Highly Efficient Production of Benzene-Free Aromatics from Methanol over Low-Si/Al-Ratio Alkali-Modified Fe/Zn/HZSM-5

Abstract

Herein, the methanol conversion to aromatic hydrocarbons was studied over a new family of mesoporous low-silica HZSM-5 (Si/Al = 11) catalysts in a fixed-bed tubular reactor under ambient pressure at 375 °C, feeding with weight hourly space velocity of 2 h^–1. The catalysts were prepared in the absence and presence of Zn and Fe in both alkaline and neutral aqueous solutions, characterized by using X-ray diffraction, X-ray fluorescence, temperature programmed desorption of ammonia, N₂ adsorption/desorption, thermogravimetric analysis, Fourier-transform infrared, transmission electron microscopy (TEM), field emission scanning electron microscopy and FE-SEM/energy dispersive X-ray spectroscopy techniques. The [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst exhibited novel selectivity for aromatics (>86 wt %), specifically for m and p-xylenes (44.7 wt %) alongside 0.1 wt % for benzene.

1. Introduction

Nowadays, aromatic hydrocarbons are significant chemical intermediates in fine chemical industries, exhibiting sharp commercial demands.^1,2 Reforming of petroleum fraction aromatic hydrocarbons is the central source of aromatics, whereas working on new nonpetroleum substitute is industrially crucial.³⁻⁵ During recent decades, the research on the role of solid acid catalysts in methanol conversion to aromatics (MTA) process has fascinated both scientists and industrial organization, as methanol is supplied from different resources as syngas, natural gas, coal, biomass, and any other carbon-based gasifiable feedstock.^2,6−8

As one of the most practical choices, ZSM-5 zeolites enjoy having acidic sites, shape-selectivity, high surface area, and adaptability to an extensive assortment of products, namely from olefin to aromatics.^9,10 From the morphological point of view, two kinds of pores, shaped by 10-membered oxygen rings could be found in the ZSM-5 structure.¹¹ The presence of these types of pores together with the “zigzag pores” interconnecting them is crucial for the formation of desirable compounds during the MTA process.¹² Meanwhile, one of the significant limitations for the application of this catalyst is the diffusion of the compounds within the pores. Thus, the modifications of ZSM-5 catalyst could give rise to subsequent enhancement of the pore diffusion properties as well as upgrading of the catalyst impact on the target process.¹² A well-known method to achieve this goal is the desilication of ZSM-5 by employing NaOH, which results in formation of the mesoporous catalysts.¹³⁻¹⁸ Recently, the alkali treatment was found as a fascinating approach for developing of mesoporous morphology from the corresponding microporous ZSM-5 zeolites.^12,17 Mochizuki et al. applied the alkali treatment method by using different concentrations of NaOH solution in desilication of ZSM-5.¹⁹ They found several crystal sizes with different external surface areas, wherein the deactivation of the catalyst was attributed to the presence of Lewis-type acidic sites generated owing to employing high concentration of NaOH. In another experiment, Bjørgen et al.²⁰ reformed ZSM-5 crystals by employing alkali-treatment method with NaOH. They reported that the modification made on ZSM-5 resulted in lifting of the diffusion restrictions, wherein the selectivity of the ZSM-5 for the gasoline cut was promoted by a factor of 1.7.

Meanwhile, the existence of metal atom sites along with the acidic sites in the structure of ZSM-5 are two efficient morphological factors influencing on the catalyst activity. For instance, significant enhancement in methanol conversion to gasoline has been reported over ZSM-5 zeolite, incorporated with some metal species.²¹ Freeman et al.²² investigated the effect of various loadings of Ga₂O₃ on ZSM-5 in catalysis of the methanol-to-gasoline process. Their observations indicated on the enhancement of the conversion of methanol over the promoted ZSM-5 catalyst, whereby the catalyst stability was also significantly guaranteed against deactivation. In another report, Zaidi and Pant²³ reported the increase in the yield of aromatic hydrocarbons, employing ZSM-5 co-impregnation with ZnO and CuO.

The influence of various Si/Al ratios in ZSM-5 catalysts has also been considered in several reports. For instance, Mn-ZSM-5 catalyst with Si/Al = 200 represented the highest reported propylene selectivity.²⁴ As another example, the Zn-ZSM-5 catalyst with Si/Al = 27 displayed improved catalytic stability in 1-hexene aromatization,²⁵ while in another report Fe/Mo/impregnated ZSM-5 exhibited higher benzene yield, compared with Mo catalyst in methane dehydroaromatization.²⁶ In the latter report, Mo/ZSM-5 was promoted with Fe and Zn, displayed better catalyst stability in terms of the coke formation. Fila et al. reported better resistance of Co/Mo/ZSM-5 (Si/Al = 30) catalyst toward deactivation by coke formation.²⁷ Higher total olefins (80%) selectivity and propylene selectivity (51%) were also found in a methanol-to-olefin process catalyzed in the presence of Fe-ZSM-5 catalyst with Si/Al = 200.²⁴ By employing Si/Al = 13, Zhao et al.²⁸ reported remarkable improvement in stability of Pt-modified nanoscale HZSM-5 catalysts. Moreover, Li et al.²⁹ showed that the catalyst lifetime improvement in methanol conversion (MTO) process, up to 14 days after postmodification of ZSM-5 with Ga. On the other hand, some reports unveiled that the yield of hydrocarbon production could be improved in methanol conversion over ZSM-5 zeolite, upgraded with some metal species such as Zn,^21,30 Ga,²⁹ and Fe.^24,31 Jiao et al. showed that the Fe-modified HZSM-5 catalyst affected on the MTO process.³² They reported loading of HZSM-5 with Fe from 0.18 to 1.75% resulted in increase of the weak acid sites. They found a maximum of total acid site on HZSM-5 at the Fe loading up to 0.35%.³²

Some reports unveiled that the yield of aromatic hydrocarbons production in MTA was improved in the presence of the upgraded HZSM-5 zeolite with some metal species.^4,33,34 As an effective element in promoting the methanol-to-hydrocarbons process, zinc is the most attractive element.³⁰ Fattahi et al. represented 77 wt % production of aromatics using a Zn-impregnated (0.75 wt %) zeolite having a SiO₂/Al₂O₃ ratio of 54. Then, the reaction was accomplished at 370 °C and ambient pressure by employing weight hourly space velocity (WHSV) equal to 4 h^–1.¹⁸ The selectivity for aromatic hydrocarbons was also improved from 36.2 to 45.8% at 390 °C and WHSV of 3.2 h^–1, applying of Zn on the ZSM-5 crystals having Si/Al = 40 ratio.³⁰ Another report¹⁷ dealt with the improvement of BTX yield (up to 48%) over a nanosize [Zn,Al]-HZSM-5 zeolite, wherein the catalytic stability decreased to 32% after 160 h on stream of methanol conversion at 437 °C and WHSV of 0.8 h^–1. Besides, Ghavipour et al.¹⁰ studied temperature dependence (375, 425, and 475 °C) of methanol dehydration over a HZSM-5 catalyst with SiO₂/Al₂O₃ = 31.5, modified under alkali condition under WHSV of 5 h^–1. Then, Yang et al.⁴ applied alkali-modified 0.8Zn-HZSM-5 (SiO₂/Al₂O₃ = 59) zeolite in the MTA reaction to boost the aromatic yield from 41.4 to 55.3%. The catalyst also demonstrated more catalyst stability at T = 400 °C and WHSV = 2.5 h^–1. Besides, Ni et al.²¹ observed the mesoporous morphology changing of ZSM-5 catalyst under NaOH treatment accompanied by Zn impregnation.

