Citation

Lu Yusheng, Huang Chaofeng, Lei Zhigang, Zhu Mingyuan. Catalytic effects of structural design in N-modified carbon materials for the hydrochlorination of acetylene[J]. Chin. Chem. Lett., 2025, 36(8): 110583  

Yusheng Lu^a, Chaofeng Huang^a, Zhigang Lei^a,b,*, Mingyuan Zhu^c,*     

Polyvinyl chloride (PVC), with its high resistance to light and chemical degradation, is a material that is widely used in various industries and relevant to people's lives [1]. Mercury-based catalysts are most commonly used in industrial carbide acetylene reactions [2]. Mercury chloride is unstable at high temperatures and is extremely susceptible to loss, leading to increasing environmental problems [3]. Therefore, it is very urgent to develop non-mercury green catalysts. Carbon materials exhibit distinctive physical and chemical characteristics, including high specific surface area, high defect density and adjustable structure [4]. These properties render them a highly promising catalyst for the replacement of traditional catalysts. Consequently, carbon materials are widely used by in acetylene hydrochlorination.

During the recent years, researchers have prepared heteroatom-doped carbon materials by employing various precursors, such as graphene [5], carbon nanotubes [6], polymers [7], biomass [8-10], metal-organic skeletons [11, 12] and ionic liquids [13]. Zhu et al. synthesized the D-GH-800 catalyst with D(+)-glucosamine hydrochloride to achieve the initial C₂H₂ conversion of 99%. The C₂H₂ conversion decreased by 30% after 60 h at 220 ℃ and 50 h^-1. Meanwhile, Zhang et al. [14] synthesized carbon nanospheres with porous structure using a self-templated method, which achieved the initial C₂H₂ conversion and the deactivation rate of 86%/h and 0.13%/h at 180 ℃ and 30 h^-1, respectively. These non-metallic carbon materials exhibited remarkable catalytic activity in acetylene hydrochlorination, which could almost compete with metal catalysts. However, the stability is still an obstacle to the progress of industrialization.

Bao et al. [15] investigated the deactivation mechanism of the PDA/SiC nanocomposites catalyst and suggested that the generation of carbon deposits is the main cause of catalyst deactivation. In order to alleviate the problem of carbon deposition, researchers had found that structural design of catalysts could be effective. Zhu et al. [16] changed the structural parameters of the carbon-nitrogen material by using ZnCl₂ as activating and reaming agent, which significantly enhanced catalytic performance. Zhao et al. [17] synthesized layered N-doped materials using ionic liquids and MgCl₂ as templates and casein as N source, which demonstrated good catalytic performance in acetylene hydrochlorination. Li et al. [18] synthesized sulfur and nitrogen co-doped carbon materials using glucose as carbon source and SBA-15 as hard template, and the mesoporous structure with high specific surface area and high N content enhanced the catalytic stability in acetylene hydrochlorination. Bao et al. [19] synthesized N-doped ordered mesoporous carbon using hard template SBA-15. No significant deactivation was detected in 100 h, indicating that the mesoporous structure could play a role in inhibiting carbon accumulation. By studying the structure of materials, it could be found that it could take a positive effect on the catalyst stability. It had been shown that microporous structure of catalyst facilitates the exposure of active sites [20] and mesoporous structure facilitates the mitigation of carbon deposition [21].

However, that is usually requires templating agents or activators for the structural modulation of carbon materials. The synthesis process is not only a complicated procedure, but also requires strong acid/base etching post-treatment, which severely limits its development. Therefore, it is very necessary to develop a green, efficient synthesis of N-doped carbon materials with high N content and tunable structure. Based on this, we synthesized structurally tunable hierarchically porous structure and defect-rich N-modified carbon materials by a one-pot hydrothermal method using cheap glucose as carbon source, following the principle of green chemistry. m-Phenylenediamine could be used not only as N source, but also as cross-linking agent to modulate the structure of material during preparation process. The carbon materials achieved excellent activity and stability in acetylene hydrochlorination, and the potential active sites and catalytic mechanism of the catalysts were revealed by various experimental characterizations and density functional theory. This work provides a new direction for future structural design of carbon materials.

