English

• The conversion of methanol to propene (MTP), as a non-petroleum route for propene production, is a potential substitute or supplement for the current steam cracking and fluid catalytic cracking units that produce propene. Several MTP processes have been industrialized, including the fixed-bed MTP of Lurgi, Gemany^[1] and the fluid-bed MTP of Tsinghua University, China^[2]; the former used nanosized high-silicon H-ZSM-5 zeolite, whereas the latter employed SAPO-34/SAPO-18 composite molecular sieves as the catalyst. Recently, great efforts have been also made in the synthesis of light olefins directly from syngas^{[3, 4]}.

  The catalytic performance of ZSM-5 zeolite in MTP was related to its textural properties, diffusion limitation, and acidity; especially, a moderate acid strength and density was in general expected^[5], which could be regulated by altering the Si/Al ratio and synthesis conditions^{[6, 7]}. An increase in the aluminum content (i.e. a decrease in the Si/Al ratio) in the MFI zeolites could enhance the propagation of the aromatic-based methylation/dealkylation cycle in comparison with that of the olefin-based methylation/cracking cycle in the methanol-to-hydrocarbons (MTH) process, leading to a higher selectivity to ethene^[8-10].By contrast, as reported by Liu et al^{[11, 12]}, an increase in the Si/Al ratio in ZSM-5 zeolite could weaken the strong acid sites and improve the propene yield and the propene/ethene (P/E) ratio in the MTP products. To further adjust the acid strength and quantity, ZSM-5 zeolite was modified with P and Zr-P^{[4, 13, 14]}; for the conversion of dimethyl ether to propene over P and Zr-P modified ZSM-5 at 723 K, 0.1 MPa and a space velocity of 4.6 h^-1, the conversion of dimethyl ether reached 100%, with a selectivity of 45.4% to propene and a P/E ratio of 9.9.

  On the other hand, the diffusion limitation in ZSM-5 zeolite is determined by the particle size, aggregation, morphology, pore size, and so on. Firoozi et al^[15] reported that in comparison with the micron-sized ZSM-5 zeolite, the nanosized ZSM-5 exhibited higher selectivity to propene, lower initial activity, and lower selectivity to C₄ hydrocarbons; moreover, with the proceeding of the MTP reaction, the activity of nanosized ZSM-5 zeolite is significantly higher than that of micron-sized ZSM-5 zeolite. In addition, the "alkali-treated" nanosized-ZSM-5 zeolite could form meso-porous structure, which was beneficial to reducing the diffusion limitation and thus enhancing its catalytic performance in MTP. Mei et al^[16] also observed that in comparison with conventional ZSM-5, the selectivity to propene over the alkali-treated ZSM-5 zeolite was increased from 33% to 42%, whereas the selectivity to ethene was decreased from 13% to 4.5%, with a P/E ratio approaching 10.

  In our previous works^{[17, 18]}, MO_x/H-ZSM-5 (M= Pb, Sb, and Bi) catalysts were prepared by a wet impregnation method with nitrate precursors and used in MTP; it was found that the modification with proper metal component could reduce the apparent pore size and the strength and quantity of strong acid sites in H-ZSM-5, which was of benefit in improving the selectivity to propene.

  For the synthesis of ZSM-5 zeolite, it is well known that a high fluoride/silicon ratio in the synthesis mixture under acidic or neutral media was able to reduce the number of strong acid sites. However, the presence of fluorine ion may also reduce the nucleation and crystal growth rates, producing relatively larger ZSM-5 crystals^[19-24]. We speculate that by using hydroxyl ions in addition to fluorine as a second mineralizer, the overgrowth of particle size could be avoided. To test this hypothesis, double mineralizers were used in this work to adjust the textural properties of ZSM-5 zeolite; the effect of F^-/Al₂O₃ molar ratio on the catalytic performance of synthesized H-ZSM-5 in MTP as well as the relationship between the structure and catalytic performance were then investigated.