Taking into account the above-mentioned results, the idea of promoting low-Si/Al zeolites with appropriate metal ions was found to be of commercial importance to prepare cheaper catalyst together with more efficient product distribution in MTA process.^18,35−39 For instance, researchers have synthesized a series of desilicated, Ga-impregnated, and low-Si/Al HZSM-5 catalysts, tested in aromatization of methanol at high temperatures, causing high aromatic yields alongside lengthy lifetime.⁴⁰

Moreover, Behbahani and Mehr⁴¹ observed the impregnation of ZSM-5 with strontium enhanced 2% in production of aromatics. In another report, the alternation of the reactivity of Brønsted acid sites (BASs) on aluminosilicate lattice of HZSM-5 as a result of variation in Si/Al ratio was explored.⁴² They showed the development of the lattice and BASs in low Si/Al ratio (Si/Al = 11) zeolites as a result of substitution of the Al groups. The aforesaid substitution was due to increase of polarization in the nearest-neighbor T-sites, whereby both the proton-transfer and the protonated-product formation processes were improved.⁴²

Very recently, Jiang et al.⁴³ demonstrated high selectivity of olefins in the MTO process in terms of C₂–C₄ over the use of nanosized ZSM-5 zeolites catalysts ([Fe,Al]NZ5), owing to the moderate acidity alongside the observed improvement in diffusion performance. They also indicated that impregnation of the catalyst with Fe could give rise to increase in the number of the BASs, leading to inhibit the olefin cycle and promote the aromatic cycle.

In this research, we undertook the preparation of alkali-modified low-Si [Fe, Zn]-HZSM-5 zeolite to develop a new catalyst to boost the best results reported for the catalytic performance of the modified HZSM-5 in the MTA process. The main aim of this work is to accomplish a systematic and comparative study of simultaneous alkali treatment as well as the best known postmodification of a low-Si/Al HZSM-5 catalyst by Fe and Zn in order to monitor the production of the aromatics in the MTA process.

In the course of the present research, the Taguchi method⁴⁴ was applied to obtain the best results by employing the minimum number of the required experiments. The Taguchi method has been a persuasive tool to facilitate research and development in order to attain prompt high quality products and productivity at reasonable costs.⁴⁵ The method covers a logical application of design and analysis of experiments in order for designing and improving the product quality. In the present study, based on the Taguchi method, the activity HZSM-5 catalyst in the MTA process was optimized using a program consisted of 2 levels and 5 factors.

2. Results and Discussion

2.1. Catalyst Characterization

Figure S1 shows the Fourier-transform infrared (FT-IR) spectra of HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5 and [0.4Fe,0.6Zn]-HZSM-5 catalysts in 400–4000 cm^–1. The spectra represented typical bands at ∼450 cm^–1 for the bending vibrations of the primary tetrahedral building units, T–O, wherein the central atom (T) was usually either a silicon or aluminum atom. The other band at ∼550 cm^–1 was assigned to double five-member rings in SiO₄ and AlO₄ units in a HZSM-5 structure. Besides, typical symmetric and asymmetric modes of vibrations of SiOSi for these bands were detected at 740–860 and 1040–1210 cm^–1. Another recorded band at 1620 cm^–1 together with the other broad band at 3450 cm^–1 was assigned to physically absorbed water and the interconnecting OH groups, respectively, akin to the other reports.^20,46,47

Furthermore, an assortment of OH groups was found in the IR spectra at 3000–3800 cm^–1, confirmed on the Si–OH functional groups within the sample. As shown in Figure S1, all the catalysts represented the band at ∼3600 cm^–1, indicating on Si–OH–Al and Al–OH groups, respectively.⁴³ The bands at 3740 cm^–1 were also attributed to the external Si–OH.^48,49

Moreover, Figure 1 represents the X-ray diffraction (XRD) patterns for the catalysts.

Figure 1.

XRD patterns of the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 zeolites.

Practically, five diffraction bands (2θ = 8.1°, 8.9°, 23.1°, 23.3°, and 23.8°) were found in the XRD patterns shown in Figure 1, assigned to the orthorhombic structure of HZSM-5.¹¹ As represented in Figure 1, modification of zeolite structure, made no alternation in the peak position of the XRD graphs nor change in the initial structure of the HZSM-5, as expected for the low-metal-loaded (below 1 wt %) HZSM-5 samples. All seven catalysts exhibited related diffraction patterns, specifying that the original crystal structure of HZSM-5 zeolite was well-preserved during the alkaline treatments, corresponding to the previous report.⁴ Although all of the catalysts exhibited high crystallinity (Table 1), but subsequent impregnation of the parent catalyst led to relatively low crystallinity, as a result of the effect of the desilication process on the decrease of the crystallinity of HZSM-5 zeolite.⁵⁰ Incidentally, high dispersion of Fe and Zn species on HZSM-5 zeolite was concluded in the absence of any evidence for the formation of a new phase in the XRD patterns in Figure 1.

Table 1. Structural Properties of the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 Zeolites.

	characteristic
sample name	SBET (m2 g–1)a	SMicro (m2 g–1)b	SMeso (m2 g–1)c	VTotal (cm3g–1)d	VMicro (cm3 g–1)e	VMeso (cm3 g–1)f	particle sizeg (nm)	crystallinityh (%)
HZSM-5	229.89	209.16	20.73	0.1235	0.0965	0.0270	26.09	100
alk-HZSM-5	349.57	291.02	58.55	0.1679	0.1056	0.0523	27.49	79
0.8Zn-HZSM-5	350.85	307.95	42.90	0.1552	0.1522	0.0030	27.11	82
0.8Zn-alk-HZSM-5	323.42	245.55	77.87	0.1823	0.0688	0.1135	18.55	76
[0.2Fe,0.3Zn]-alk-HZSM-5	279.84	220.32	59.52	0.1636	0.0619	0.1017	21.44	68
[0.2Fe,0.3Zn]-HZSM-5	336.82	312.90	23.92	0.1413	0.1214	0.0199	29.30	83
[0.4Fe,0.6Zn]-HZSM-5	335.31	311.47	23.84	0.1371	0.1295	0.0076	29.91	75

^aThe BET method was applied in terms of adsorption data for determination of total surface areas in P/P₀, extending from 0.05 to 0.25.

^bMicropore surface area assessed by the t-plot method.

^cMesopore surface area was measured by using S_BET–S_Micro.

^dTotal pore volumes were judged by the adsorbed amount at P/P₀ = 0.99.

^eMicropore volume evaluated by the t-plot method.

^fMesopore volume calculated using V_Total–V_Micro.

^gParticle size estimated by Scherer’s equation.

^hThe intensity fraction for a portion of the XRD pattern for the sample was considered with respect to the intensity of the corresponding portion in the reference HZSM-5 pattern (ASTM D5758-01), which was applied as XRD relative crystallinity.

Furthermore, the HZSM-5 zeolite represented relatively high crystallinity among the synthesized samples (Table 1).

