Effect of various g-C3N4 precursors on the catalytic performance of alkylorganotin-based catalysts in acetylene hydrochlorination

A series of alkylorganotin-based catalysts (Sn-g-C3N4 /AC) was prepared by wet impregnation in ethanol using different g-C3N4 precursors and alkylorganotin compounds. The structure, texture, surface composition, and adsorption properties of the as-prepared catalysts were extensively characterized. Then, the obtained samples were evaluated for their catalytic performance in hydrochlorination of acetylene. The results provided by the X-ray photoelectron spectroscopy, acetylene temperature-programmed desorption, and HCl adsorption confirmed the nature of the active sites (i.e. Sn-Nx) involved in the reactant adsorption, and hence in the improved catalytic performance. These active sites were also related to the improved lifetime of alkylorganotin-based catalysts in the hydrochlorination of acetylene. At a constant reaction temperature of 200 °C with an acetylene gas hourly space velocity (C2H2 -GHSV) of 30 h-1 , Sn-g1 -C3N4 /AC-550 exhibited the highest acetylene conversion (~98.0%) and selectivity toward the vinyl chloride monomer (>98.0%). From the catalytic test results, it was reasonably concluded that the hexamethylenetetramine is the most suitable N precursor, as compared to the dicyandiamide and urea, to prepare high-performance catalysts. From the BET specific surface area of fresh and used catalysts, it was suggested that, in contrast to dicyandiamide and urea, hexamethylenetetramine could delay the deposition of coke on alkylorganotin-based catalysts, which is reflected by the extended lifetime.


Introduction
Owing to the widespread applications of polyvinyl chloride (PVC) in all human activities, there is an increased demand for vinyl chloride monomer (VCM) as precursor for PVC manufacture [1,2]. The hydrochlorination of acetylene is the main technology for producing VCM in several countries, especially in China, due to their rich coal reserves. However, there are several drawbacks of the use of carbon-supported mercuric chloride catalysts for the synthesis of VCM via acetylene hydrochlorination. Specifically, stringent government policies and severe mercury pollution have urged the researchers to explore alternative catalysts [3]. Hence, the design and development of mercury-free catalysts is extensively investigated [4].
In this study, hexamethylenetetramine, urea, and dicyandiamide were selected as precursors of g-C 3 N 4 , and the effect of these precursors on the catalytic properties of alkylorganotin-based catalysts was investigated.
The as-obtained catalysts were characterized by X-ray diffraction (XRD), N 2 adsorption-desorption isotherms, thermogravimetric analysis (TGA), and derivative thermogravimetric (DTG) analysis, X-ray photoelectron spectroscopy (XPS), HCl adsorption, and acetylene temperature-programmed desorption (C 2 H 2 -TPD). The results indicated that hexamethylenetetramine is the best g-C 3 N 4 precursor as Sn-N X in alkylorganotin serves as catalytically active sites.