  1.   Experimental

  1.1   Catalyst preparation

  ZSM-5 was synthesized by a static hydrothermal approach, using tetraethoxysilane (chemically pure, Sinopharm Chemical Reagent Co., Ltd), sodium aluminate, and tetrapropylammonium hydroxide (TPAOH, 40%) as the silicon source, aluminum source, and template agent, respectively. NaOH (chemically pure, Sinopharm Chemical Reagent Co., Ltd) and NH₄F (chemically pure, with a content of 96.6%, Tianjin Damao Chemical Reagent Factory) were used as the mineralizers. The molar composition of the synthesis mixture is F^-:Al₂O₃:SiO₂:TPAOH :Na₂O:H₂O = x :1:150:15:7:100; x = 0, 6, 12, and 18, where the corresponding catalysts are numbered as S-1, S-2, S-3, and S-4, respectively. To form the synthesis mixture, sodium aluminate, TPAOH, NaOH, and given amount of NH₄F were dissolved in water to form a solution, which was stirred for 0.5 h at room temperature; tetraethoxysilane was then added to this solution under stirring. The mixture was further stirred to obtain a homogeneous gel, with a pH value of higher than 7. The gel was transferred to a stainless-steel autoclave with a polytetrafluoroethene (PTFE) liner; the sealed autoclave was placed in an air oven maintained at 443 K for 24 h. After cooling and centrifugal separation, the sediment was washed with water until the pH value reached 8. The sediment was dried overnight at 373 K and then calcinated at 823 K for 8 h to remove the organic template and to obtain the Na-ZSM-5 zeolite powder. Ion exchange was carried out twice with 1 mol/L hydrochloric acid at 363 K for 4 h, followed by calcination at 823 K in air for 8 h to give the H-form ZSM-5 products. Finally, the ZSM-5 catalysts of 20-40 mesh were prepared by tableting, crushing, and sieving.

  1.2   Catalyst characterization

  X-ray diffraction (XRD) patterns were collected on a Rigaku D/max-2500 powder diffractometer with Ni-filter Cu Kα radiation (λ= 0.15418 nm). The diffraction patterns were recorded in the angle range of 2θ = 5°-35° with steps of 0.02° and an interval of 0.5 s at 40 kV (tube voltage) and 30 mA (tube current). The relative crystallinity of ZSM-5 was calculated based on the intensity of the peaks in 2θ = 23°-25°, using S-1 as the reference:

  $ C\left( \% \right) = \frac{{\sum I(2\theta = 23^\circ - 25^\circ )}}{{\sum {I_0}(2\theta = 23^\circ - 25^\circ )}} \times 100\% $

  (1)

  Elemental analysis for the Si/Al ratio was performed on an inductively coupled plasma-atomic emission spectrometer (ICP-AES, Shimadzu ICPE-9000).

  The morphology and size of the catalysts were characterized by a Jeol 6300-F scanning electron microscope (SEM, Shimadzu Corporation) operated at 10-15 kV. The catalyst was placed on a carbon tape without any metal coating.

  N₂ adsorption/desorption was performed at 77 K in an ASAP 2000 apparatus. The total specific surface areas were calculated according to the Brunauer-Emmett-Teller (BET) method, and the micropore surface areas were determined by the t-plot method.

  NH₃-TPD analysis was performed on a TL5000-2 multi-functional adsorption instrument (Tianjin Xianquan Industry and Trade Company) using a thermal conductivity detector to monitor the NH₃ desorption. The H-ZSM-5 catalyst (0.1 g) was heated at 873 K in a He flow (30 mL/min) for 3 h, and then cooled to 373 K. NH₃ adsorption was performed under a flow of 10% NH₃/He (30 mL/min) for 1 h. Then, the catalyst was flushed with He gas at a rate of 30 mL/min for 1 h. The NH₃-TPD was promptly started at a heating rate of 10 K/min from 373 to 873 K.

  FT-IR spectra of adsorbed pyridine were measured on an IR 560 FT-IR spectrometer (Nicolet). The self-supporting disk (15-20 mg) of the catalyst was placed in an IR cell equipped with a vacuum system, and pretreated by evacuation (10^-2 Pa) at 623 K for 2 h. It was then cooled to 473 K and kept at this temperature for 10 min. The spectrogram at 473 K was scanned as the background. Pyridine was adsorbed onto the catalyst disk at 298 K for 1 h, physically desorbed at 473 K for 1 h, and then the infrared spectra within 1700-1400 cm^-1 were recorded.

