MoO3 Nanobelt-Modified HMCM-49 Zeolite with Enhanced Dispersion of Mo Species and Catalytic Performance for Methane Dehydro-Aromatization

The methane dehydro-aromatization reaction (MDA) is a promising methane valorization process due to the conversion of methane to value-added aromatics (benzene, toluene and naphthalene). However, one of the major disadvantages of utilizing zeolite in MDA is that the catalyst is rapidly inactivated due to coke formation, which eventually causes the activity and aromatic selectivity to decrease. Consequently, the process is not conducive to large-scale industrial applications. The reasonable control of Mo site distribution on the zeolite surface is the key factor for partially inhibiting the coking of the catalyst and improving stability. Here, MoO3 nanobelts can be used for alternative Mo precursors to prepare MDA catalysts. Catalysts modified with MoO3 nanobelts present higher activity (13.4%) and benzene yield (9.2%) than those catalysts loaded with commercial MoO3.


Introduction
Natural gas is a promising alternative to oil as an industrial carbon source. Methane is a clean and environmentally friendly fossil fuel because it mainly emits CO 2 and water vapor after burning [1]. In addition, natural gas is usually transported by pipelines or in the form of liquefied natural gas. However, the construction of pipelines is extremely expensive and there is a risk of leakage. Consequently, there is a need to explore an efficient method for methane conversion into value-added chemicals from the perspective of methane utilization and environmental friendliness. MDA is one of the most attractive reactions, because the products are high-value petrochemicals (benzene, toluene and xylene) that can be applied to produce industrial raw materials [2].
Since Wang et al. reported the pioneering work that Mo/HZSM-5 catalyst can effectively enhance the catalytic performance, particularly the stability in MDA for the commercial viability of the process [3], there has been a great deal of research indicating that HZSM-5 zeolite is an excellent catalyst to provide a shape-selective condition for the conversion of methane to benzene. However, MDA reaction is a highly endothermic process: CH 4 → C 6 H 6 + H 2 , ∆ r H m = 532 KJ·mol −1 . Strong heat of adsorption means that higher benzene yields can be obtained at higher temperatures. The temperatures in the 300-1100 • C range are thermodynamically more inclined to produce graphite carbon (coke) [4,5]. Therefore, the efficient MDA catalyst should have the following conditions [6]: (i) the catalysts effectively activate the stable C-H bond in the CH 4 molecule; (ii) the prepared materials selectively produce aromatics and minimize carbon deposition; and (iii) the catalysts have good stability at high temperature.
It is generally believed that the dispersion of metal Mo species in zeolite plays an important role in catalyst stability [7]. Many strategies have revealed improvements, including the addition of accelerators [2,8,9], modulation of the pH value of the impregnation solution [10], and modification of the zeolite surface by desilication/dealumination or silicification [11] to enhance the dispersity of Mo species. Furthermore, many alternative zeolitic structures have been developed and investigated with the aim of reducing carbon deposition at the active site in methane non-oxidative aromatization. Various zeolite carriers such as ZSM-5 [12], ITQ-13 [13], MCM-22 [14], MCM-49 [15], IM-5 [16] and TNU-9 [17] zeolites have been tested. The results show that HZSM-5 and MCM series zeolite remain effective carriers for promoting MDA activity. HMCM-49 exhibits higher stability and benzene selectivity than HZSM-5; HMCM-49 has a unique pore structure design with two independent pore systems: a two-dimensional 10-membered ring sinusoidal pore system (4.1 × 5.9 Å) and a larger three-dimensional 12-membered ring super cage system (7.1 × 7.1 × 9.1 Å). The two systems are interconnected by a 10-membered ring channel (4.1 × 5.4 Å). The type of pore structure and the existence of super cages may be beneficial for the high carbon accumulation capacity in HMCM-49 zeolite [18] and maintaining the selectivity to aromatics [19].
Nanostructured MoO 3 materials have attracted extensive attention in photoelectric materials [20], heterogeneous catalysis [21,22], biochemical sensing [23], photothermal therapy [24], and other aspects because of its unique physical and chemical properties. The Ce-doped nano-MoO 3 modification of ZSM-5 catalyst shows excellent SCR performance [25], the ordered mesoporous α-MoO 3 nanocrystalline applied to thin film pseudocapacitors exhibits excellent capacitive charge storage performance [26]. Therefore, this study explores the use of MoO 3 nanobelts as the precursor for Mo-based HMCM-49 to induce better dispersion of the active metal MoO x in the pore channel of zeolite, thus effectively improving MDA catalytic performance.

