The Preparation of Cu-g-C3N4/AC Catalyst for Acetylene Hydrochlorination

A novel catalyst based on g-C3N4/activated carbon was prepared by adding CuCl2. The catalytic performance of the as-prepared catalyst was investigated in the acetylene hydrochlorination reaction. X-ray photoelectron spectroscopy, temperature programmed desorption, low temperature N2 adsorption/desorption (Brunauer–Emmett–Teller), and thermal gravity analysis showed that Cu-g-C3N4/AC significantly enhanced the catalytic performance of the original catalyst by increasing the relative pyrrolic N content. Cu-g-C3N4/AC also affected the adsorption of hydrogen chloride and acetylene, as well as inhibited the coke deposition during acetylene hydrochlorination.


Introduction
The use of plastic products is increasing to date. Polyvinyl chloride (PVC) has become the second most frequently used general plastic worldwide. Vinyl chloride monomer (VCM) is used for PVC synthesis. In China, use of acetylene as a raw material to produce VCM still dominates local practices [1]. Mercuric chloride (HgCl 2 ) supported on activated carbon (AC) is used as the catalyst in the current industrial production process. This approach holds major shortcomings: the catalyst loses volatile active components during the reaction, while its high toxicity is harmful to human health and seriously pollutes the environment. On the other hand, the catalyst cannot generally be reactivated after deactivation. To achieve the efficient and clean production of VCM, the development of new non-mercury catalyst must be urgently addressed [2].
Numerous attempts have been made to develop a non-mercury catalyst, and precious metal catalysts have attracted the attention of researchers, especially the gold-containing catalyst. Hutchings et al. [3] studied the catalytic mechanism of the Au catalyst and found that the activity of acetylene hydrochlorination was improved by adding adsorbed hydrogen chloride (HCl). The authors also noted that the increase in acetylene (C 2 H 2 ) adsorption can deactivate the catalyst. Specifically, HCl adsorption is the reaction rate control step. Li et al. [4] discovered that the enhanced catalytic performance could be attributed to the introduction of polypyrrole (PPy) into multiwall carbon nanotubes (MWCNTs), which increased the adsorption of HCl. Consequently, AuCl 3 /PPy-MWCNT demonstrated excellent catalytic performance compared with AuCl 3 /MWCNT during acetylene hydrochlorination. However, the price and stability of the Au-based catalyst limit the widespread use of this catalyst [5].
Carbon nitride material has recently attracted attention because of its excellent chemical stability and unique semiconductor band structure. Thus, carbon nitride material has been widely used for photolysis of aquatic hydrogen, which produces oxygen; light-catalyzed organic synthesis and the photocatalytic degradation of organic pollutants [6,7]. Dai et al. [8] searched for a graphitic carbon nitride catalyst (g-C 3 N 4 /AC), their study revealed that the nitrogen atom of the g-C 3 N 4 /AC catalyst is the active site for the HCl; thus, the prepared g-C 3 N 4 /AC catalyst displayed considerable catalytic performance. However, this catalyst rapidly deactivates given its poor stability. This catalyst entails further study.
To solve the high cost and rapid deactivation of catalyst, we introduced the non-precious metal CuCl 2 to the g-C 3 N 4 /AC catalyst by a simple preparation method. The properties of this new catalyst were explored by acetylene hydrochlorination. Characterization confirmed the success of material preparation; the test results showed that the catalyst possessed good catalytic activity, thereby promoting further research on experimental and theoretical references.