  1.3   Catalytic test

  The MTP reaction was carried out at atmospheric pressure in a fixed-bed stainless steel reactor with an inner diameter of 15 mm and length of 550 mm. The catalyst powder was compressed to wafers, then crushed and sieved to 40-60 mesh particles. In all tests, 0.5 g catalyst was placed in the middle of the reactor and activated in situ at 823 K for 3 h under an N₂ flow of 30 mL/min. The products were analyzed online with a gas chromatograph (GC-950) equipped with a flame ionization detector and an HP-Plot Q capillary column (30 m, 0.53 mm i.d., stationary phase thickness of 40 μm). Both methanol and dimethyl ether (DME) were regarded as reactants in the calculations.

  2.   Results and discussion

  2.1   Effect of F^-/Al₂O₃ ratio on the structure and acidity of H-ZSM-5 catalyst

  Figure 1 shows that the zeolite catalysts synthesized with or without fluoride ions all have similar XRD patterns, with the characteristic diffraction peaks of H-ZSM-5 at 7.9°, 8.8°, 23.1°, 23.3°, 23.9°, and 24.4°^[25]. The sample S-4 (F^-/Al₂O₃=18) displays an exceptionally strong diffraction peak at 21.8°, which is the characteristic peak of microcrystalline a-SiO₂^[26]. The amorphous zeolites lacking in long-range order with compositional and chemical similarities to the parent MFI zeolite exhibits a peak at 18.5°^{[27, 28]}. Moreover, the diffraction peaks shift slightly toward lower angles with the increase of the F^-/Al₂O₃ ratio, due to an increase in the Si/Al ratio of H-ZSM-5, as given in Table 1. In addition, with an increase in the F^-/Al₂O₃ ratio, the relative crystallinities of the synthesized H-ZSM-5 zeolites (S-1, S-2, S-3, and S-4) are decreased (100%, 64%, 63%, and 23%, respectively). However, the SEM images shown in Figure 2 suggest that all the zeolite products display mainly spherical or nearly spherical morphology.

  Figure 1

  

  Figure 1.  XRD patterns of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片

  Table 1

  

  Table 1.  Relative crystallinity, SiO₂/Al₂O₃ ratio, and textural properties of H-ZSM-5 prepared with different F^-/Al₂O₃ ratios

  下载: 导出CSV

  Catalyst	Relative crystallinitya/%	SiO2/Al2O3b	Surface area c A/(m2·g-1)				Pore volume v/(cm3·g-1)
  			total	micro	meso		totalf	microg
  S-1	100	285	388.0	356.0	32.0		0.2422	0.1774
  S-2	64	295	386.7	348.7	37.9		0.2277	0.1740
  S-3	63	359	154.8	126.8	27.9		0.1238	0.0657
  S-4	23	395	109.7	92.6	17.1		0.0797	0.0476
  note: a: the relative crystallinity of ZSM-5 was calculated based on the intensity of the diffraction peaks at 2θ of 22°-25°, using S-1 as the reference; b: the SiO2/Al2O3 ratio was determined by ICP-AES; c: total BET surface area was obtained by the BET method using adsorption data in p/p0 ranging from 0.05 to 0.25, whereas micro and meso-surface areas were determined by the t-plot method; d: the total pore volume was estimated from the adsorbed amount at p/p0 = 0.99, whereas the micro pore volume was determined by the t-plot method

  Figure 2

  

  Figure 2.  SEM images of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片

  The crystallinity decrease of ZSM-5 synthesized in OH^--F^- system can be attributed to four reasons. First, fluoride ions may not only promote the hydrolysis of tetraethoxysilane, but also accelerate the generation of Si-O^- structural units^[29]. The fluorine (and OH^-) -silicon (aluminum) complex is formed in the presence of silicon and aluminum sources^[30], leading to a decrease in the amount of formed Si-O-Al^- structural units. Therefore, the double mineralizers could slow down the consumption of aluminum source, leading to an increase in the Si/Al ratio of synthesized H-ZSM-5 zeolite; as given in Table 1, the Si/Al ratio of H-ZSM-5 decreases significantly with an increase in the F^-/Al₂O₃ ratio of the synthesis mixture. Second, the electrostatic interaction between TPA⁺ and fluorine (and OH^-) -silicon (aluminum) complex may weaken the interactions among Si-O^-, Si-O-Al^-, and O-Al^-, which can then reduce the template effects of TPA⁺ and then alter the aluminum distribution. Third, the presence of fluoride ion may also reduce the nucleation and crystal growth rates^[31]; the crystallization of ZSM-5 zeolite may then need a longer time and microcrystalline SiO₂ is then formed during the crystallization process, as shown in Figure 2. Lastly, microcrystalline SiO₂ may also partly cover the crystal face of H-ZSM-5 zeolite (Figure 2). The crystalline grains shown in Figure 2 are basically the same, with the average size smaller than 100 nm. However, they are rather different in the surface morphology: S-1 and S-2 have smooth surface, whereas S-3 and S-4 show some crystallite-like structure, which cannot be removed even by the hydrochloric acid treatment. The XRD results (Figure 1) suggest that such microcrystalline structure is made of SiO₂.