Synthesis of MoO 3 Nanobelts
First, 1 g (NH 4 ) 6 Mo 7 O 24 ·4H 2 O and 0.08 g citric acid were added to a 100 mL beaker, and 10 mL of deionized water was introduced and stirred thoroughly to fully dissolve the solid. Then, the pH value was regulated to 1 with 3 mol/L HCl solution; the mixture was ultrasonicated for 30 min and then transferred to a Teflon-lined stainless autoclave under static crystallization at 180 • C for 2 d. The reaction product was cooled to room temperature, vacuum pumped and filtered, and washed with anhydrous ethanol and deionized water several times. The obtained products were dried for 30 min in the drying oven at 100 • C, and blue powder products were obtained.

Preparation of Na-MCM-49 Zeolite
MCM-49 zeolite was fabricated by hydrothermal synthesis by applying hexamethylenediamine (HMI) as a hard template and silica sol as a silicon source. Firstly, a mixture of HMI, water, aluminum nitrate, sodium hydroxide and silica sol in a certain proportion was stirred for 12 h. The mole ratio of gel composition is SiO 2 :0.04Al 2 O 3 : 0.12NaOH:0.35HMI: 25H 2 O, and the mixture was transferred to a Teflon-lined stainless autoclave under static crystallization at 170 • C for 3 d. The resulting gel was vacuum pumped and filtered, washed with deionized water to neutralize, and dried at 100 • C for 30 min. Then, 1 g MCM-49 precursor was dissolved in 50 mL 30 wt% H 2 O 2 solution in a round-bottomed flask at 90 • C for 12 h to remove the templating agent and then vacuum extraction and drying were performed. The above operations were repeated twice, and the product was calcined at 550 • C for 5 h to obtain Na-MCM-49 zeolite.

The Fabrication of HMCM-49 Zeolite
The Na-MCM-49 zeolite and ammonium nitrate (NH 4 NO 3 ) were weighed at a mass ratio of 1:16 and then placed in a 250 mL round-bottomed flask, 100 mL of deionized water

The Preparation of Mo-Based HMCM-49 Zeolite
HMCM-49 zeolite and commercial MoO 3 or MoO 3 nanobelt powder were mechanically mixed according to the mass ratio of 1:0.06, and then calcined in a muffle furnace at 550 • C for 5 h to obtain Mo-based MCM-49, which was respectively recorded as C-Mo-MCM-49 and N-Mo-MCM-49.

Catalyst Characterization
The XRD characterization of all samples was performed on a Smart Lab equipped with Cu Ka radiation (λ = 0.1541 nm; scan speed of 6 min −1 ; 2θ = 5-40 • ). The morphology and particle size of each sample were tested by FESEM XL-30 field emission scanning electron microscope. Fourier transform infrared spectroscopy (FT-IR) was performed with KBr as a support plate on a Shimadzu Lab Tatal spectrophotometer in the range of 400-4000 cm −1 . The specific surface and pore size distribution of the samples were recorded by N 2 adsorption-desorption isotherms at −196 • C on ASAP-2020 equipment. NH 3 -TPD was used to test the acidity of the catalyst on an Auto Chem 2920. Thermogravimetric analysis was performed on Shimadzu DTG-60, the sample was heated to 900 • C at a rate of 5 • C/min. Transmission electron microscopy (TEM) was performed by using a JEM 2100Plus microscopes, the catalyst was dispersed in ethanol using an ultrasonicator and dropped on carbon. H 2 -TPR was performed on an AutoChem 2720 instrument. The sample was treated at 350 • C for 0.5 h in a flow of Ar and cooled down. In the end, the sample was heated. X-ray photoelectron spectra (XPS) of the catalyst was tested with an ESCALAB250XI ray photoelectron spectrometer. Raman analysis of the catalyst was conducted to identify the carbonaceous deposit species of the catalyst.

Catalytic Tests
The catalytic activity of methane non-aromatization was measured in a fixed-bed reactor equipped with a quartz tube with an inner diameter of 1 cm, in which 0.5 g catalyst (40-60 mesh) was injected. The mass flow meter was used to control the introduction of raw gas (CH 4 (92.5%) and N 2 (7.5%)) mixture into the reactor at an airflow rate of 1500 mL g −1 h −1 using a mass flow controller, and the reaction temperature was controlled at 700 • C. Methane conversion and product yield during the reaction were analyzed using a Tianmei SCION gas chromatograph equipped with a 6 m × 3 mm HayeSep D 80/100 stainless steel column connected to a thermal conductivity detector (TCD) for the analysis of H 2 , N 2 , CO, CH 4 , CO 2 , C 2 H 4 and C 2 H 6 ; the other was a Shimazu CBP1-M50-025 non-polar capillary quartz column to detect benzene, toluene and naphthalene using a hydrogen flame detector. The system used for chromatographic analysis was Lab Solution, and the quantitative basis of methane conversion and hydrocarbon selectivity were calculated based on carbon mass balance using helium as carrier gas and N 2 as the internal standard [27].