Catalytic Performance
The catalytic activities of g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC are shown in Figure 1. Reaction conditions were as follows: temperature (T) = 180 • C; C 2 H 2 hourly space velocity (GHSV) = 72 h −1 ; Feed volume ratio V(HCl)/V(C 2 H 2 ) = 1.15. As a carrier of acetylene hydrochlorination reaction, AC exhibited an initial activity of up to 45%, which declined to 20% after 450 min of reaction ( Figure 1a). Under the same conditions, the acetylene conversion of the g-C 3 N 4 /AC catalyst decreased from 67% to 55% after 450 min of reaction. Therefore, the g-C 3 N 4 /AC catalyst enhanced the acetylene hydrochlorination reaction to some extent compared to AC. The acetylene conversion by the Cu-g-C 3 N 4 /AC catalyst reached 79%, and remained almost constant after 450 min of reaction. Figure 1b displays the selectivity to vinyl chloride in the acetylene hydrochlorination reaction, where the selectivity to VCM of the g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC catalyst exceed 99.5%. The above-mentioned results show that g-C 3 N 4 /AC has better catalytic activity, and Cu-g-C 3 N 4 /AC catalytic performance is further improved compared with AC.   Figure 2 is the TEM image of the morphology and structure of g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC. The layered plate junction of the g-C 3 N 4 load on the carrier can be clearly observed, and this junction is consistent with results reported in the literature [8] (Figure 2a). Figure 2b is notably similar to Figure 2a. A layered structure was produced after CuCl 2 was added to g-C 3 N 4 /AC, which suggests that CuCl 2 did not alter the texture or structure of carbon-nitrogen compounds. The TEM elemental mapping images of Cu-g-C 3 N 4 /AC confirm the presence of C, N, O, and Cu in the Cu-g-C 3 N 4 /AC catalyst with homogeneously distribution (Figure 2c). Importantly, Cu and N highly overlay in the carbon structure from the elemental mapping signal, implying the presence of Cu-N coordination [9,10].  Figure 2 is the TEM image of the morphology and structure of g-C3N4/AC and Cu-g-C3N4/AC. The layered plate junction of the g-C3N4 load on the carrier can be clearly observed, and this junction is consistent with results reported in the literature [8] (Figure 2a). Figure 2b is notably similar to Figure 2a. A layered structure was produced after CuCl2 was added to g-C3N4/AC, which suggests that CuCl2 did not alter the texture or structure of carbon-nitrogen compounds. The TEM elemental mapping images of Cu-g-C3N4/AC confirm the presence of C, N, O, and Cu in the Cu-g-C3N4/AC catalyst with homogeneously distribution (Figure 2c). Importantly, Cu and N highly overlay in the carbon structure from the elemental mapping signal, implying the presence of Cu-N coordination [9,10].   Figure 3 shows the XRD spectra of g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC before and after AC loading. As shown in Figure 2a, two obvious diffraction peaks at 12.9 • and 27.4 • were observed for g-C 3 N 4 /AC. The strong peak at 27.19 • is the (002) characteristic peak for g-C 3 N 4 , which is a characteristic inter planar stacking peak of aromatic systems. The peak at 12.9 • is the (100) characteristic peak for g-C 3 N 4 , which is typical of highly oriented melons [11]. The characteristic peak for the copper-modified g-C 3 N 4 is slightly wider and weaker than that of pure g-C 3 N 4 [12]. Notably, the XRD spectra of Cu-g-C 3 N 4 /AC do not show the characteristic peak related to the copper element, such as that of copper or copper oxide. This observation suggests that the metal and g-C 3 N 4 /AC formed a chemical complex (M-N) [13], which is consistent with the TEM mapping results. The two obvious characteristic peaks for AC at 2θ = 24.4 • and 43.7 • in Figure 2b, corresponding to the (002) and (101) crystal planes [14]. Moreover, the broad peaks for the g-C 3 N 4 /AC catalysts were positively shifted when g-C 3 N 4 was thoroughly dispersed on the support surface. A similar finding was noted with the Cu-g-C 3 N 4 /AC catalyst.  Figure 3 shows the XRD spectra of g-C3N4/AC and Cu-g-C3N4/AC before and after AC loading. As shown in Figure 2a, two obvious diffraction peaks at 12.9° and 27.4° were observed for g-C3N4/AC. The strong peak at 27.19° is the (002) characteristic peak for g-C3N4, which is a characteristic inter planar stacking peak of aromatic systems. The peak at 12.9° is the (100) characteristic peak for g-C3N4, which is typical of highly oriented melons [11]. The characteristic peak for the copper-modified g-C3N4 is slightly wider and weaker than that of pure g-C3N4 [12]. Notably, the XRD spectra of Cu-g-C3N4/AC do not show the characteristic peak related to the copper element, such as that of copper or copper oxide. This observation suggests that the metal and g-C3N4/AC formed a chemical complex (M-N) [13], which is consistent with the TEM mapping results. The two obvious characteristic peaks for AC at 2θ = 24.4° and 43.7° in Figure 2b, corresponding to the (002) and (101) crystal planes [14]. Moreover, the broad peaks for the g-C3N4/AC catalysts were positively shifted when g-C3N4 was thoroughly dispersed on the support surface. A similar finding was noted with the Cu-g-C3N4/AC catalyst.  Figure 4 shows the Raman spectroscopy of the AC, g-C3N4/AC, and Cu-g-C3N4/AC catalysts. The ID/IG ratios for the AC, g-C3N4/AC, and Cu-g-C3N4/AC catalysts are 1.097, 1.187, and 1.253, respectively. The G band of the AC was red-shifted after Cu-g-C3N4 was supported on the AC, implying that electron transfer occurs between the Cu species and the g-C3N4 matrix [15]. The Raman spectrum analysis results suggest that the Cu species are chemically coordinated to the g-C3N4 matrix.  Figure 4 shows the Raman spectroscopy of the AC, g-C 3 N 4 /AC, and Cu-g-C 3 N 4 /AC catalysts. The I D /I G ratios for the AC, g-C 3 N 4 /AC, and Cu-g-C 3 N 4 /AC catalysts are 1.097, 1.187, and 1.253, respectively. The G band of the AC was red-shifted after Cu-g-C 3 N 4 was supported on the AC, implying that electron transfer occurs between the Cu species and the g-C 3 N 4 matrix [15]. The Raman spectrum analysis results suggest that the Cu species are chemically coordinated to the g-C 3 N 4 matrix.