Firstly, the morphology and structure of catalysts were observed and analyzed by scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffractometer (XRD) and Raman characterizations. As shown in Figs. 1a and b, it could be clearly observed that the C-800 material has a regular dense spherical structure, which was formed by the self-polymerization of glucose during the hydrothermal process [22]. A blocky structure resulting from the incorporation of mPDA was observed in amine-modulated derived carbon materials in Figs. 1c-f and Fig. S1 (Supporting information). This is due to the fact that the polycondensation reaction between amino group of mPDA and aldehyde group of glucose inhibited polymerization of glucose itself [23], thus disrupting the generation of globular structure. Furthermore, all amine-modulated carbon materials were composed of irregular nanoparticles, which allowed N element to become better distributed. The morphological structure of 0.3PmPD-C material did not change significantly after heat treatment as shown in Figs. 1e and f, indicating that high temperature pyrolysis did not destroy the original morphological structure of carbon materials. In Fig. S2 (Supporting information), all carbon materials had no obvious ordered carbon lattice stripes inside, indicating that carbon materials were amorphous structure. The edges of hydrothermally treated 0.3PmPD-C catalyst were graphitized and formed regular stripes with a spacing of 0.42 nm. However, the layer streaks at the edges of 0.3PmPD-C-800 materials were much more abundant and ordered, suggesting that high temperature carbonization further enhanced the graphitization of materials. The carbonization treatment accelerated the conversion of 0.3PmPD-C carbon precursor to carbon material. As shown in Fig. 1g and Fig. S3 (Supporting information), the Mapping images and XPS full spectrum diagram of 0.3PmPD-C-800 catalyst showed that C, N and O elements were present and uniformly distributed, which showed no other impure elements. This is the fact that mPDA not only acted as a cross-linker during glucose polymerization, but also as an N source.

Figure 1

	Download: JPG larger image
Fig. 1. SEM images of (a, b) C-800, (c, d) 0.3PmPD-C, (e, f) 0.3PmPD-C-800 catalysts. (g) The HAADF-mapping images of 0.3PmPD-C-800 catalyst.

The crystallinity and phase structure characteristics of carbon materials could be further confirmed by XRD and Raman spectroscopy. The XRD patterns showed (002) diffraction crystal plane corresponding to disordered carbon layer around 2θ = 25° and (101) diffraction crystal plane for graphite lattice at 2θ = 43° [24]. Broad diffraction peaks were observed for all catalyst samples in Fig. S4a (Supporting information), which was a typical diffraction of an amorphous structure, suggesting that the materials are amorphous carbon [25]. Furthermore, compared with 0.3PmPD-C catalyst, the (101) diffraction peaks of the 0.3PmPD-C-800 catalyst were more pronounced and had higher graphitization, indicating that the carbonization treatment further accelerated the graphitization of the material. Raman spectra could be used to analyze defects of material. In Fig. S4b (Supporting information), it could be found that D peak at 1350 cm^-1 represents disorder/defects in carbon material and G peak at 1580 cm^-1 represents sp² structure in ordered graphite structure [26]. In general, I_D/I_G values indicate the degree of defects. There was basically no difference in the I_D/I_G values of xPmPD-C-800 catalysts, indicating that the addition of mPDA did not have a significant effect on defects and crystal structure [27]. However, compared with the xPmPD-C-800 catalysts, the C-800 catalyst exhibited the lowest value of I_D/I_G (0.98). This could be attributed the fact that the xPmPD-C-800 catalysts had more defects due to the introduction of N.