  Table 1 gives the textural properties of H-ZSM-5 prepared with different F^-/Al₂O₃ ratios. Both the surface area and pore volume decrease with the increase of F^-/Al₂O₃ ratios, from 388 m²/g and 0.1774 cm³/g in S-1 to 109.7 m²/g and 0.0476 cm³/g in S-4, respectively. The XRD, N₂ sorption, and SEM results indicate that as the H-ZSM-5 pore openings and channels are modified by microcrystalline SiO₂, the relative crystallinity, surface area, and pore volume all are decreased with the increase of F^-/Al₂O₃ ratio.

  It is well known that a higher Si/Al ratio can reduce the strength and number of strong acid sites in the H-ZSM-5 zeolite. Figure 3 shows that S-1, S-2, S-3, and S-4 all are similar in their characteristic NH₃-TPD profiles. Two desorption peaks are observed, one at low temperature (478 K) corresponds to the weak acid sites, and another at high temperature (713 K) to the strong acid sites^[32]. The peak temperature is related to the acid strength, while the peak area to the quantity of acid sites. Obviously, with the increase of F^-/Al₂O₃ ratio, the desorption peak, i.e. the amount of strong acid sites is signifcantly decreased.

  Figure 3

  

  Figure 3.  NH₃-TPD profiles of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片

  Figure 4 shows the Pyridine-FTIR (Py-FTIR) spectra of S-1, S-2, S-3, and S-4 at 473 K. The peaks for Brønsted and Lewis acid sites lie at 1545 and 1447 cm^-1, respectively. Both Brønsted and Lewis acid sites have a peak at 1493 cm^-1, while physical adsorption displays a peak at 1599 cm^{-1[33, 34]}. S-1, S-2 and S-3 display all these four peaks, whereas S-4 only shows an extremely weak peak at 1447 cm^-1. It means that probably no strong acidic sites in S-3 and S-4 can be detected by Py-FTIR, in contrast with the NH₃-TPD results. Similar results were reported by Campo et al^[35] for SAPO-34 and H-ZSM-5. These results suggest that the H-ZSM-5 pore openings and channels are modified and most of the surfaces are covered by the microcrystalline SiO₂.

  Figure 4

  

  Figure 4.  Py-FTIR spectra of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片

  2.2   Effect of F^-/Al₂O₃ ratio on the catalytic performance of H-ZSM-5 zeolite

  Figure 5(a) shows the methanol conversion versus the time on stream for MTP over various H-ZSM-5 zeolites. With the exception of S-4, all other H-ZSM-5 catalysts show complete methanol conversion at steady period (from 4 to 36 h), which remains above 96% after 48 h. After that, the methanol conversion decreases gradually with the time on stream, which becomes increasingly pronounced, following the order of S-2, S-1, S-3, and S-4. In contrast, the methanol conversion over S-4 drops to 95% at 36 h, which is attributed to the synergic effect of the acidity and textural properties of the catalysts. S-1 and S-2 have similar textural properties, but S-2 has weaker and fewer strong acid sites than S-1, which may reduce the hydrogen transfer and cyclization reactions and alleviate the coke formation in S-2^[11]. As a result, S-2 is catalytically more stable than S-1. Although S-3 and S-4 have fewer strong acid sites than S-2 and S-1, the modification by microcrystalline SiO₂ hinders the diffusion of methanol and product in the zeolite channels; S-3 and S-4 are then probably deactivated due to carbonaceous deposition just like the small pore molecular sieves such as SAPO-34.