Catalytic performance of Sn-g-C 3 N 4 /AC
In our previous study [14], the change in the calcination temperature of the g-C 3 N 4 precursor in the range of 400°C-650°C had a positive effect on the hydrochlorination activity of organotin. Thus, to investigate the effect of various g-C 3 N 4 precursors on the performance of alkylorganotin-based catalysts for acetylene hydrochlorination, the calcination of Sn-g-C 3 N 4 /AC catalyst precursors was carried out at 400, 450, 500, 550, and 650°C. Tables 2 and 3 summarize the BET specific surface areas and total pore volumes of Sn-g-C 3 N 4 /AC catalysts. Clearly, the BET specific surface areas and total pore volumes of Sn-g-C 3 N 4 /AC catalysts were lower than those of g- , indicating that the pores of g-C 3 N 4 /AC were filled with alkylorganotin. Hence, BET specific surface areas between 210 and 653 m 2 g −1 are obtained depending on the calcination temperature. First, with the increase in the calcination temperature, the values of the textural properties increased and then decreased, demonstrating the effect of the temperature on the textural properties of Sn-g-C 3 N 4 /AC. To identify the optimal calcination temperature of Sn-g-C 3 N 4 /AC, the acetylene hydrochlorination over all samples was performed. Figure [14] were higher than that of Sn/AC (89.1%) (Figure 2d) [14]. Therefore, the enhancement in activity of alkylorganotinbased catalysts correlates to the g-C 3 N 4 precursor, and hexamethylenetetramine is obviously the best g-C 3 N 4 precursor. Usually, XRD is used to investigate the dispersion of metal species and eliminate the interference of metal agglomeration [18,41]. As shown in Figure 2e, the XRD patterns of Sn- and Sn-g 3 -C 3 N 4 -400 display 3 obvious peaks at 26.7°, 33.9°, and 51.7°, which are characteristic to Sn-g-C 3 N 4 .
However, there are no typical peaks of tin metal (PDF#18-1380) and SnO 2 phases (PDF#33-1374) in Sn-g-C 3 N 4 , inferring that the high temperature favoured the formation of Sn-g-C 3 N 4 . Figure  , 2f), suggesting that Sn-g 1 -C 3 N 4 -550, Sn-g 2 -C 3 N 4 -450, and Sn-g 3 -C 3 N 4 -400 are well dispersed on the AC surface. In line with the conditions used to perform acetylene hydrochlorination at the industrial level, the reaction temperature and acetylene gas hourly space velocity were controlled in the range of 130°C-180°C and 30-50 h −1 , respectively [42]. However, when the reaction temperature was 180°C , the acetylene conversion over the Sn-g-C 3 N 4 /AC did not achieve~98%. When the temperature increased to 200°C, the acetylene conversion over Sn-g-C 3 N 4 /AC increased to~98%, which is close to the activity of HgCl 2 /AC. Therefore, the stability of catalysts was tested in acetylene hydrochlorination at 200°C.
Pore size (nm) Sn Pore size (nm) Sn  [14]. Sn-g 1 -C 3 N 4 /AC-550, Sn-g 2 -C 3 N 4 /AC-450, and Sn-g 3 -C 3 N 4 /AC-400 were highly selective toward VCM (>98.0%, Figure 3b). Moreover, the selectivity did not change over the entire reaction period. After the reaction, the BET specific surface area and pore volume of the used catalysts decreased in comparison with those of the fresh catalysts (Table 4) due to the deposition of coke on the catalyst surface. Thus, the percentage decrease in the BET specific surface area of Sn-g 1 -C 3 N 4 /AC-550, Sn-g 2 -C 3 N 4 /AC-450, Sn-g 3 -C 3 N 4 /AC-400, and Sn/AC are 54, 68, 77, and 79%, respectively. This phenomenon is mainly attributed to the polymerization of acetylene and vinyl chloride during acetylene hydrochlorination [43][44][45][46][47][48]. Consequently, the g-C 3 N 4 precursor may prevent the loss of the BET specific surface area of alkylorganotin-based catalysts in the hydrochlorination of acetylene, thereby extending the lifetime of the alkylorganotin-based catalysts. Furthermore, from this point of view, among the 3 g-C 3 N 4 precursors, the hexamethylenetetramine is proved to be the optimum one.

Chemical states of Sn and N
XPS was employed to investigate the chemical states of Sn, N, C, and Cl on the Sn-g 1 -C 3 N 4 -550, Sn-g 2 -C 3 N 4 -450, and Sn-g 3 -C 3 N 4 -400 catalyst surfaces. Sn 3d , C 1s , N 1s , and Cl 1s signals were detected in all 3 samples (Figure 4a and Table 5).

Thermal stability of Sn-g-C 3 N 4 /AC
To investigate the thermal stability of Sn-g 1 -C 3 N 4 /AC-550, Sn-g 2 -C 3 N 4 /AC-450, and Sn-g 3 -C 3 N 4 /AC-400, TGA curves were recorded under nitrogen. Figures 5a, 5b, and 5c show the TG-DTG curves recorded between 25°C and 800°C for the investigated catalysts. All 3 catalysts exhibited a similar weight loss trend.

C 2 H 2 -TPD and HCl adsorption
Both Sn 4+ and HCl as electron-acceptors do not react with each other [58], inferring that alkylorganotin firstly prefers to interact with C 2 H 2 in acetylene hydrochlorination and then reacts with HCl to generate vinyl chloride. Therefore, reactant adsorption of catalysts behaves a significant impact on catalytic performance. As shown in Figure 6a, the acetylene adsorption capacity follows the order of Sn-g 1 -C 3 N 4 -550 >Sn-g 2 -C 3 N 4 -450 >Sn-g 3 - and Sn-g3-C3N4/AC-400 (138.4°C) (Figure 6a). Therefore, the strength of acetylene adsorption on Sn-g1-C3N4/AC-550 is higher as compared to those on Sn-g2-C3N4/AC-450 and Sn-g3-C3N4/AC-400. The C 2 H 2 -TPD profiles of Sn-g 1 -C 3 N 4 /AC-550, g 1 -C 3 N 4 /AC-550, and Sn/AC are illustrated in Figure 6b. Although Sn-C and Sn-Cl x co-exist in Sn/AC and Sn-g 1 -C 3 N 4 /AC-550, the last sample exhibits higher acetylene adsorption capacity than Sn/AC-550. Furthermore, Figure 6b shows that the acetylene adsorption capacity and the adsorption strength both increased in Sn-g 1 -C 3 N 4 /AC-500 samples, improvement was associated with the existence of Sn-Nx in these catalysts. As shown in Table 6, hexamethylenechloride as nitrogen precursor can stabilize the content of Sn and thus, an enhanced catalytic performance was obtained for this sample in comparison with those of the catalysts obtained with urea and dicyandiamide. The corresponding catalytic results are depicted in Figure 2d.
A previous study [59] has reported that hydrogen chloride adsorption is the rate determining step of  6d). The higher hydrogen chloride adsorption capacity of Sn-g 1 -C 3 N 4 /AC-550 is attributed to the coexistence of pyridinic N [24] and Sn-N x sites.