The N₂ adsorption/desorption isotherms of all catalysts were shown in Fig. 2a. According to the IUPAC classification, 0.3PmPD-C and xPmPD-C-800 catalysts exhibited a combination of type Ⅰ and type Ⅳ isotherms, indicating the presence of both micro- and mesoporous structures in the catalysts. The distinctive feature was that the N₂ adsorption showed a sharp increase at low relative pressures (P/P₀ < 0.1), indicating an abundant microporous structure. Secondly, a significant hysteresis loop occurs at high pressures, which was related to the presence of mesopores. The adsorption isotherm of 0.3PmPD-C-800 catalyst after pyrolysis was significantly higher than that of 0.3PmPD-C catalyst, indicating that the 0.3PmPD-C-800 catalyst has higher specific surface area, pore volume, and especially micropores. As shown in the Table S1 (Supporting information), the 0.3PmPD-C catalyst had the lowest specific surface area, microporous specific surface area, total pore volume and microporous volume of 208 m²/g, 23 m²/g, 0.22 cm³/g and 0.01 cm³/g, respectively. After pyrolysis treatment, the specific surface area, microporous specific surface area, total pore volume and microporous volume of 0.3PmPD-C catalyst were elevated to 446 m²/g, 255 m²/g, 0.34 cm³/g and 0.14 cm³/g, respectively. This was mainly due to the fact that unstable groups and backbone atoms during high-temperature pyrolysis were lost and tiny holes were formed, which could be linked to a decrease of the average pore size. Interestingly, the catalysts showed H4-type hysteresis loops at high relative pressures (P/P₀ = 0.40–0.99), indicating the presence of both mesopores and micropores in these catalysts. However, the 0.1PmPD-C-800 catalyst showed a different H3-type hysteresis loop, which was due to the formation of large pores by the stacking of catalyst particles, which could also be confirmed by TEM as shown in Fig. S1c (Supporting information). The pore size distribution curve shown in Fig. 2b further confirmed these results. It was clear that all catalysts contained mesopores and micropores, indicating a hierarchical pore structure of the catalyst.

Figure 2

	Download: JPG larger image
Fig. 2. (a) The adsorption/desorption isotherm and (b) the pore size distribution of 0.3PmPD-C and xPmPD-C-800 catalysts.

It is well known that the synthesis conditions are important factors affecting the structural properties of porous polymers [28, 29]. The amount of cross-linking agent mPDA affects the degree of cross-linking of carbon precursors, which has an impact on the carbonization process of the catalyst and the pore structure of the catalyst [30]. The specific surface area of the catalysts showed a varying degree of decrease as the mPDA dosage was elevated, which was mainly caused by the clogging of the pores due to polymerization of excess mPDA [31]. This was further confirmed by the obvious finding from the pore size distribution in Fig. 2b that the mesopore distribution was gradually shifting towards smaller pore sizes.

In summary, it was shown that the introduction of mPDA could not only serve as a N source for heteroatom doping, but also had an important effect on the carbon network and pore structure of the catalyst. These structures containing micropores and mesoporous could facilitate reactive gas adsorption, mass transfer and promote kinetic processes in multi-phase catalytic applications [32, 33].