  Figure 5

  

  Figure 5.  Methanol conversion (a), selectivity to ethene (b), selectivity to propene (c), and the P/E ratio (d) for MTP over various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片

  Figure 5(b) and 5(c) indicate the selectivity to ethene and propene versus the time on stream, respectively. The selectivity to propene is divided into two groups; over S-1 and S-2 it increases gradually, whereas over S-3 and S-4 it decreases with the time on stream. Nevertheless, at the same time, the selectivity to propene over S-3 and S-4 (> 45%) is far higher than that over S-1 and S-2 (< 40%), although the methanol conversion over the former catalysts is somewhat lower. There are two main possible reasons for the high selectivity to propene over S-3 and S-4. On one hand, they have fewer strong acid sites; as higher alkenes are much more active than ethene and propene^[36], the side-reactions of propene is then suppressed. It was also reported that the propene conversion over H-ZSM-5 is reduced with the decrease of strong acid sites^[37]. On the other hand, polymethylbenzene as the "carbon hydrocarbon pool" active centers can be cracked to ethene and propene^[38]. Modification of the pore openings and channels in H-ZSM-5 by microcrystalline SiO₂ improves the cracking activity of polymethylbenzene (as illustrated in Figure 6) and/or the alkenes methylation/cracking (decreasing the selectivity to ethene), leading to a further increase in the selectivity to propene.

  Figure 6

  

  Figure 6.  Possible reaction pathways from mesitylene to ethene and propene

  下载: 全尺寸图片 幻灯片

  The selectivity to ethene over S-1 and S-2 is much higher than that over S-3 or S-4, which is gradually decreased with the time on stream. Figure 5(d) further illustrates that the P/E ratio of all catalysts increases with the time on stream.

  It is now well-accepted that the conversion of methanol to olefins (MTO) follows the "carbon-hydrocarbon pool" mechanism^{[35, 39]}; the actual active hydrocarbon pool species are dependent on the topology, textural properties, and acid strength and density of the molecular sieves^{[38, 40-50]}. Over the H-ZSM-5 catalysts, ethene and propene were formed from polymethylbenzenes such as dimethylbenzene, trimethylbenzene and tetra-methylbenzene^{[49, 51]} (aromatics-based cycles), whereas higher alkenes (including propene) were formed by alkenes methylation/cracking reactions (alkenes based cycles)^{[36, 49, 52]}. Dimethylbenzene or trimethylbenzene should be more likely to accumulate within the channels of S-3 and S-4, according to aromatics-based cycles, with higher selectivity to ethene than that over S-1 and S-2. Moreover, as the reaction proceeds, more and more cokes should also be deposited near the pore openings and in their channels, prolonging the retention time of dimethylbenzene and trimethylbenzene within the channels and thus increasing the selectivity to ethene. However, our experimental results turn out to be quite opposite to these predictions.

  For the "carbon hydrocarbon pool" mechanism over large pore molecular sieves, the "ring expansion-contraction pairing" route was predominantly supported by experiment and theoretical calculation^[42]. First, the carbocations such as 1, 3-dimethylcyclopentenyl cation, 1, 2, 3-trimethylcyclopentenyl cation, 1, 3, 4-trimethyl cyclopentenyl cation and pentamethylbenzenium ion have been detected by ¹³C magic-angle spinning nuclear magnetic resonance (MAS NMR)^{[50, 53]} and Raman spectroscopy^[54], which were the active hydrocarbon pool species (active intermediates or transients) for the formation of light olefins. Based on the theoretical calculation, the deprotonation of 1, 3-dimethylcyclopentenyl cation produced 1, 3-dimethyl-1, 3-cyclopentadiene^[53], for which the activation energy barrier was only 2.2 kcal/mol higher than that for the parent ions^[50]. The activation energy barrier of 1, 3-dimethylcyclopentenyl cation reacting with trimethyl oxonium ion (forming dimethyl ether and 1, 3, 4-trimethyl cyclopentenyl cation) was only 33 kcal/mol, far below that needed for ethene or propene methylation. The methylation of dimethyl cyclopentadiene was much easier than that of ethene, propene, or polymethylbenzene. Second, as found by Cui et al^[55], toluene could react with deuterium-methoxy, forming deuterated aromatics. The possible mechanism might be exactly the "ring expansion-contraction pairing". Third, the 1, 3-dimethylcyclopentenyl cation could undergo a series of reactions, forming toluene^[56].