Conclusions
According to the previous study, MF-600 with 94.5% acetylene conversion was prepared using melamine and toxic formaldehyde [57]. In addition, Sn/AC and g-C 3 N 4 /AC exhibited acetylene conversion of 89.1% and 76.5%, respectively, in acetylene hydrochlorination [14,21]. To meet the industrial requirement of activity and the green route development of chemical industry and further study the properties of Sn-based catalysts during acetylene hydrochlorination, in this study, Sn-g-C 3 N 4 /AC as novel nonprecious metal-based catalyst was prepared with alkylorganotin and g-C 3 N 4 precursors by wet impregnation, as well as exhibited higher acetylene conversion (97.8%). The excellent performance was mainly attributed to the coexistence of Sn-Nx and pyridinic N in Sn-g-C 3 N 4 /AC catalysts. The results of XPS, TG-DTG, C 2 H 2 -TPD, HCl adsorption, N 2 physisorption, and stability tests confirmed that the g-C 3 N 4 precursors can stabilize the Sn species provided by the alkylorganotin precursors, improve the thermal stability of Sn species and adsorption capacity of the resulted catalysts. In addition, the coke deposition is delayed over these alkylorganotin-based catalysts, which is favourable for a longer lifetime, as compared with Sn/AC. Among the 3 g-C 3 N 4 precursors, hexamethylenetetramine, having higher nitrogen content, proved to be the best g-C 3 N 4 precursor.

g-C 3 N 4 /AC preparation
Carbon supports were initially washed with 0.01 mol L −1 HCl to remove the impurities and then dried overnight at 100°C in an oven. The obtained carbon material was denoted as AC. N 1 (2.0 g) and AC (14.0 g) were mixed with ethanol (100 mL) and then stirred at 80°C for 3.5 h.
Afterwards, the mixture was dried overnight at 100°C. Finally, the sample was subjected to calcination at 550°C for 4 h to obtain the g 1 -C 3 N 4 /AC-550 sample. The other 2 catalysts (g 2 -C 3 N 4 /AC-550 and g 3 -C 3 N 4 /AC-550, respectively) were prepared in a similar way.

Sn-g-C 3 N 4 /AC preparation
Sn-g-C 3 N 4 /AC was prepared according to one of our previous studies by using an optimum molar ratio of 1 labelled as Sn-g 1 -C 3 N 4 /AC-450. The other 2 catalysts (Sn-g 2 -C 3 N 4 /AC and Sn-g 3 -C 3 N 4 /AC, respectively) were prepared in a similar way.

Catalyst characterization
Powder X-ray diffraction patterns were recorded on a Shimadzu XRD-6000 instrument with Cu-K α radiation operated at 40kV.
The textural properties of samples were analysed by nitrogen physisorption on a Nova2000e instrument (Quantachrome) after degassing the samples at 150°C for 3h.
Thermogravimetric analysis was performed on a NETZSCH STA 449F3 analyser. The temperature was increased from room temperature to 800°C at a heating rate of 15°C min −1 under air at a flow rate of 30 mL min −1 . X-ray photoelectron spectroscopy was performed with an EscaLab 250Xi spectrometer using a monochromatic Al K α source.
Acetylene temperature-programmed desorption (C 2 H 2 -TPD) was performed with a FINESORB-3010 chemisorption analyser. TPD experiments were carried out with~50 mg of sample, which was first treated at 200°C for 1.5 h under Ar. After cooling, it was continually flushed with C 2 H 2 at a flow rate of 25 mL min − 1 for 1 h and then heated from room temperature to 500°C at a heating rate of 10°C min − 1 .
Hydrogen chloride adsorption experiments were performed in a fixed-bed reactor. Catalysts were initially pretreated at 200°C for 1 h under Ar. Then, hydrogen chloride was fed into the reactor at a flow rate of 30 mL min − 1 for 1 h. Finally, the samples were heated from 200°C to 650°C under Ar, and desorbed hydrogen chloride was removed using deionized water (1000 mL). The amount of hydrogen chloride in the final solution was evaluated by titration [60].

Catalyst performance
The performance of catalysts (4.0 mL) was tested in a fix-bed reactor (d = 10mm). To activate the catalyst and remove the air and physisor bed water, hydrogen chloride gas was initially passed through the reaction system for 40 min. Then, a mixture of hydrogen chloride and acetylene (V HCl /V C2H2 = 1.1) was passed through the reactor (reaction temperature = 150°C-200°C, C 2 H 2 -GHSV = 30 h −1 ) . The unreacted hydrogen chloride in the product gas was adsorbed on limestone. The clean gas was then analysed online using a gas chromatograph equipped with a GDX-301column and TCD.