The methane conversion and benzene, toluene, naphthalene, and coke selectivity were calculated according to following mathematical expressions: F, X and N carbon represent total gas flow rate, mole fraction and carbon number in a molecule, respectively.

Structure and Morphology of the Synthesized Catalysts
XRD reveals the composition and crystal phase structure of the MoO3 nanobelts (Figure 1a). As displayed in Figure 1a, the diffraction peaks of commercial MoO3 located at 2θ = 12.8°, 23.3°, 25.7°, 27.2°, 38.9° and 67.7° correspond to the (020), (110), (040), (021), (060) and (010) crystal planes, respectively, which is identical to the orthorhombic MoO3 (JCPDS card no. 01-0706). All of the peaks over the MoO3 nanobelts can be indexed to the hexagonal MoO3 ( JCPDS card no. 01-0569) without any impurities, demonstrating that the MoO3 nanobelts were successful fabricated [28]. Figure 1b Figure 3 presents the SEM images of the catalyst before and after modification. It can be seen that HMCM-49 is a flake crystal with a regular shape [31]; its length is less than 500 nm and its thickness is less than 50 nm. After the modification of the MoO3 nanobelts, the morphology of MCM-49 does not change significantly. However, the crystal   Figure 3 presents the SEM images of the catalyst before and after modification. It can be seen that HMCM-49 is a flake crystal with a regular shape [31]; its length is less than 500 nm and its thickness is less than 50 nm. After the modification of the MoO 3 nanobelts, the morphology of MCM-49 does not change significantly. However, the crystal particles become smaller on the surface of HMCM-49 after modification with different MoO 3 . This can be attributed to the complete grinding of the HMCM-49 crystal and MoO 3 nanobelts, or the fact that the large surface area of nano-MoO 3 bonds easily with the acid center in the zeolite, thus destroying the flake crystal [32]. The above results further indicate that MoO 3 nanobelts can easily sublimate into the pores of the material after mechanical mixing of MoO 3 nanobelts and HMCM-49 under high-temperature calcination [33]. Therefore, it is particles become smaller on the surface of HMCM-49 after modification with different MoO3. This can be attributed to the complete grinding of the HMCM-49 crystal and MoO3 nanobelts, or the fact that the large surface area of nano-MoO3 bonds easily with the acid center in the zeolite, thus destroying the flake crystal [32]. The above results further indicate that MoO3 nanobelts can easily sublimate into the pores of the material after mechanical mixing of MoO3 nanobelts and HMCM-49 under high-temperature calcination [33]. Therefore, it is speculated that the MoO3 nanobelts can achieve good distribution in the HMCM-49 zeolite without damaging the overall morphology of the crystal.   particles become smaller on the surface of HMCM-49 after modification with different MoO3. This can be attributed to the complete grinding of the HMCM-49 crystal and MoO3 nanobelts, or the fact that the large surface area of nano-MoO3 bonds easily with the acid center in the zeolite, thus destroying the flake crystal [32]. The above results further indicate that MoO3 nanobelts can easily sublimate into the pores of the material after mechanical mixing of MoO3 nanobelts and HMCM-49 under high-temperature calcination [33]. Therefore, it is speculated that the MoO3 nanobelts can achieve good distribution in the HMCM-49 zeolite without damaging the overall morphology of the crystal.    [34]. The peak at 443.62 cm −1 is caused by T−O bending vibration in TO 4 (T is Si or Al), and the above-infrared absorption peak indicates that the synthesized material is HMCM-49 zeolite [35]. Furthermore, it is revealed that when Mo species are loaded onto HMCM-49 zeolite, the position of the sample characteristic absorption peak does not change, and the absorption intensity decreases, indicating that the Mo species have been dispersed into the zeolite channel, which may also damage the material structure to a certain extent.