Table 1 also demonstrates that the Brunauer–Emmett–Teller (BET) surface area (S_BET) as well as the total pore volume (V_total) of the HZSM-5 zeolite samples, while in general the alkali treatment alongside impregnation with Zn and Fe on the HZSM-5 gave rise to remarkable increase in the corresponding effective surface area (S_Meso). Basically, higher surface area and pore volumes for the alkali-treated samples could be observed as a result of the desilication treatment, generating more mesopores.

On the other hand, Table 1 represents a significant increase in V_Meso which was observed in conjunction with decrease in V_Micro. Evidently, the observed growth in the number of mesopores was occurred on the expense of the decrease in the corresponding micropore population. In conclusion, both of alkali treatments and metal impregnation on the parent ZSM-5 generally have promoted the key surface and catalytic factors, namely S_BET, S_Meso, and V_Total. Taking into account the abovementioned conclusions, a prominent catalytic effect for [0.2Fe,0.3Zn]-alk-HZSM-5 with regard to the parent HZSM-5 zeolite was expected. The textural parameters for [0.2Fe,0.3Zn]-alk-HZSM-5, however, were denoted on the reduced values for S_BET as well as V_total with respect to 0.8Zn-alk-HZSM-5 zeolite. Previously, some researchers endorsed the phenomenon of blockage of pores by metallic species to serve as the reducing factor in S_BET and V_total.^9,51,52

The field emission scanning electron microscopy (FE-SEM) micrograph of the catalyst samples in Figure 2 revealed the morphology of (a) HZSM-5, (b) alk-HZSM-5, (c) 0.8Zn-HZSM-5, (d) 0.8Zn-alk-HZSM-5, (e) [0.2Fe,0.3Zn]-alk-HZSM-5, (f) [0.2Fe,0.3Zn]-HZSM-5, and (g) [0.4Fe,0.6Zn]-HZSM-5 zeolites.

Figure 2.

FE-SEM images of (a) HZSM-5, (b) alk-HZSM-5, (c) 0.8Zn-HZSM-5, (d) 0.8Zn-alk-HZSM-5, (e) [0.2Fe,0.3Zn]-alk-HZSM-5, (f) [0.2Fe,0.3Zn]-HZSM-5, and (g) [0.4Fe,0.6Zn]-HZSM-5 zeolites.

Figure 2 displays a spongelike for these samples very similar to the recently reported morphology for HZSM-5 and 0.8Zn-HZSM-5.^30,31 In order to display the distribution of the elements over the catalyst, energy dispersive X-ray spectroscopy (EDXS) dot-mapping analysis was shown in Figure S2, confirming on their perfect homogeneous distribution on the samples. Besides, Figure 3 exhibits transmission electron microscopy (TEM) images of the catalysts.

Figure 3.

TEM images of (a) HZSM-5, (b) alk-HZSM-5, (c) 0.8Zn-alk-HZSM-5, and (d) [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites. The marked yellow circles signify the observed changes.

As shown in Figure 3a, the HZSM-5 catalyst exhibits irregular and spherical-like morphology. The TEM image in Figure 3b indicates that the HZSM-5 zeolite morphology undergoes intracrystalline mesoporosity, owing to desilication during the alkali treatment. Meanwhile, because of the alkali treatment, the etched holes and pores are observable in Figure 3b, suggesting a partial deterioration of external surface of the parent HZSM-5 zeolite. Because the XRD method demonstrated no indication on the formation of new phases pertaining to Fe and Zn compounds on the HZSM-5 due to the low loading of the metals, TEM micrographs were used to monitor any new morphological changes on the surface of 0.8Zn-alk-HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites (Figure 3c,d), resulted from the corresponding modification on the precursor zeolite. As Fe and Zn oxides have a higher average molecular weight than the other components on the parent HZSM-5 zeolite, they should display darker contrast in the bright-field images compared to the support. As shown in Figure 3d several new phases, ranging from 30 to 200 nm, can be seen at the edges of the HZSM-5 crystals, attributable to new phases on the HZSM-5 support.

To investigate surface structure of the samples in more detail, N₂ adsorption/desorption isotherms were inspected (Figure 4).

Figure 4.

N₂ adsorption/desorption isotherms for the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 zeolites.

The isotherms of the catalysts in Figure 4 represented an adsorption isotherm type IV. Furthermore, at relatively high pressures (P/P₀ > 0.5) in Figure 4, restricted gas uptake during capillary condensation in mesopores provided type IV isotherms together with the observed H4 hysteresis loops specified narrow ink-bottle-type pores with irregular shapes.⁵³ In these experiments, at relative low pressures, that is P/P₀ < 0.2 for HZSM-5, <0.35 for 0.8Zn-alk-HZSM-5, <0.34 for alk-HZSM-5, <0.36 for 0.8Zn-HZSM-5, <0.24 for [0.2Fe,0.3Zn]-alk-HZSM-5, <0.35 for [0.2Fe,0.3Zn]-HZSM-5, and <0.33 for [0.4Fe,0.6Zn]-HZSM-5 zeolites, the isotherms indicated on the formation of Langmuir-type monolayer adsorption, occurred as a result of the limiting gas uptake by the available micropores. The variation of the utmost monolayer P/P₀ values for these samples at low pressures, however, represents the following order:

If one takes the extent of monolayer formation in Figure 4 (at the abovementioned P/P₀ values) as an indication of surface area, the aforesaid order suggested that alkali treatment of the parent HZSM-5 (3.1 mmol g^–1) resulted in a significant surface enlargement (4.4 mmol g^–1 for 0.8Zn-alk-HZSM-5), wherein subsequent metal impregnations served as a fair decreasing factor for surface area (3.7 mmol g^–1 for [0.2Fe,0.3Zn]-alk-HZSM-5). These evidence led us to propose that the impregnation process had a somehow negative effect on the monolayer adsorption of [0.2Fe,0.3Zn]-alk-HZSM-5. However, the remarkable change in the N₂ adsorption/desorption isotherms observed for [0.2Fe,0.3Zn]-alk-HZSM-5 zeolite could be attributed to the developed mesoporosity on the zeolite during the alkali treatment process.

Barrett–Joyner–Halenda (BJH) pore size distribution (PSD) curves are demonstrated in Figure 5.

Figure 5.

BJH PSD curves of the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 zeolites; the presentation of the PSD at: (a) <6, (b) 30–40 nm, and (c) the complete range.

The meso-PSD was obtained from the adsorption isotherm by means of the BJH method. The BJH PSDs of HZSM-5 zeolite was concentrated at only 2.5–6 nm, while corresponding values for alk-HZSM-5 zeolite encompass a large population of pores at 1.5–10 nm range, a low population of medium-size pores at ∼4 nm, and a new very low population of large pore centered at a 35 nm (Figure 6a,b). Impregnation of both HZSM-5 as well as alkali-HZSM-5 with Zn afforded the pore-size distributions very similar to alk-HZSM-5 zeolite for HZSM-5, while under the same condition, corresponding small pores at 2.5–6 nm in alk-ZSM-5 were completely substituted with the pores with ∼4 and ∼35 nm. Moreover, the co-impregnation of 0.8Zn-HZSM-5 with Fe did not significantly altered the pore-size distribution, even at high concentration of Fe, comparing the PSD curves for [0.2Fe,0.3Zn]-HZSM-5 and [0.4Fe,0.6Zn]-HZSM-5 zeolites in Figures 5a and 6b. The co-impregnation of 0.8Zn-alk-HZSM-5 with Fe, however, resulted in a sensible collapsing of small-size pores, that is ∼4 nm (see Figure 5b), to large-size pores at 30–40 nm. It means that the co-impregnation of 0.8Zn-alk-HZSM-5 with Fe could afford more large-size pores, maybe by facilitation of more Si–O–Si bond cleavage. Conclusively, any catalytic changes for the alkali-treated HZSM-5 samples could be described in terms of newly formed 4 nm pores.