In order to assess the performance of catalysts in acetylene hydrochlorination, a series of catalysts treated with different mPDA content were evaluated. Reaction conditions: T = 220 ℃, GHSV(C₂H₂) = 150 h^-1, V_HCl/V_C = 1.15. As shown in Fig. 3a, it could be found that the initial C₂H₂ conversion of C-800, 0.3PmPD-C and 0.3PmPD-C-800 catalysts were 45.28%, 63.39% and 96.38%, respectively. Comparing the un-pyrolyzed 0.3PmPD-C and un-modified C-800 catalysts, the 0.3PmPD-C-800 catalyst showed a significant enhancement, suggesting that both N introduction and high-temperature treatment had some effect on the catalytic performance. The initial C₂H₂ conversion of all xPmPD-C-800 catalysts reaching > 80% and the selectivity of vinyl chloride (VCM) close to 100% in Fig. S5a (Supporting information). As shown in Fig. 3b, the average deactivation rate of catalyst gradually decreased with increasing molar ratio (x) of mPDA and D(+)-glucose. The 0.3PmPD-C-800 catalyst had the lowest deactivation rate of 0.434%/h. However, the stability of catalyst was inhibited at molar ratio greater than 0.3, which could be attributed to the polymerization of the excess amine blocking the original pores and inhibiting the gas transport during the reaction. Subsequently, the C₂H₂ conversion and space-time yield (STY) of the optimal 0.3PmPD-C-800 catalyst was evaluated at different gas hourly space velocity, as shown in Figs. S5b and c (Supporting information). The STY of 0.3PmPD-C-800 catalyst could reach 0.138 g_VCMmL_cat^-1h^-1 at 220 ℃ and 50 h^-1. Furthermore, as shown Fig. S5d (Supporting information), the stability of the C-800 and 0.3PmPD-C-800 catalysts was tested under reaction conditions (T = 220 ℃, GHSV(C₂H₂) = 30 h^-1, V_HCl/V_C = 1.15). The 0.3PmPD-C-800 catalyst exhibited good stability, with the C₂H₂ conversion maintained at around 98% in 100 h. However, the initial C₂H₂ conversion of C-800 catalyst could only reach 74.32%, which was directly reduced by 41.65% after 100 h, with very severe deactivation.

Figure 3

	Download: JPG larger image
Fig. 3. (a) The C2H2 conversion and (b) deactivation rates of C-800 (A), 0.3PmPD-C (B) and xPmPD-C-800 (C-I) catalysts. (c) The STY and (d) deactivation rates of the C-800 and 0.3PmPD-C-800 catalysts and other catalysts reported in 220 ℃.

For comparison with previous work in the field, we calculated the STY and average deactivation rate of the catalysts under similar reaction conditions, as shown in Figs. 3c and d, Tables S2 and S3 (Supporting information). Clearly, the STY of 0.3PmPD-C-800 catalyst is higher than most of the catalysts reported in the literature, demonstrating excellent catalytic activity. The STY of 0.3PmPD-C-800 catalyst is essentially the same as that of the D-GH-800 (No. 1), CN-2 (No. 9) and NPC-800 (No. 16) catalysts. In addition, comparing the average deactivation rates of the catalysts, the 0.3PmPD-C-800 catalyst has a very low deactivation rate, which is significantly lower than that of the D-GH-800, CN-2 and NPC-800 catalysts. In summary, by comparing the STY and the average deactivation rate of catalysts, it was shown that the 0.3PmPD-C-800 catalyst has very excellent catalytic performance.

In order to analyze the factors affecting the activity and stability of the catalysts, it was explored the relationship between N content, S_micro/S_BET and catalytic performance. As the amount of mPDA added increased, xPmPD-C-800 catalysts showed an increasing proportion of N in Table S4 (Supporting information). When the proportion reached 0.4, the N element content showed a slight change. As shown in Fig. 4a, it was clearly found that the N content was positively correlated with the initial C₂H₂ conversion. Both N content and pore structure are important factors affecting the stability of the catalyst, as shown in Fig. 4b. When the structural parameters of the catalysts were similar, it was clearly found that the elevation of the N content significantly suppressed the catalyst deactivation; when the N content was kept the same, the percentage of microporous structures in the catalysts was positively correlated with the average deactivation rate, and the deactivation was accelerated by the excessive microporous structures. The synergistic effect of high N content and moderate microporous structure significantly enhanced the activity and stability of the 0.3PmPD-C-800 catalyst.

Figure 4

	Download: JPG larger image
Fig. 4. (a) The relationship between N content and C2H2 conversion. (b) The relationship between Smicro/SBET, N content and the average deactivation rates. (c) The relationship between content of different N species and C2H2 conversion in xPmPD-C-800 catalysts. (d) The reaction energy profile and bond distances of the substances with pyridinic N in acetylene hydrochlorination.