symmetric stretching vibration peaks. The absorption peaks at 675.02 and 549. caused by the double six-ring (D6R) vibration in the HMCM-49 zeolite, and characteristic absorption peaks of microporous zeolite [34]. The peak at 443 caused by T−O bending vibration in TO4 (T is Si or Al), and the above-infrared peak indicates that the synthesized material is HMCM-49 zeolite [35]. Further revealed that when Mo species are loaded onto HMCM-49 zeolite, the position ple characteristic absorption peak does not change, and the absorption intensity indicating that the Mo species have been dispersed into the zeolite channel, w also damage the material structure to a certain extent. The structures of HMCM-49 and HMCM-49 modified with different MoO termined by N2 adsorption-desorption isotherms, as shown in Figure 5. It can b that all of the samples show typical I-type curves with microporous material c tics in the region of low specific pressure (P/P0 < 0.01), and the adsorption capac ues to increase in the region with high specific pressure (P/P0 = 0.86~1.0), and a type hysteresis ring appears [18]. This may be caused by the adsorption of N2 tercrystalline pore and capillary condensation on the crystal surface, indicatin material contains a certain amount of mesopores. The surface area and pore str rameters of the material are shown in Table 1. The pore size distributions sho pore size of the zeolite HMCM-49 after MoO3 nanobelt modification increases. I the specific surface area of the material decreases significantly after Mo modifi the specific surface area of HMCM-49 modified by MoO3 nanobelts decrease compared to that of the parent zeolite, indicating that Mo species migrate into of the zeolite and bond with the acid center to promote the dissolution of part eton of the zeolite, thus increasing the pore size. Meanwhile, MoOx species ga surface of the catalyst and block the pore of the zeolite, reducing the specific su [36]. The above analysis results indicate that MoO3 nanobelts are easy to sub migrate into the pores of HMCM-49 zeolite during the calcination process, res significant decrease in catalyst surface area [37].  The structures of HMCM-49 and HMCM-49 modified with different MoO 3 were determined by N 2 adsorption-desorption isotherms, as shown in Figure 5. It can be observed that all of the samples show typical I-type curves with microporous material characteristics in the region of low specific pressure (P/P 0 < 0.01), and the adsorption capacity continues to increase in the region with high specific pressure (P/P 0 = 0.86~1.0), and a small H1-type hysteresis ring appears [18]. This may be caused by the adsorption of N 2 on the intercrystalline pore and capillary condensation on the crystal surface, indicating that the material contains a certain amount of mesopores. The surface area and pore structure parameters of the material are shown in Table 1. The pore size distributions show that the pore size of the zeolite HMCM-49 after MoO 3 nanobelt modification increases. In addition, the specific surface area of the material decreases significantly after Mo modification, and the specific surface area of HMCM-49 modified by MoO 3 nanobelts decreases to 80.6% compared to that of the parent zeolite, indicating that Mo species migrate into the pores of the zeolite and bond with the acid center to promote the dissolution of part of the skeleton of the zeolite, thus increasing the pore size. Meanwhile, MoO x species gather on the surface of the catalyst and block the pore of the zeolite, reducing the specific surface area [36]. The above analysis results indicate that MoO 3 nanobelts are easy to sublimate and migrate into the pores of HMCM-49 zeolite during the calcination process, resulting in a significant decrease in catalyst surface area [37].

The Surface Acid Studies
Acid sites play a vital role in the preparation of excellent MDA catalysts, not only affecting the adsorption of metal sites, but also optimizing the reaction path of aromatic hydrocarbon generation [38]. The surface acidity of the synthesized catalyst was characterized by NH 3 -TPD ( Figure 6 and Table 2). There is a new peak after fitting the curve for HMCM-49 and Mo-doped HMCM-49 catalyst. Three NH 3 desorption peaks are respectively recorded as peak H, peak M and peak L [39]. The center of peak L, located at about 187 • C, is assigned to the physical adsorption of NH 3 species on Lewis acid and/or NH 3 stays on non-exchangeable cations, The center of peak M at about 278 • C is attributed to ammonia desorption on the exchangeable protonic acidic sites, and the peak H centered at 417 • C is due to NH 3 bound to strong Brønsted acid centers [40]. As shown in Figure 6, peak H of the N-Mo-HMCM-49 catalyst shifts at lower temperatures, while peak M increases when HMCM-49 is loaded with different MoO 3 , and peak H of the N-Mo-HMCM-49 catalyst is clearly lower than that of the C-Mo-HMCM-49 catalyst. Combined with the results presented in Table 2, this indicates that Brønsted acid strength decreases significantly. In addition, the N-Mo-HMCM-49 catalyst contributes to the outstanding MDA to a certain extent due to the generation of Mo-O-Al by the interaction of small MoO 3 with Brønsted acid [41].