Figure 6.

NH₃-TPD profiles for the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 zeolites.

To investigate the effect of the strength and distribution of different acid sites on the catalysts, temperature-programmed desorption of ammonia (NH₃-TPD) experiments were carried out. Corresponding NH₃-TPD profiles are shown in Figure 6.

These profiles represented the TPD desorption peaks at two different temperature ranges consisting of 215–280 and 470–540 °C, designated to the chemical desorption of NH₃ from weak and strong acidic sites, respectively.²⁵ As evident from Figure 6, each profile could be deconvoluted to two Gaussian distributions. Primarily, comparison of these profiles suggested that the parent HZSM-5 had rather more acid site combined with stronger acid character (i.e. higher desorption temperature). Then, because of the remarkable decrease in acid-site concentration (Figure 6), the NH₃-TPD results showed that the alkali treatment with NaOH in alk-HZSM-5, 0.8Zn-alk-HZSM-5, and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites resulted in preferential desilication of the parent HZSM-5. Basically, removal of some Si atoms from the zeolite framework gave rise to destruction of the zeolite structure with subsequent loss of some acid sites.²¹ Similarly, Yang et al.⁴ attributed the growing number of weak acid sites on H–Zn/ZSM-5 to the drop in the number of the strong acid sites at 430 °C. They showed that the alkali treatment and Zn incorporation could lessen the coke formation process on the zeolite.

Furthermore, in Table 2, the density of the acid site on the catalysts was calculated from the peak areas shown in Figure 6.

Table 2. NH₃-TPD Data for the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 Zeolites.

	distribution and concentration of acid sites (mmol NH3 g–1)				peak temperature (°C)
catalyst	region I (weak)	region II (strong)	total	strong/weak	Td1	Td2
HZSM-5	1.103	1.199	2.302	1.087	258.8	474.2
0.8Zn-alk-HZSM-5	1.107	0.583	1.690	0.527	256.4	487.8
alk-HZSM-5	0.996	0.541	1.537	0.543	249.1	490.0
0.8Zn-HZSM-5	1.748	0.842	2.590	0.482	265.2	520.7
[0.2Fe,0.3Zn]-HZSM-5	1.867	0.867	2.734	0.464	218.5	512.5
[0.4Fe,0.6Zn]-HZSM-5	2.095	0.866	2.961	0.413	247.6	538.3
[0.2Fe,0.3Zn]-alk-HZSM-5	1.293	0.511	1.804	0.395	270.4	486.7

The NH₃-TPD results were collected in Table 2 showed that total acid sites of the parent zeolite significantly decreased during the impregnation of alk-HZSM-5, 0.8Zn-alk-HZSM-5, and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites, establishing the increase in total acid sites with metal incorporation. Furthermore, the parent HZSM-5 exhibited relatively more acid sites (2.302 mmol NH₃ g^–1) than 0.8Zn-alk-HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5. Where Zn and Fe were incorporated, both bands were thermally enlarged, decreasing in the intensity of the strong acid band (region II). Meanwhile, the acid site densities of catalysts were reduced from 1.199 in HZSM-5 to 0.583 in 0.8Zn-alk-HZSM-5 and 0.511 mmol NH₃ g^–1 in [0.2Fe,0.3Zn]-alk-HZSM-5 in region II. In Conclusion, the integration of Zn and Fe species into alk-HZSM-5 zeolite resulted in reduction of the strong acid sites, meaning that the acid character could be less important factor than the pore-size consideration in our chemically treated ZSM-5 samples for describing any observable catalytic differences.

Moreover, Figure S3 displays the pyridine-IR spectra, adsorbed on the HZSM-5, alk-HZSM-5, 0.8Zn-HZSM-5, 0.8Zn-alk-HZSM-5, [0.2Fe,0.3Zn]-alk-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, and [0.4Fe,0.6Zn]-HZSM-5 catalysts. Three characteristic peaks at 1450, 1545, and 1492 cm^–1 represented the amount of the Lewis acid sites (LASs), BASs alongside both BAS and LAS in the samples, respectively. Figure S3 also illustrates the subsequent increase in the amount of the LASs together with decreasing in the BASs after metal impregnation on the parent HZSM-5. Additionally, the alkali treatment on the parent zeolite gave rise to decrease in the amount of LASs and BASs. The estimated peak area for each of the said acid sites on the catalysts in Figure S3 generally registered corresponding significant increases in the proportion of the BAS (recorded at 1545 cm^–1) and the LAS (recorded at 1450 cm^–1) for the samples, suggesting the following order:

• (a)Suggested order for BAS order recorded at 1545 cm^–1

• (b)Suggested order for LAS recorded at 1450 cm^–1

These observations were also in agreement with NH₃-TPD results.

2.2. Catalytic Performance of the Zeolite Catalysts in MTA

Table S1 shows a L₁₆ Taguchi orthogonal array nominated for the test plan because the optimization process to obtain the best results for catalytic performance of the MTA process over low-Si/Al-ratio practically required 32 experiments to obtain the best resulted for the catalyst in terms of Si/Al, metal, alkali treatment, WHSV, and temperature parameters. Fundamentally, optimization of the parameters by Taguchi method is a simple way to attain the best operational condition, needless to do unnecessary experiments, as long as the parameters are unaffected by alternation in environmental conditions as well as the other noise factors. To achieve this goal, a superior design of orthogonal arrays is required by employing the Taguchi method, wherein a loss function should be considered for the deviation of the experimental from the desired values.⁵⁴ Meanwhile, the loss function values were also converted into a signal-to-noise (S/N) ratio. Normally, there are three categories of performance, treated as the index of S/N ratio in the analysis. The S/N ratio for each level of the catalyst preparation parameters was calculated on the basis of the S/N analysis. Hence, the finest level of a typical parameter would be the highest S/N ratio. According to the Taguchi method, four catalysts were prepared and tested in our experiments. Meanwhile, in the next step, three other samples were synthesized, given that the results of the previous phase. The selected catalyst with the highest activity was considered to be studied in MTA processes in detail. Then, the collected data from Table S1 were analyzed using Minitab-18 software to calculate the effect of each projected factor on the optimization conditions. The results were summarized in Table S2 and Figure S4. Figure S4 shows that HZSM-5 with Si/Al ratio < 20, Zn impregnated, alkali treated as well as MTA experiments at the lowest temperature and WHSV advocated the best performance for the catalyst.

In order to discuss catalytic performance of the prepared samples in MTA process, the liquid hydrocarbons of the products were evaluated by gas chromatography (GC) and GC–mass spectrometry (GC–MS) methods (Table 3).