The catalytic sites of carbon-nitrogen materials in the hydrochlorination of acetylene were investigated. The high-resolution N 1s spectrum of xPmPD-C-800 shown in Fig. S6 (Supporting information), the four peaks of pyridinic N, pyrrolic N, graphitic N and oxidized N could be fitted to be at 398.4, 400.1, 401.0 and 402.6 eV, respectively [26]. The respective proportions of N species were revealed in Fig. 4c. From the comparison of relationship between different N species and the initial C₂H₂ conversion, it could be noticed that only pyridinic N content was positively correlated with the C₂H₂ conversion. It indicated that pyridinic N was the main active site. Meanwhile, the adsorption of reactants and the activation capacity of active sites are very crucial for catalytic reactions [34]. As shown in Fig. S7a (Supporting information), the HCl-TPD revealed that xPmPD-C-800 samples showed a great enhancement in HCl adsorption compared to C-800, which was extremely favorable for reaction. The enhancement of HCl adsorption capacity was mainly due to the fact that the introduction of N enhanced the number of adsorption sites in xPmPD-C-800 catalysts. In addition, the introduction of N improved desorption capacity, which was mainly related to adsorption and activation of HCl by electron cloud density around N atoms. The C₂H₂-TPD in Fig. S7b (Supporting information) showed that desorption capacity of xPmPD-C-800 catalysts were lower than that of the C-800 catalyst, indicating that the introduction of N species reduced adsorption capacity for C₂H₂. The lower desorption temperature and moderate adsorption capacity of reaction gas C₂H₂ were known to inhibit formation of coke deposition [34, 35] and thus to facilitate electrophilic addition reaction. In addition, the C₂H₂ desorption temperature (183.9 ℃) was significantly lower than HCl binding temperature (214.7 ℃) in 0.3PmPD-C-800 catalyst, indicating that the catalyst has a strong adsorption capacity for HCl.

To investigate catalytic mechanism of different N sites in acetylene hydrochlorination, four typical model structures of pyridinic N, pyrrolic N, graphitic N and oxidized N were calculated by density functional theory (DFT) in Fig. S8 (Supporting information). As shown in Fig. S9a (Supporting information), the adsorption capacity of all N sites for HCl was greater than that for C₂H₂, which was consistent with the results of TPD. The adsorption energies of HCl and C₂H₂ on pyridinic N sites were −12.83 and −5.28 kcal/mol, respectively. The adsorption energy of HCl in pyridinic N showed more negative values, indicating that pyridinic N site preferred to adsorb HCl rather than C₂H₂ [36]. From the changes in bond lengths of HCl and C₂H₂ before and after adsorption, the bond lengths of C₂H₂ changed weakly in Fig. S8. However, changes in bond lengths for HCl adsorption at different N sites: pyridinic N (1.39 Å) > oxidized N (1.37 Å) > pyrrolic N (1.31 Å) > graphitic N (1.29 Å), and that was a greater variation compared with the original bond length of HCl (1.29 Å). Once again, it was demonstrated that pyridinic N played a preferential role for activating HCl [37]. As shown in Table S5 (Supporting information), the electrons transferred by pyridinic N to HCl and C₂H₂ were 0.239 eV and 0.076 eV by Mulliken charge analysis, respectively. The results showed that pyridinic N transfers electrons to HCl more readily compared to C₂H₂. This result was consistent with the adsorption energy (E_a), suggesting that pyridinic N preferentially activated HCl molecule over C₂H₂ [38]. The projected density of states showed that pyridinic N, pyrrolic N, graphitic N and oxidized N could all exhibit localized electronic states below Fermi energy level in Fig. S9b (Supporting information). Pyridinic N species had a higher electron density and energy level, which was very favorable for the adsorption of HCl in reaction gas [39]. As shown in Fig. S10 (Supporting information), the density of states of all N sites showed overlapping peaks after adsorption of HCl, suggesting that electron redistribution occurred after formation of adsorption bonds. The bonding orbitals of pyridinic N site had higher energies than those of other N species, demonstrating that pyridinic N site had a higher adsorption energy for HCl, which was consistent with adsorption energy data.