Molecules 2022, 27, x FOR PEER REVIEW

The Surface Acid Studies
Acid sites play a vital role in the preparation of excellent MDA catalysts, n affecting the adsorption of metal sites, but also optimizing the reaction path of a hydrocarbon generation [38]. The surface acidity of the synthesized catalyst was terized by NH3-TPD ( Figure 6 and Table 2). There is a new peak after fitting the cu HMCM-49 and Mo-doped HMCM-49 catalyst. Three NH3 desorption peaks are tively recorded as peak H, peak M and peak L [39]. The center of peak L, located a 187 °C, is assigned to the physical adsorption of NH3 species on Lewis acid and/ stays on non-exchangeable cations, The center of peak M at about 278 °C is attrib ammonia desorption on the exchangeable protonic acidic sites, and the peak H c at 417 °C is due to NH3 bound to strong Brønsted acid centers [40]. As shown in F peak H of the N-Mo-HMCM-49 catalyst shifts at lower temperatures, while pea creases when HMCM-49 is loaded with different MoO3, and peak H of the N-Mo-H 49 catalyst is clearly lower than that of the C-Mo-HMCM-49 catalyst. Combined w results presented in Table 2, this indicates that Brønsted acid strength decreases cantly. In addition, the N-Mo-HMCM-49 catalyst contributes to the outstanding M a certain extent due to the generation of Mo-O-Al by the interaction of small Mo Brønsted acid [41].     (Figure 7). The different reduction features of MoO x species in Mo-based HZSM-5 zeolite have been reported, and the reduction located at 370-580 • C can be ascribed to the reduction of MoO 3 to MoO 2 , whereas the reduction of MoO 2 to Mo occurs at higher temperatures-above 600 • C [42]. It can be observed that when moving from N-Mo-HMCM-49 to C-Mo-HMCM-49, the reducibility of initial MoO 3 to MoO 2 in N-Mo-HMCM-49 catalyst increases, which can be considered to be due to the change in the anchoring sites of MoO 3 species inside the HMCM-49 channels with respect to different kinds of MoO 3 loading. Lower reducibility over N-Mo-HMCM-49 can be observed, which is due to the strong interaction of MoO x species with HMCM-49 [2]. Therefore, the variation in the reducibility of MoO x species, as shown in the H 2 -TPR analysis, verifies that the MoO 3 species change in the HMCM-49 zeolite significantly influences the binding of MoO x species.    Figure 8. Molybdenum carbide (Mo 2 C) dispersed in zeolite channels is the active site of C-H bond activation, and is formed during the induction stage of the MDA reaction process [5,43]. It can be observed that the Mo 3d binding energies at 236.6 and 233.4 eV can respectively be assigned to the Mo 6+ (3d 3/2 ) and Mo 6+ (3d 5/2 ) molybdenum oxide species for fresh catalysts, whereas the binding energy at 228.9 eV can be attributed to the molybdenum carbide species (Mo 2 C) formed for the catalysts after the reaction under CH 4   C 1s region XPS spectra of the C-Mo-HMCM-49 and N-Mo-HMCM-49 catalysts after reaction under CH4 flow at 700 °C for 430 min confirmed that the C-Mo-HMCM-49 catalyst has greater carbon deposition. As shown in Figure 10, the C-C bond due to hard coke formed by polyaromatic or pre-graphitic species at strong Brønsted acid sites, while the C-O-C and O-C=O bonds are ascribed to the graphitic coke formed at the oxycarbide active sites (MoOxCy) [45]. These results demonstrate that the higher Mo distribution in the case of the N-Mo-HMCM-49 sample, which also inhibits the formation of hard coke, maybe conducive to a different catalyst deactivation mechanism.   C 1s region XPS spectra of the C-Mo-HMCM-49 and N-Mo-HMCM-49 catalysts after reaction under CH4 flow at 700 °C for 430 min confirmed that the C-Mo-HMCM-49 catalyst has greater carbon deposition. As shown in Figure 10, the C-C bond due to hard coke formed by polyaromatic or pre-graphitic species at strong Brønsted acid sites, while the C-O-C and O-C=O bonds are ascribed to the graphitic coke formed at the oxycarbide active sites (MoOxCy) [45]. These results demonstrate that the higher Mo distribution in the case of the N-Mo-HMCM-49 sample, which also inhibits the formation of hard coke, maybe conducive to a different catalyst deactivation mechanism.  