X-ray Photoelectron Spectroscopy (XPS)
The catalyst's elemental content and chemical composition, as well as the states of the nitrogen atoms in the catalysts, were characterized with XPS. The XPS results are shown in Table 1 and Figure 5. Table 1 lists the elemental contents of the AC, g-C3N4/AC, and Cu-g-C3N4/AC catalysts. The copper content of the Cu-g-C3N4/AC catalyst is 4.15%, which is lower than the actual value (4.3%). The nitrogen content of the g-C3N4/AC and Cu-g-C3N4/AC catalysts is nearly close to that listed in Table 1. The C1s, N1s, and O1s signals were observed from the XPS survey spectrum of the g-C3N4/AC and Cu-g-C3N4/AC catalysts (Figure 5a), proving that C, N, and O are present in the catalyst. Furthermore, a Cu2p signal was detected from the Cu-g-C3N4/AC catalyst. The above-mentioned results show that the Cu species were successfully introduced into the g-C3N4/AC material. The XPS deconvolution of the N1s of g-C3N4/AC and Cu-g-C3N4/AC is shown in Figure 5b, and Table 1 presents related contents of nitrogen species of the catalyst. Graphitic N, pyrrolic N, and pyridinic N were obtained. The graphitic N corresponded to nitrogen atoms that, in the grapheme layer, linked to three carbon atoms, the pyrrolic N bonds in the five-membered rings had two carbon atoms, and the pyridinic N bonds had one p-electron localized with two carbon atoms [16,17]. The contents of pyridinic N, pyrrolic N, and graphitic N were 49.21%, 22.56%, and 28.22%, respectively ( Table 2). These results indicate that pyridinic N is the main N style in the g-C3N4/AC catalyst. For the Cu-g-C3N4/AC catalyst, the contents of pyrrolic N increased to 40.63%, which is due to the fact that the transitional metal of Cu can promote the formation of pyrrolic N-species [18]. Related research has revealed that the type of N, particularly pyrrolic N, greatly influences the catalytic performance of catalysts in an acetylene hydrochlorination reaction. A previous investigation by our group demonstrated that pyrrolic N ranked first in the order of nitrogen species' role for this reaction [19]. Bao et al. also reported, based on experimental and theoritical results, that pyrrolic N in carbon is the active site for acetylene hydrochlorination [20]. Therefore, the pyrrolic N content of the Cu-g-C3N4/AC catalyst increased relative to that of the g-C3N4/AC catalyst, which is consistent with the catalytic activity.