Table 3. Analysis of the Hydrocarbon Cut for MTA Process, Catalyzed by the Zeolites Prepared in This Work at T_R = 375 °C.

	catalyst
component (wt %)	HZSM-5	0.8Zn-HZSM-5	alk-HZSM-5	0.8Zn-alk-HZSM-5	[0.2Fe,0.3Zn]-HZSM-5	[0.4Fe,0.6Zn]-HZSM-5	[0.2Fe,0.3Zn]-alk-HZSM-5
i-C5	1.7	1.1	4.5	1.0	1.3	1.3	0.9
C6 saturated	0.3	0.2	4.4	0.3	0.1	0.1	0.2
benzene	3.5	3.4	0.0	1.4	5.1	4.8	0.1
C7 saturated	0.5	0.4	0.2	0.7	0.1	0.1	0.6
toluene	21.2	28.4	13.8	19.4	30.1	30.9	15.9
ethyl benzene	1.8	1.9	1.8	2.4	1.5	1.5	2.4
m & p-XYL	20.6	23.4	30.1	22.6	21.4	22.6	33.9
o-XYL	9.6	10.6	9.3	14.6	8.4	9.1	10.8
M-E-BZ	2.4	2.9	3.6	4.7	1.9	2.1	4.2
3M-BZ	8.2	9.2	7.3	7.7	8.9	9.3	10.5
Ar-C10	10.0	7.9	9.4	9.9	8.6	7.9	9.0
C10+	10.8	5.1	7.3	5.7	8.1	7.8	7.9
others	9.4	5.5	8.3	9.6	4.5	2.5	3.6
sum	100	100	100	100	100	100	100
BTX	54.9	65.8	53.2	58.0	65.0	67.4	60.7
aromatics	77.3	87.7	75.3	82.7	85.9	88.2	86.8

Table 3 certifies the highest level of the aromatics (>85%) for the impregnated HZSM-5 samples, explicitly 0.8Zn-HZSM-5, [0.2Fe,0.3Zn]-HZSM-5, [0.4Fe,0.6Zn]-HZSM-5, and [0.2Fe,0.3Zn]-alk-HZSM-5. Moreover, Table 3 reserves the highest aromatic selectivity together with the lowest benzene content over the bimetallic catalysts for [0.2Fe,0.3Zn]-alk-HZSM-5. As shown in Table 3, the weight percent of benzene and xylenes for alk-HZSM-5 was 0.0 and 39.4 wt %, respectively. Moreover, this table demonstrates the weight percent of benzene and xylenes for [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst as 0.1 and 44.7 wt %, respectively. The observed low level of benzene for alk-HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 may largely be due to the synergistic effect between its higher mesoporous volume and stronger acid sites as the alkaline low Si/Al-HZSM-5 zeolites.

As seen in Figure 7, among the aforesaid samples, the highest aromatic selectivity together with the lowest benzene content over the bimetallic catalysts was found for [0.2Fe,0.3Zn]-alk-HZSM-5, higher than the parent HZSM-5.

Figure 7.

Liquid hydrocarbon products over HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites. (Reaction condition: T = 375 °C, WHSV = 2 h^–1).

Figure 7 compares the liquid hydrocarbon cut selectivity for [0.2Fe,0.3Zn]-alk-HZSM-5 with respect to the parent HZSM-5 in detail.

Along with difference in their catalyst activity, the comparative changes of the product distribution in these two catalysts were also notable. The diagrams in Figure 7 represent higher percentage of xylenes in the product, applying [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst than the parent HZSM-5. However, the most outstanding feature of the aromatic selectivity for [0.2Fe,0.3Zn]-alk-HZSM-5 zeolite was the lowest percentage of benzene (<1%) found for in the product mixture.

The constituents of liquid hydrocarbons produced in the presence of HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 are also separately illustrated in Tables 4 and 5.

Table 4. Analysis of the Hydrocarbon Cut for the MTA Process Catalyzed by HZSM-5 at T_R = 375 °C.

time (h)	3	4	5	6	8	average
component	wt %	wt %	wt %	wt %	wt %	wt %
i-C5	1.1	1.6	1.7	1.8	2.2	1.7
C6 saturated	0.1	0.2	0.2	0.3	0.6	0.3
benzene	4.4	4.0	3.4	3.1	2.5	3.5
C7 saturated	0.1	0.2	0.3	0.6	1.2	0.5
toluene	27.4	25.4	21.1	18.6	13.5	21.2
ethyl benzene	1.6	1.6	1.7	2.0	2.0	1.8
m & p-XYL	21.3	21.0	21.1	21.2	18.4	20.6
o-XYL	8.3	8.4	8.7	9.7	12.9	9.6
M-E-BZ	2.0	1.9	2.3	2.8	3.2	2.4
3M-BZ	8.9	8.5	8.5	8.2	6.9	8.2
Ar-C10	8.2	8.2	9.5	10.1	14.0	10.0
C10+	9.2	10.7	10.9	10.7	12.3	10.8
others	7.4	8.3	10.5	10.9	10.3	9.4
sum	100	100	100	100	100	100

Table 5. Analysis of the Hydrocarbon Cut for the MTA Process, Catalyzed by [0.2Fe,0.3Zn]-alk-HZSM-5 at T_R = 375 °C.

time (h)	2	3	4	5	6	average
component	wt %	wt %	wt %	wt %	wt %	wt %
i-C5	0.6	0.9	1.0	1.1	1.0	0.9
C6 saturated	0.1	0.2	0.2	0.3	0.3	0.2
benzene	0.1	0.1	0.1	0.1	0.2	0.1
C7 saturated	0.3	0.4	0.5	0.7	0.9	0.6
toluene	18.3	17.0	17.0	14.4	12.9	15.9
ethyl benzene	2.4	2.3	2.5	2.5	2.4	2.4
m & p-XYL	35.0	33.7	33.9	33.1	33.5	33.9
o-XYL	10.5	10.4	10.7	10.8	11.6	10.8
M-E-BZ	4.0	3.8	4.0	4.3	4.7	4.2
3M-BZ	11.5	10.8	10.3	10.1	9.9	10.5
Ar-C10	7.7	8.4	9.0	9.7	10.4	9.0
C10+	7.7	9.2	6.9	8.5	7.0	7.9
other	1.8	2.8	3.9	4.3	5.2	3.6
sum	100	100	100	100	100	100

Tables 4 and 5 show that the dominant hydrocarbons found in the products are 63.8 and 77.7% alkyl aromatics, that is toluene, ortho, meta, and para xylenes (o, m, and p-XYL), methyl ethyl benzene (M-E-BZ), and tri-methylbenzene (3M-BZ) for the parent HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5, respectively. The products exhibited a growth in weight percent of alkyl aromatics up to 13.9% after improvement. Compared with the parent HZSM-5 catalyst, Tables 4 and 5 indicate that the selectivity versus aromatic product increases in [0.2Fe,0.3Zn]-alk-HZSM-5. Referring to the previous reports revealed that Fattahi et al.¹⁸ represented average mass fraction of alkyl aromatic components for HZSM-5 was up to 70.70 wt %, wherein the existence of zinc element on the improved catalyst upgraded this value up to 77.30 wt %. In this report,¹⁸ a maximum increase in weight percent of alkyl aromatics up to 6.6% was observed over the Zn-impregnated (0.75 wt %) catalyst, with SiO₂/Al₂O₃ ratio of 54 at WHSV equal to 4 h^–1, while in our experiments a 13.9% increase in weight percent of alkyl aromatics was observed for [0.2Fe,0.3Zn]-alk-HZSM-5. As revealed in Table 5, the analysis of the hydrocarbon cut in the MTA process on the HZSM-5 catalyst indicates on the reduction of the weight percent of toluene from 27.4 to 13.5 wt %, after 8 h, meanwhile executing the process on the [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst exhibited a weight percent decrease from 18.3 to 12.9 wt % for the production of toluene after 6 h. The weight percent of 3M-BZ also reduced from 8.9 to 6.9 wt %, after 8 h for HZSM-5, while the corresponding change in the presence of [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst was observed as 11.5 to 9.9 wt %, after 6 h.