C 1s region XPS spectra of the C-Mo-HMCM-49 and N-Mo-HMCM-49 catalysts after reaction under CH 4 flow at 700 • C for 430 min confirmed that the C-Mo-HMCM-49 catalyst has greater carbon deposition. As shown in Figure 10, the C-C bond due to hard coke formed by polyaromatic or pre-graphitic species at strong Brønsted acid sites, while the C-O-C and O-C=O bonds are ascribed to the graphitic coke formed at the oxycarbide active sites (MoO x C y ) [45]. These results demonstrate that the higher Mo distribution in the case of the N-Mo-HMCM-49 sample, which also inhibits the formation of hard coke, maybe conducive to a different catalyst deactivation mechanism.  C 1s region XPS spectra of the C-Mo-HMCM-49 and N-Mo-HMCM-49 catalysts after reaction under CH4 flow at 700 °C for 430 min confirmed that the C-Mo-HMCM-49 catalyst has greater carbon deposition. As shown in Figure 10, the C-C bond due to hard coke formed by polyaromatic or pre-graphitic species at strong Brønsted acid sites, while the C-O-C and O-C=O bonds are ascribed to the graphitic coke formed at the oxycarbide active sites (MoOxCy) [45]. These results demonstrate that the higher Mo distribution in the case of the N-Mo-HMCM-49 sample, which also inhibits the formation of hard coke, maybe conducive to a different catalyst deactivation mechanism.  The catalysts of C-Mo-HMCM-49 and N-Mo-HMCM-49 after reaction under CH 4 flow at 700 • C for 430 min have almost identical Raman spectra, as shown in Figure 11. They exhibit five-peak-based deconvolution of the intense signals associated with carbon species. It is clear that the Raman shift located at 1200 cm −1 can be ascribed to aliphatic C-H bounds, the Raman shift around 1312 cm −1 represents the D band, which is attributed to the aromatic structures or amorphous coke, the D 3 band around 1405 cm −1 is due to structural defects of aromatic domains with poor organization, and the G band located at 1590 cm −1 is assigned to graphite-like coke while D 2 belongs to the D echo band. It can be clearly seen that the ratio between D and G band intensities is lower for N-Mo-HMCM-49 (I D /l G = 0.62) than that of C-Mo-HMCM-49 (I D /l G = 1.01), demonstrating that the degree of the disordered nature of the surface carbon species is more remarkable for C-Mo-HMCM-49 [46]. Furthermore, the coke produced from MoC x is well known to be amorphous, which promotes the deactivation of the catalyst [47,48]. Meanwhile, HMCM-49 modified with MoO 3 nanobelts may prevent excessive coke formation on MoC x and would improve catalytic stability. The catalysts of C-Mo-HMCM-49 and N-Mo-HMCM-49 after reaction under CH4 flow at 700 °C for 430 min have almost identical Raman spectra, as shown in Figure 11. They exhibit five-peak-based deconvolution of the intense signals associated with carbon species. It is clear that the Raman shift located at 1200 cm −1 can be ascribed to aliphatic C-H bounds, the Raman shift around 1312 cm −1 represents the D band, which is attributed to the aromatic structures or amorphous coke, the D3 band around 1405 cm −1 is due to structural defects of aromatic domains with poor organization, and the G band located at 1590 cm −1 is assigned to graphite-like coke while D2 belongs to the D echo band. It can be clearly seen that the ratio between D and G band intensities is lower for N-Mo-HMCM-49 (ID/lG = 0.62) than that of C-Mo-HMCM-49 (ID/lG = 1.01), demonstrating that the degree of the disordered nature of the surface carbon species is more remarkable for C-Mo-HMCM-49 [46]. Furthermore, the coke produced from MoCx is well known to be amorphous, which promotes the deactivation of the catalyst [47,48]. Meanwhile, HMCM-49 modified with MoO3 nanobelts may prevent excessive coke formation on MoCx and would improve catalytic stability.  Figure 12 and Table 3 display the methane conversion, aromatics yield, and selectivity of different products (benzene, toluene, naphthalene and coke) over C-Mo-HMCM-49 and N-Mo-HMCM-49 in MDA [49]. It can be clearly observed from Figure 12a that the methane conversion rate demonstrates an increasing trend during the initial reaction stage of less than 130 min, implying the existence of an induction period for the MDA reaction during which active site molybdenum carbide species (Mo2C or MoOxCy) are generated [50]. The highest methane conversion rates of C-Mo-HMCM-49 and N-Mo-HMCM-49 are 11.9% and 13.4%, respectively, and the corresponding aromatic hydrocarbon yields are 7.5% and 9.2% (Figure 12b). Furthermore, the value TOF of 15.1 h −1 over N-Mo-HMCM-49 is higher than that of C-Mo-HMCM-49 (14.2 h −1 ). Compared to the 6Mo(N)-MCM-22 catalyst previously reported, N-Mo-HMCM-49 exhibits a better TOF value. It is worth noting that the