X-ray Photoelectron Spectroscopy (XPS)
The catalyst's elemental content and chemical composition, as well as the states of the nitrogen atoms in the catalysts, were characterized with XPS. The XPS results are shown in Table 1 and Figure 5. Table 1 lists the elemental contents of the AC, g-C 3 N 4 /AC, and Cu-g-C 3 N 4 /AC catalysts. The copper content of the Cu-g-C 3 N 4 /AC catalyst is 4.15%, which is lower than the actual value (4.3%). The nitrogen content of the g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC catalysts is nearly close to that listed in Table 1. The C 1s , N 1s , and O 1s signals were observed from the XPS survey spectrum of the g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC catalysts (Figure 5a), proving that C, N, and O are present in the catalyst. Furthermore, a Cu 2p signal was detected from the Cu-g-C 3 N 4 /AC catalyst. The above-mentioned results show that the Cu species were successfully introduced into the g-C 3 N 4 /AC material. The XPS deconvolution of the N 1s of g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC is shown in Figure 5b, and Table 1 presents related contents of nitrogen species of the catalyst. Graphitic N, pyrrolic N, and pyridinic N were obtained. The graphitic N corresponded to nitrogen atoms that, in the grapheme layer, linked to three carbon atoms, the pyrrolic N bonds in the five-membered rings had two carbon atoms, and the pyridinic N bonds had one p-electron localized with two carbon atoms [16,17]. The contents of pyridinic N, pyrrolic N, and graphitic N were 49.21%, 22.56%, and 28.22%, respectively ( Table 2). These results indicate that pyridinic N is the main N style in the g-C 3 N 4 /AC catalyst. For the Cu-g-C 3 N 4 /AC catalyst, the contents of pyrrolic N increased to 40.63%, which is due to the fact that the transitional metal of Cu can promote the formation of pyrrolic N-species [18]. Related research has revealed that the type of N, particularly pyrrolic N, greatly influences the catalytic performance of catalysts in an acetylene hydrochlorination reaction. A previous investigation by our group demonstrated that pyrrolic N ranked first in the order of nitrogen species' role for this reaction [19]. Bao et al. also reported, based on experimental and theoritical results, that pyrrolic N in carbon is the active site for acetylene hydrochlorination [20]. Therefore, the pyrrolic N content of the Cu-g-C 3 N 4 /AC catalyst increased relative to that of the g-C 3 N 4 /AC catalyst, which is consistent with the catalytic activity.

Temperature-Programmed Desorption (TPD)
The TPD characterization of the catalysts was conducted to analyze their adsorption of reactants ( Figure 6). The desorption temperature shows the adsorption strength, and the desorption peak area is widely believed to represent the amount of adsorption. In Figure 6a, the HCl desorption temperature of g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC are 200.5 and 215 • C, respectively. The desorption peak area of the Cu-g-C 3 N 4 /AC catalyst was larger than that of the g-C 3 N 4 /AC catalyst. That is to say, the Cu-g-C 3 N 4 /AC catalyst has a greater capacity to absorb HCl, which is consistent with its relative catalytic activity in the acetylene hydrochlorination reaction. In Figure 6b, the C 2 H 2 desorption temperature of both g-C 3 N 4 /AC and Cu-g-C 3 N 4 /AC is 145 • C. However, the desorption peak area of the Cu-g-C 3 N 4 /AC catalyst is smaller than that of the g-C 3 N 4 /AC catalyst. The TPD results indicate that the CuCl 2 -modified g-C 3 N 4 /AC catalyst improves the ability for HCl adsorption, but greatly weakens the ability for C 2 H 2 adsorption.

Temperature-Programmed Desorption (TPD)
The TPD characterization of the catalysts was conducted to analyze their adsorption of reactants ( Figure 6). The desorption temperature shows the adsorption strength, and the desorption peak area is widely believed to represent the amount of adsorption. In Figure 6a, the HCl desorption temperature of g-C3N4/AC and Cu-g-C3N4/AC are 200.5 and 215 °C, respectively. The desorption peak area of the Cu-g-C3N4/AC catalyst was larger than that of the g-C3N4/AC catalyst. That is to say, the Cu-g-C3N4/AC catalyst has a greater capacity to absorb HCl, which is consistent with its relative catalytic activity in the acetylene hydrochlorination reaction. In Figure 6b, the C2H2 desorption temperature of both g-C3N4/AC and Cu-g-C3N4/AC is 145 °C. However, the desorption peak area of the Cu-g-C3N4/AC catalyst is smaller than that of the g-C3N4/AC catalyst. The TPD results indicate that the CuCl2-modified g-C3N4/AC catalyst improves the ability for HCl adsorption, but greatly weakens the ability for C2H2 adsorption.