The product distributions in the gas phase over [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst are demonstrated in Table 6.

Table 6. Product Distributions in Gas Phase for the MTA Process, Catalyzed by [0.2Fe,0.3Zn]-alk-HZSM-5 at T_R = 375 °C.

time (h)	1	2	3	4	5	6	8	11
component	mole %	mole %	mole %	mole %	mole %	mole %	mole %	mole %
methane	17.6	14.7	17.7	15.1	12.5	11.8	19.3	31.3
ethane	3.1	2.4	2.5	2.4	1.9	2.2	3.8	1.8
ethene	9.8	15.3	20.6	28.3	47.9	57.6	49.6	19.3
propane	29.1	20.4	22.6	15.2	8.92	0.1	0.1	0.1
normal butane	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.2
iso-butane	0.0	0.0	0.0	0.0	0.0	0.0	0.1	0.0
butene	1.4	2.1	2.8	4.1	4.5	4.9	2.7	1.9
di methyl ether	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
methanol	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
other	39.0	45.1	35.2	34.9	24.3	23.4	24.4	45.4
sum	100	100	100	100	100	100	100	100

Table 6 summarizes the gas-phase product distribution, namely methane, ethane, ethene, and propane. The data in Table 6 signify that the mole percent of ethylene increases with time on stream (TOS), for 6 h, attributed to the enhancement of the so-called shape-selective effects of the catalyst by occurring coke deposition in the cavities of the catalyst, leading to suppression of the formation of the larger molecules. Meanwhile, the gaseous product distributions over the [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst are showed in Figure 8.

Figure 8.

Product distributions in gas phase for the MTA process, catalyzed by [0.2Fe,0.3Zn]-alk-HZSM-5 at T_R = 375 °C.

Figure 8 indicates the higher methane formation after 8 h, due to the formation of more carbon deposition, leading to more demethylation reactions.

The average production of the aromatic components (BTX) over HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 samples is demonstrated in Figure 9.

Figure 9.

Average production of BTX components over the parent HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 samples for 8 h on stream.

Evidently, the [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst exhibits great improvements in selectivity for aromatic hydrocarbons, in especial for xylenes (Figure 9). The selectivity of the aromatic components over HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 is also addressed in Table 7.

Table 7. Aromatics Distributions of the MTA Reaction on the Various Catalysts, Prepared in the Present Work^a.

	selectivity (%)
catalyst	benzene	toluene	xylenes	3M-BZ	ethyl benzene	M-E-BZ	Ar-C10
HZSM-5	3.5	21.2	30.2	8.2	1.8	2.4	10.0
[0.2Fe,0.3Zn]-alk-HZSM-5	0.1	15.9	44.7	10.5	2.4	4.2	9.0
alk-HZSM-5	0.0	13.8	39.4	7.3	1.8	3.6	9.4

^aReaction conditions: 375 °C, 1 atm, WHSV = 2 h–1; data obtained at 8 h on stream.

Table 7 represents that alk-HZSM-5 catalyst produced 75.3 wt % aromatics along with xylenes and benzene selectivity as 39.4 and 0.0%, respectively. Although the alkali treatment of HZSM-5 to give alk-HZSM catalyst could considerably promote the aromatic products in the MTA process, but co-impregnation of alk-HZSM-5 to [0.2Fe,0.3Zn]-alk-HZSM-5 was capable to upgrade the hydrocarbon cut to: 86.8 wt % aromatics, consisting of xylenes selectivity 44.7%, higher than the best results reported before.³³ More selectivity for xylenes (44.7%) alongside very low benzene content (0.1%) in the products of [0.2Fe,0.3Zn]-alk-HZSM-5 zeolite recommended it as an eligible candidate for MTA process.

Figure 10 demonstrates that higher benzene (4.8–5.1%), toluene (>30%), as well as lower xylenes (<32%) selectivity were observed for [0.2Fe,0.3Zn]-HZSM-5 and [0.4Fe,0.6Zn]-HZSM-5, wherein the metal content of the catalyst increased.

Figure 10.

Selectivity of the aromatic products on various catalysts for the MTA reaction. (Reaction condition: T = 375 °C, WHSV = 2 h^–1).

Figure 10 denotes the selectivity for aromatic hydrocarbons 66.9 wt %, benzene selectivity 5.1 wt %, and toluene selectivity 30.1% of the [0.2Fe,0.3Zn]-HZSM-5 catalyst. By increasing the metal content of this catalyst to afford [0.4Fe,0.6Zn]-HZSM-5 catalyst, the aforesaid selectivity did not experience any significant changes. Figure 10 also establishes that the alkali-treated sample, [0.2Fe,0.3Zn]-alk-HZSM-5, exhibits the following selectivity as: 64.9 wt % for aromatic hydrocarbons, 0.1 wt % for benzene and 15.9% for toluene. In these cases, the decrease in toluene and benzene contents in the product has evidently been compensated with increasing in the xylene content, explicitly m- and p-xylenes. These assumptions were also in agreement with our previous conclusion about the PSDs shown in Figure 5 for these samples. As evident from Figure 5, all of those samples having pores with 40 nm were capable to produce more xylenes at the expense of reducing in benzene and toluene contents. The mechanism of this proposal could be attributed to eligibility of 40 nm pore size for more alkylation of benzene and toluene to xylenes. Although, Jiang et al.⁴³ showed that the impregnation of Fe on the ZSM-5 catalyst resulted in more BASs, but the efficiency of Fe-impregnation to promote the Friedel–Crafts alkylation as well as the effect of acid sites on the samples should not be ignored.

Furthermore, Figure 10 shows that alkaline treatment on the parent HZSM-5 resulted in the production of benzene-free product, having low toluene (13.8%) and high xylenes (39.4%) selectivity. Among the xylenes, m- and p-xylenes experienced more increase, owing to rising in metal ion contents on the parent HZSM-5. Olson and Haag⁵⁵ reported the relationship between the diffusion time of o-xylene over various HZSM-5 catalysts on p-xylene selectivity. It is well known that MTA is an acid catalytic reaction, enjoying from the synergetic effect between BAS and LAS.⁵⁶ Consistent with this proposal, the stronger BASs, the more active sites for the reaction. An increase in the amount of acid sites, however, improves the olefin cyclization reaction together with deep alkylation reaction of the aromatic products, leading to a subsequent increase in the formation of polyalkylaromatics in the reaction.

Figure 11 compares the conversion of methanol as a function of TOS at HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites.

Figure 11.

Conversion of methanol as a function of TOS at HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 zeolites. (Reaction condition: T = 375 °C, WHSV = 2 h^–1).