methane conversion rate for all catalysts continues to decline with increasing time on stream (TOS), meaning that catalyst inactivation occurs continuously during the reaction, which may be attributed to the sintering and separation of carbon and metal carbonization species accumulated on the zeolite [51]. Additionally, it can be clearly seen from Table 3 that 11.5% methane conversion is obtained on N-Mo-HMCM-49, which exceeds that (9.8%) on C-Mo-HMCM-49 after 430 min, and the aromatics yields of C-Mo-HMCM-49 and N-Mo-HMCM-49 are respectively 6.2% and 8.0% at this point. The MCM-49 zeolite has a set of 12-ring super cages and 10-ring channels connected through 10-ring  Figure 12 and Table 3 display the methane conversion, aromatics yield, and selectivity of different products (benzene, toluene, naphthalene and coke) over C-Mo-HMCM-49 and N-Mo-HMCM-49 in MDA [49]. It can be clearly observed from Figure 12a that the methane conversion rate demonstrates an increasing trend during the initial reaction stage of less than 130 min, implying the existence of an induction period for the MDA reaction during which active site molybdenum carbide species (Mo 2 C or MoO x C y ) are generated [50]. The highest methane conversion rates of C-Mo-HMCM-49 and N-Mo-HMCM-49 are 11.9% and 13.4%, respectively, and the corresponding aromatic hydrocarbon yields are 7.5% and 9.2% (Figure 12b). Furthermore, the value TOF of 15.1 h −1 over N-Mo-HMCM-49 is higher than that of C-Mo-HMCM-49 (14.2 h −1 ). Compared to the 6Mo(N)-MCM-22 catalyst previously reported, N-Mo-HMCM-49 exhibits a better TOF value. It is worth noting that the methane conversion rate for all catalysts continues to decline with increasing time on stream (TOS), meaning that catalyst inactivation occurs continuously during the reaction, which may be attributed to the sintering and separation of carbon and metal carbonization species accumulated on the zeolite [51]. Additionally, it can be clearly seen from Table 3 that 11.5% methane conversion is obtained on N-Mo-HMCM-49, which exceeds that (9.8%) on C-Mo-HMCM-49 after 430 min, and the aromatics yields of C-Mo-HMCM-49 and N-Mo-HMCM-49 are respectively 6.2% and 8.0% at this point. The MCM-49 zeolite has a set of 12-ring super cages and 10-ring channels connected through 10-ring windows. The unique pore systems are beneficial for benzene and aromatics formation. Mo carbides located on the surface of zeolite mainly generate heavy hydrocarbon and carbon deposition, leading to a decrease in methane conversion and aromatic hydrocarbon yield [51]. Furthermore, as illustrated in Figure 13, the weight loss of MoO 3 nanobelts is 7.5% below 900 • C, which is higher than the weight loss (4.4%) of commercial MoO 3 . Thermogravimetric analysis shows that MoO 3 nanobelts sublimate more easily than commercial MoO 3 . The improvement in catalytic activity may be related to the good dispersibility of active Mo sites. It is likely that the structure of Mo precursors plays a significant role in catalytic performance. Mo precursors supported on carrier material have multiple structures, which results in complex and widened signals for most spectroscopic techniques [1]. Combined with the results of acid tests, MoO 3 nanobelts can be effectively transferred to MCM-49 channels by regulating the acidity distribution and interacting with strong acid sites to generate more effective active Mo sites and achieve better catalytic performance [30]. lectivity (60.3%) than C-Mo-HMCM-49 (54.6%) after reaction for 130 min, and the corresponding coke selectivities are 37.0% and 31.4%, respectively. The improved selectivity of benzene modified by MoO3 nanobelts may be partly due to the strong inhibition of coking by external Brønsted sites. However, the deposition of carbonaceous species eventually blocks the zeolite channel, making it difficult for products to diffuse out of the zeolite channel [52]. Combined with the results of Raman spectra, HMCM-49 modified with MoO3 nanobelts would inhibit excessive coke deposition and thus improve catalytic stability.    