Low-Temperature N 2 Adsorption/Desorption Experiment (BET)
The pore structure parameters of the pores in the fresh and used catalysts are shown in Table 3. After introducing CuCl 2 to the g-C 3 N 4 /AC catalyst, the BET surface area, the average pore diameter, and the total pore volume of the g-C 3 N 4 /AC mainly decreased because the active components covered part of the space, decreasing from 857.68 to 521.26 m 2 ·g −1 , and the total pore volume decreased from 0.47 to 0.29 cm 3 ·g −1 . By contrast, the BET surface area and total pore volume for the Cu-g-C 3 N 4 /AC catalyst also decreased. This result may have been the result of the carbon deposition on the catalyst or pore [21].

Thermal Gravity Analysis (TG-DTG)
Thermal gravity analysis is conducted under oxidizing conditions, and this analysis is used to obtain information regarding carbonaceous material deposition on catalysts [22]. The extent of carbon deposition on the catalysts is shown in Table 4. The extent of carbon deposition decreased for the Cu-g-C 3 N 4 /AC catalyst. Specifically, the effect of Cu(II) addition on the inhibition of carbon deposition was conspicuous to some extent.

Catalyst Preparation
The catalysts were prepared as reported previously [12]. First, at 80 • C, dicyandiamide was mixed with deionized water, stirred for 30 min. AC was added to the above solution. Remove water by heating. The sample was calcined at 500 • C for 4 h under a flowing nitrogen atmosphere. The sample was denoted as g-C 3 N 4 /AC. The preparation of Cu-g-C 3 N 4 /AC was similar to that of g-C 3 N 4 /AC, but CuCl 2 ·2H 2 O was added after dicyandiamide was mixed with deionized water.

Characterization of Physical Parameters
Low-temperature N 2 adsorption/desorption experiments on surface area and porosity were performed with a Micromeritics ASAP 2020C instrument (Micromeritics Instrument Ltd., Norcross, GA, USA). The samples were first degassed at 150 • C and analyzed at −196 • C. Raman spectrum analyses were conducted using a Raman spectrometer (Fei Tecai G2F20, FEI, Hillsboro, OR, USA) with a 633 nm Ar laser. XRD patterns were collected with a Bruker D8 advanced X-ray diffractometer (Bruker Biosciences Corporation, Billerica, MA, USA) at wide (10 • -90 • in 2θ) angles. TEM was used to examine the sample morphologies with a JEM2010 electron microscope (FEI, Hillsboro, OR, USA). XPS analysis was conducted using a Kratos AXIS Ultra DLD spectrometer (Kratos, Manchester, UK). TPD was performed with a Micromeritic chemisorb 2720 (Micromeritic ASAP 2720, Micromeritics Instrument Ltd, Norcross, GA, USA). The temperature was increased from 50 to 500 • C. TG analysis of the samples was conducted with a TG DSC simultaneous thermal analyzer (NETZSCH STA 449F3 Jupiter1, Netzsch, Selb, Germany) in air. The temperature was increased from 50 to 850 • C at a heating rate of 10 • C·min −1 .

Catalytic Performance Evaluation
The catalytic activity tests were performed in a fixed-bed micro-reactor (internal diameter, 10 mm). Prior to the reaction, the system was purged with nitrogen gas to remove water and air for approximately 30 min. Subsequently, the catalyst was activated by passing HCl gas through the reactor. After the temperature was heated to 180 • C, C 2 H 2 (2.1 mL·min −1 ) and HCl (2.4 mL·min −1 ) were fed into the heated reactor, which contained 2 mL of catalyst, with a GHSV of 72 h −1 . Gas chromatography (GC-2014C, Shimadzu, Kyoto, Japan) was used for the reaction products.

Conclusions
In this paper, a Cu-g-C 3 N 4 /AC catalyst was prepared with a simple method for an acetylene hydrochlorination reaction. The evaluation of catalytic performance showed that the Cu-g-C 3 N 4 /AC catalyst's catalytic performance was superior to the g-C 3 N 4 /AC catalyst. The introduction of CuCl 2 to the g-C 3 N 4 /AC catalyst increased the relative pyrrolic N content, and enhanced the adsorption capacity of HCl. Moreover, CuCl 2 inhibited coke deposition on the catalyst.