scale, which could be the main reason why this catalyst showed the highest activity.
Under the conditions of GHSV of 50000h -1 , DMDS concentration of 1000ppm and oxygen concentration of 5%, the prepared 5%Cu-0.3%Pt/γ-Al2O3-CeO2 catalyst has the highest catalytic activity of DMDS, and 262 o Ccan achieve complete conversion of DMDS. In addition, the 5%Cu-0.3%Pt/γ-Al2O3-CeO2 catalyst has a conversion of about 100% in the 1000-hour stability test at a space velocity of 30000 h -1 , and the SO2 yield is above 97%, which is an ideal catalyst for catalytic combustion of DMDS.

Z.
Funding: This research was funded by Beijing science and technology projects Z181100005418011.
Conflicts of Interest: The authors declare no conflict of interest.