Figure 12d indicates the selectivity of various products for both Mo-based catalysts. In combination with Table 3, it can be seen that N-Mo-HMCM-49 has higher benzene selectivity (60.3%) than C-Mo-HMCM-49 (54.6%) after reaction for 130 min, and the corresponding coke selectivities are 37.0% and 31.4%, respectively. The improved selectivity of benzene modified by MoO 3 nanobelts may be partly due to the strong inhibition of coking by external Brønsted sites. However, the deposition of carbonaceous species eventually blocks the zeolite channel, making it difficult for products to diffuse out of the zeolite channel [52]. Combined with the results of Raman spectra, HMCM-49 modified with MoO 3 nanobelts would inhibit excessive coke deposition and thus improve catalytic stability.  As shown in Figure 14, the weight loss below 220 °C is due to the evaporation of water. The weight loss between 220 and 400 °C is caused by the aromatics of physical absorption, and it is noteworthy that the carbon deposits on zeolite are burned and caused weight loss in the range of 400 to 800 °C [53]. Benzene is mainly formed in the channels of the zeolite, whereas naphthalene is produced on the external surface and/or the pore mouth of the zeolite [1,54]. Carbon deposition in the zeolite channel will not only block the zeolite channel, but also poison the catalytically active center, eventually leading to catalyst inactivation [52]. The weight-loss rates of C-Mo-MCM-49 and N-Mo-MCM-49 after 430 min are 5.76% and 3.47%, respectively, indicating that N-Mo-MCM-49 has better anti-carbon deposition ability than C-Mo-MCM-49 [55]. Consequently, N-Mo-MCM-49 shows better catalytic stabilization in MDA reactions.

Conclusions
In summary, MCM-49 zeolite modified with MoO3 nanobelts exhibits better MDA performance than commercial MoO3-modified zeolite. XRD, FT-IR, N2 As shown in Figure 14, the weight loss below 220 • C is due to the evaporation of water. The weight loss between 220 and 400 • C is caused by the aromatics of physical absorption, and it is noteworthy that the carbon deposits on zeolite are burned and caused weight loss in the range of 400 to 800 • C [53]. Benzene is mainly formed in the channels of the zeolite, whereas naphthalene is produced on the external surface and/or the pore mouth of the zeolite [1,54]. Carbon deposition in the zeolite channel will not only block the zeolite channel, but also poison the catalytically active center, eventually leading to catalyst inactivation [52]. The weight-loss rates of C-Mo-MCM-49 and N-Mo-MCM-49 after 430 min are 5.76% and 3.47%, respectively, indicating that N-Mo-MCM-49 has better anti-carbon deposition ability than C-Mo-MCM-49 [55]. Consequently, N-Mo-MCM-49 shows better catalytic stabilization in MDA reactions. thane conversion, (b) aromatics yield, (c) the comparison of catalyst performance over both catalysts and (d) the selectivity of various products. Reaction conditions: T (temperature) = 700 °C, P (pressure) = 1 atm, GHSV (gas firing hourly space velocity) = 1500 h -1 . As shown in Figure 14, the weight loss below 220 °C is due to the evaporation of water. The weight loss between 220 and 400 °C is caused by the aromatics of physical absorption, and it is noteworthy that the carbon deposits on zeolite are burned and caused weight loss in the range of 400 to 800 °C [53]. Benzene is mainly formed in the channels of the zeolite, whereas naphthalene is produced on the external surface and/or the pore mouth of the zeolite [1,54]. Carbon deposition in the zeolite channel will not only block the zeolite channel, but also poison the catalytically active center, eventually leading to catalyst inactivation [52]. The weight-loss rates of C-Mo-MCM-49 and N-Mo-MCM-49 after 430 min are 5.76% and 3.47%, respectively, indicating that N-Mo-MCM-49 has better anti-carbon deposition ability than C-Mo-MCM-49 [55]. Consequently, N-Mo-MCM-49 shows better catalytic stabilization in MDA reactions.

Conclusions
In summary, MCM-49 zeolite modified with MoO 3 nanobelts exhibits better MDA performance than commercial MoO 3 -modified zeolite. XRD, FT-IR, N 2 adsorption/desorption NH 3 -TPD, XPS, TEM, H 2 -TPR and Raman spectra indicate that MoO 3 nanobelts are easy to sublimate and distribute into the channel of HMCM-49 zeolite and combine with the strong acid center located on the zeolites' inner surface to form the active Mo site, thus promoting methane activation, inhibiting the coke deposition formed by external Brønsted sites, and increasing the yield of aromatics. The methane conversion rate of MoO 3 nanobelt-modified HMCM-49 can reach 13.4%, and the corresponding aromatic hydrocarbon yield can reach 9.2%. This work lays the groundwork for the usage of nano-metal precursors to develop new catalysts and further enhance the MDA performance.