The possible reaction pathways were simulated using the established Re model. Fig. 4d showed the ground state structure and energy distribution of the atoms during the reaction. When C₂H₂ and HCl molecules were together in close proximity to pyridinic N site together, the H atom of HCl was preferentially captured by the lone pair of electrons on N atom in pyridinic N. The formation of hydrogen bond between lone pairs of electrons and H atom resulted in bond length of HCl being stretched from 1.290 Å to 1.412 Å. Co-ads were formed and −18.67 kcal/mol energy was released. Subsequently, C₂H₂ molecule attacked the activated HCl molecule, and one C atom of C₂H₂ molecule was attached to the Cl atom of HCl to form transition state Ts1. This process required the provision of external supply and total energy barrier of 20.54 kcal/mol, which was the key control step in acetylene hydrochlorination reaction. Subsequently, the C₂H₂ molecule formed a solid chemical bond with Cl atom to give the intermediate Im1, and the H atom on N atom of the intermediate was attached to another carbon atom of C₂H₂ to form transition state Ts2. With the generation of VCM molecule, final state Fs was formed. Finally, the desorption of VCM was demonstrated, resolving the energy produced to be −35.21 kcal/mol. Fig. S11 (Supporting information) showed reaction pathways of catalytic process in pyrrolic N, graphitic N and oxidized N. The comparison revealed that the energy barrier of pyridinic N (20.54 kcal/mol) was the lowest compared with that of pyrrolic N (29.66 kcal/mol), graphitic N (33.28 kcal/mol) and oxidized N (35.63 kcal/mol). From the pathway analysis, it was concluded that the trapping and activation of pyridinic N site for HCl was the key to catalytic reaction.

In summary, non-metallic N-modified carbon were synthesized using a facile one-pot hydrothermal method to assemble mPDA and glucose. The material structure could be tuned by adding N source, mPDA, to form defect-rich carbon materials. The catalytic activity was significantly improved, which was derived from synergistic effect of hierarchically ordered porous structure and abundant N active sites. The C₂H₂ conversion could remained essentially at about 98% in 100 h tests, which was in some aspect superior to most non-metallic catalysts. The reaction pathways of pyridinic N, pyrrolic N, graphitic N and oxidized N were compared by a combination of experimental characterizations and DFT, demonstrating that pyridinic N was the predominant active site with the lowest energy barrier. Meanwhile, deactivation of catalysts was mainly due to formation of carbonaceous substances in reaction progress, which was evidenced by Brunner-Emmet-Teller (BET) and thermogravimetry analysis (TG) characterizations. The synthesis method proposed in this work provides a reference for development of morphologically and structurally oriented catalysts in acetylene hydrochlorination, which has a promising application.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Yusheng Lu: Writing – review & editing, Writing – original draft, Validation, Investigation, Formal analysis. Chaofeng Huang: Writing – review & editing, Resources. Zhigang Lei: Writing – review & editing, Supervision, Resources. Mingyuan Zhu: Writing – review & editing, Supervision, Resources.

This work is supported by the Tianchi Innovation Leading Talent Development Fund (No. CZ002710) in Xinjiang, the Taishan Scholars Program of Shandong Province (No. tsqn202103051), the Project of Science and Technology Development of Yantai City (No. 2023JCYJ073), Natural science foundation of Shandong province (No. ZR2023MB064), special funds for over provincial level leading talent of Yantai city, the Start-Up Foundation for High-level Professionals of Shihezi University (No. RCZK201932), Tianshan Talents Training Program of Xinjiang (Science and Technology Innovation Team, No. 2022TSYCTD0021).

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2024.110583.