Practically, [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst in Figure 11 were found to best active sample in MTA process so that their initial catalytic activities remained fully constant up to 4 h, for example keeping >97% of its activity after 8 h (Figure 11). A typical 75.8% catalytic activity was also obtained for [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst after 12 h, checked as the best catalyst for MTA processes within our zeolite series. The observed stability was also consistent with several recent reports.^38,50 For instance, as the latest similar work,⁴⁰ Lai et al. studied the methanol aromatization over Ga-doped desilicated HZSM-5 with an SiO₂/Al₂O₃ ratio of 23. They showed that methanol conversion over the parent HZSM-5 was 80% only for 11.5 h TOS. In another study,⁵⁸ Zhang et al. reported the cadmium-modified HZSM-5 for the selective conversion of methanol for 4 h.

Moreover, the coke formation in the fresh and used [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst was studied by thermogravimetric analysis (TGA) (Figure 12).

Figure 12.

TGA curve of [0.2Fe,0.3Zn]-alk-HZSM-5 zeolite.

The weight loss at 25–150 °C was attributed to the elimination of the physically adsorbed water in the catalysts. The coked [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst exhibited a typical weight loss of 10% after 8 h on stream. Besides, the coke content of the spent catalyst resulted in a weight loss at ca. 150–800 °C. Meanwhile, the fresh [0.2Fe,0.3Zn]-alk-HZSM-5 sample represented 8.6% weight loss after 8 h on stream. These observations suggested that the coke formation process was a little more than the reported value for Fe-ZSM-5, which was attributed to more extensive porosity of [0.2Fe,0.3Zn]-alk-HZSM-5. Moreover, the weight loss in the temperature range of 500–600 °C could be attributed to removing of the organic template, tetra propyl ammonium ion (TPA⁺) in fresh noncalcinated-[0.2Fe,0.3Zn]-alk-HZSM-5 zeolite. According to the abovementioned results, it is deduced that [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst displays remarkable catalytic performance in the MTA process in comparison to the parent HZSM-5 zeolite based on more production of aromatics, explicitly m- and p-xylenes.

3. Conclusions

The following conclusions could be drawn from this research:

• (1)By employing the Taguchi method alongside analysis of data, the number of the required experiments to accomplish a systematic study on the effect of such parameters as Si/Al, the type of metal impregnation, alkali treatment, WHSV, and temperature was reduced from 32 to 16 parameters.

• (2)The postmodification of alkali-modified low-Si/Al HZSM-5 by Fe(NO₃)₃ and Zn(NO₃)₂ resulted in a new bimetallic [0.2Fe,0.3Zn]-alk-HZSM-5 catalyst, enjoying from: (a) larger surface area and (b) lower pore diameter and particle size with respective to the parent HZSM-5.

• (3)The catalyst lifetime for both HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5 catalysts was significantly high enough to keep >97% of its activity after 8 h and 75.8% after 12 h in MTA process.
• (4)The incorporation of Fe decreases the acidity of the catalyst resulted in subsequent low coke formation.
• (5)The combination of long lifetime for the catalyst and high aromatics selectivity for [0.2Fe,0.3Zn]-alk-HZSM-5 makes this work a leading study on application of metal impregnation process on a low-Si/Al HZSM-5 zeolite as a significant candidate for conversion of methanol to benzene-free aromatics, specifically m- and p-xylenes
• (6)Mechanistically, the large pores, that is 40 nm, found in metal-impregnated alkali-treated HZSM-5 were proposed as the leading structural factor for the observed promotional catalytic effect.

4. Experimental Section

4.1. Catalyst Preparation

4.1.1. Materials

TPAOH, 40 wt % aqueous solution, aluminum isopropoxide (97 wt %), tetraethyl orthosilicate (98 wt %), iron(III) nitrate (Fe(NO₃)₃·9H₂O), zinc nitrate (Zn(NO₃)₂·6H₂O), ammonium nitrate (NH₄NO₃), nitric acid (HNO₃), and sodium hydroxide (NaOH) were obtained from Merck company (Germany). All chemical reagents were used without any further purification.

4.1.2. HZSM-5 Synthesis

The HZSM-5 zeolite was prepared via the hydrothermal method as stated by the general technique reported by Karimi et al.⁵⁷ with few modifications. The gel molar composition was 1Al₂O₃/22SiO₂/2.7TPAOH/5Na₂O/2500H₂O. Finally, the calcined NaZSM-5 (at 550 °C for 8 h) was undergone ion exchange treatment by stirring with a 0.8 M NH₄NO₃ solution (by the 1:8 mass ratio) at 80 °C for 12 h prior to washing with distilled water. The NH₄ZSM-5 zeolite was obtained after repeating the same procedure for three times on NaZSM-5. The proton-form of the zeolite powder was also prepared after drying of the NH₄ZSM-5 sample at 110 °C overnight prior to calcination at 550 °C for 5 h.

4.1.3. Alkaline Treated HZSM-5

The NaZSM-5 zeolite was desilicated by employing the method reported before,²¹ the NaZSM-5 catalyst was stirred with 0.3 M NaOH solution (the ratio of 8 mL NaOH solution/1 g zeolite) at 80 °C for 2 h. Later, the suspension was cooled down prior to filtering and neutralizing the remaining of NaOH on the zeolite by diluted HNO₃. The zeolite sample was dehydrated at 110 °C for 5 h before stirring in a 0.8 M NH₄NO₃ solution at 80 °C for 12 h in 1:8 mass ratio. The slurry was filtered and decanted with distilled water as well as executing the aforesaid procedure on the sample for three times. After drying the sample at 110 °C overnight, the calcination was accomplished at 550 °C for 5 h to form alk-HZSM-5 zeolite.

4.1.4. Catalysts Promotion

The catalyst was promoted by Zn via wet impregnation of alk-HZSM-5 and HZSM-5 zeolites with aqueous solutions of Zn(NO₃)₂·6H₂O to achieve 0.8 wt % Zn loading, respectively. The reaction mixtures were stirred for 24 h, and the impregnated samples dried at 110 °C overnight and then calcined at 550 °C for 6 h.²¹ The resultant samples were denoted as 0.8Zn-alk-HZSM-5 and 0.8Zn-HZSM-5, respectively.

The initial wetness method with an aqueous solution of Zn(NO₃)₂·6H₂O and Fe(NO₃)₃·9H₂O on HZSM-5 and alk-HZSM-5 zeolites was employed to achieve 0.2 wt % Fe and 0.3 wt % Zn loading as the other bimetallic-promoted catalysts, while the mixtures were stirred for 24 h. Subsequently, the samples dried at 110 °C for 16 h prior to aerobic calcination at 550 °C for 6 h. The catalysts were designated as [0.2Fe,0.3Zn]-HZSM-5 and [0.2Fe,0.3Zn]-alk-HZSM-5, respectively. The same method was applied to make the final promoted catalyst ([0.4Fe,0.6Zn]-HZSM-5) by employing co-impregnation of HZSM-5 to achieved typical loading as 0.4 wt % Fe and 0.6 wt % Zn. In all of the abovementioned cases, the metals loading on HZSM-5 zeolite were determined by EDXS analysis.

4.2. Catalytic Characterization

The catalyst samples were characterized by employing XRD, X-ray fluorescence (XRF), N₂ adsorption/desorption, NH₃-TPD, TGA, FT-IR, FE-SEM, and FE-SEM/EDXS techniques. The applied instruments were as follows:

4.2.1. X-Ray Diffraction

To determine the extent of crystallinity for different samples, the XRD patterns were carried out within 2θ = 5–80° by a D8 ADVANCE Bruker X-ray diffractometer having Cu K_α radiation (λ = 1.5406 Å).

4.2.2. N₂ Adsorption/Desorption Isotherms

The porosity of the zeolites was monitored by means of the nitrogen physisorption at −196 °C using a Micromeritics ASAP 2010 instrument. Total specific surface areas (S_BET) of the catalysts were calculated by employing nitrogen adsorption isotherms using the BET method in the P/P₀ range 0.05–0.25. Furthermore, in order to measure the surface area of the catalysts, prior to adsorption of N₂, 250 mg of the samples was degassed at 300 °C for 3 h. The total pore volume (V_Total) was evaluated by the amount of adsorbed nitrogen at a relative pressure of ∼0.99. The size distribution of the mesopores was calculated by the adsorption isotherm using the BJH method. In order to calculate the micropore surface area (S_Micro) and the micropore volume (V_Micro) in P/P₀ = 0.1–0.4, the t-plot method was applied. The mesopore volume (V_Meso) could be calculated based on the difference between the calculated total volume and the relevant micropore volume.

4.2.3. Field-Emission Scanning Electron Microscopy

A Tescan MIRA3-LMU scanning electron microscope was employed to determine the particle size and morphology of the zeolites, working at a potential difference of 15 kV. Moreover, an EDXS system is coupled with the FE-SEM chamber. In order to prepare the samples, the zeolite crystals were deposited on a silicon wafer using dispersion, whereby the samples were subsequently covered with a gold film.

4.2.4. NH₃-TPD

In order to determine the total acidity along with the type (weak and strong) of acidic sites on the catalyst samples, the NH₃-TPD method was employed by using a Micromeritics TPD/TPR 2900 chemisorption analyzer. In these experiments, 0.1 g of samples was pretreated at 500 °C for 2 h prior to measure the desorption of NH₃ in the range of 100–700 °C.

4.2.5. X-ray Fluorescence

The Si/Al molar ratio of the HZSM-5 zeolite was accomplished on an Axios XRF spectrometer.

4.2.6. FT-IR Spectroscopy

FT-IR data were collected at room temperature by using KBr pellets technique on an RXI-PerkinElmer spectrometer with the resolution of 4 cm^–1 to address surface functional groups, in the range of 500–4000 cm^–1.

4.2.7. Thermogravimetric Analysis

The TGA system was hired to monitor the thermal decomposition behavior for the template within the fresh catalyst alongside the coke content in corresponding spent catalyst. The samples were run on a PerkinElmer Diamond TGA instrument at 30–800 °C by a heating rate of 10 °C min^–1. After heating, 6.5 mg of catalyst up to 150 °C, the system was also purged by a nitrogen flow (6 L min^–1) prior to heating (2 °C min^–1) under the air flow (6 L min^–1) from 30 to 800 °C.

4.2.8. TEM Images

TEM images were recorded on Zeiss EM900 equipment, operated at 100 kV. The samples were prepared by sonicating the powder in ethanol for 15 min and dropping the suspension onto the carbon-coated Cu grids.

4.2.9. FT-IR of Pyridine Adsorption Spectroscopy (Py-IR)

Py-IR spectra of samples were recorded on an RXI-PerkinElmer FT-IR spectrometer. The samples were activated under a vacuum of 5 × 10^–3 Pa using by Alcatel Adixen 2005SD Pascal vacuum pump at 175 °C for 3 h. Then, pyridine was adsorbed onto the samples at 30 °C for 1 h, and the samples were heated up to 150 °C prior to running their Py-IR spectra.

4.3. Catalytic Performance

A fixed-bed tubular reactor was applied to determine the efficiency of the catalyst samples in the MTA process at 375 °C under ambient pressure. A graduated burette attached to a micro tube pump (prep pump, Chem Tech Co. Ltd.) working at a flow rate of 0.1 mL/min was applied for pumping of liquid methanol at WHSV of 2 h^–1. Methanol stream was preheated at 120 °C prior to delivering to the setup. Meanwhile, this stream was mixed with nitrogen and delivered to the reactor. Figure 13 represents the experimental setup.

Figure 13.

Representation of the lab scale setup.

The tubular reactor is composed of a stainless steel tube (length 380 mm; i.d. 9 mm), heating at three zones by employing an electric furnace at 375 °C. In order to measure the axial temperature profile, it was equipped with two thermocouples attached to the top and the bottom of the reactor. Besides, a perforated circular plate was welded to the upper thermo-well to serve as the catalyst holding support. The catalyst was located in the middle part of the reactor, wherein the isothermal conditions were established. A temperature of the condenser was maintained at −1 °C served to remove the condensable gases from the products passing through the reactor, while the noncondensable were led through a product separator, a bubble flow meter prior to venting. The gas phase was sent to the gas flow meter and a gas chromatograph Agilent 7890A, analyzing the gaseous phase with nitrogen as carrier gas. A GC-Varian 3800 instrument was applied to analyze offline the liquid products that were collected after every hour at the end of the experiments. Furthermore, the liquid hydrocarbons collected over a period of 8 h were analyzed by GC–MS instrument. For this purpose, a Hewlett-Packard (HP, Palo Alto, USA) HP 6890 series GC fitted with a split/splitless injector and a HP 5973 mass-selective detector system were applied, whereby the catalytic performance of the reaction was measured. Henceforth, the methanol conversion (x_MeOH) in the MTA reaction was calculated by eq 1.

1

where N signifies the number of moles, superscript “i” denotes the inlet of the reactor and superscript “o” refers to the outlet.

Glossary

Abbreviations

AIP

aluminum isopropoxide

Ar-C10

C₁₀ aromatics

BAS

Brønsted acid sites

BET

Brunauer–Emmett–Teller

BJH

Barrett–Joyner–Halenda

DME

dimethyl ether

EDX

energy dispersive X-ray

FE-SEM

field emission scanning electron microscopy

FT-IR

Fourier transforms infrared

GC

gas chromatography

GC–MS

gas chromatography–mass spectrometry

LASs

Lewis acid sites

M

mol dm^–3

3M-BZ

tri-methylbenzene

M-E-BZ

methyl ethyl benzene

m & p-XYL

meta and para xylenes

MTA

methanol to aromatics

MTG

methanol to gasoline

MTH

methanol to hydrocarbons

MTO

methanol to olefins

NH₃-TPD

temperature-programmed desorption of ammonia

o-XYL

ortho xylene

PSD

pore size distribution

TEM

transmission electron microscopy

TEOS

tetraethyl orthosilicate

TGA

thermogravimetric analysis

TOS

time on stream

TPAOH

tetrapropylammonium hydroxide

WHSV

weight hourly space velocity

XRD

X-ray diffraction

XRF

X-ray fluorescence

Supporting Information Available

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b01380.

• FTIR spectra of the samples; FESEM/EDS analysis; L₁₆ experimental plan; response table for the mean; and experimental data used for effect of each design parameter on the mean of aromatics percent (PDF)

The authors declare no competing financial interest.