  We then speculate that the MTP or MTO reactions may follow the "methylation-based ring expansion-contraction pairing" mechanism, as illustrated in Figure 6, with mesitylene as the active centers that form ethene and propene. The difference between methanol-to-propene and methanol-to-ethene lies in the position of mesitylene methylation (C or G) and corresponding transition states (3 or 5), which depends on the pore channels and acidity of H-ZSM-5. For H-ZSM-5 with large pores, the methylation occurs probably at the ortho position of methyl (decided by orientation effects of replacement base on aromatic hydrocarbon), thereby forming the 1, 2, 3, 5-tetramethyl arenium-ion(4). In small pores, the 2, 2, 4, 6-tetramethyl arenium-ion(2) are formed instead due to the steric hindrance. Subsequently, through contraction and elimination reactions (D or H), small molecules such as propene or ethene and the corresponding intermediate (transient) cations such as 1, 3-dimethylcyclopentenyl cation (3) or 1, 3, 4-trimethyl cyclopentenyl cation (5) are formed. Methyl cyclopentenyl cations (3 or 5), through ring expansion and hydride transfer (E or I), then produce toluene or xylene. The methylation and isomerization of toluene and xylene create the active centers of mesitylene, thereby completing the reaction cycle.

  3.   Conclusions

  A series of nanosized SiO₂-ZSM-5 zeolites were synthesized in the media of F^--OH^- and the effect of F^-/Al₂O₃ molar ratio on their catalytic performance in MTP was investigated. The results indicate that an increase in the F^-/Al₂O₃ molar ratio of the synthesis mixture leads to an increase in the surface content of microcrystalline SiO₂, accompanying with a decrease in the relative crystallinity, surface area, pore volume, and acid strength and density. With a F^-/Al₂O₃ molar ratio of 12, the SiO₂-ZSM-5 zeolite exhibits the best catalytic performance in MTP, with a selectivity of 45% to propene and a propene/ethene (P/E) ratio of greater than 10.

  On the basis of the "dual-acid-site active center" model and "methylation-based ring expansion-contraction pairing" mechanism, it was demonstrated that the transition state shape selectivity plays an important role in determining the product selectivity in MTP. The high quality MTP catalyst has two characteristics: appropriate acidity and narrow pore channels with large openings.

• 1. [1]

     AGUDAMU , SUN Y, ZHANG F. Product distribution of methanol to propene reaction found in Shenhua Ningxia coal group[J]. Coal Chem Ind (in Chinese), 2013, 164(1):  58-60.

  2. [2]

     Successful testing of manufacturing olefins from methanol in fluidized bed at Huaihua group[J]. China Pet Process Petrochem Technol, 2009, 11(4): 61-61.

  3. [3]

     ZHONG L, YU F, AN Y, ZHAO Y, SUN Y, LI Z, LIN T, LIN Y, QI X, DAI Y, GU L, HU J, JIN S, SHEN Q, WANG H. Cobalt carbide nanoprisms for direct production of lower olefins from syngas[J]. Nature, 2016, 538(5623):  84-87.

  4. [4]

     JIAO F, LI J, PAN X, XIAO J, LI H, MA H, WEI M, PAN Y, ZHOU Z, LI M, MIAO S, LI J, ZHU Y, XIAO D, HE T, YANG J, QI F, FU Q, BAO X. Selective conversion of syngas to light olefins[J]. Science, 2016, 351(6277):  1065-1068. doi: 10.1126/science.aaf1835

  5. [5]

     YANG Y, SUN C, DU J, YUE Y, HUA W, ZHANG C, SHEN W, XU H. The synthesis of endurable B-Al-ZSM-5 catalysts with tunable acidity for methanol to propene reaction[J]. Catal Commun, 2012, 24(26):  44-47.

  6. [6]

     LIANG T, CHEN J, QIN Z, LI J, WANG P, WANG S, WANG G, DONG M, FAN W, WANG J. Conversion of methanol to olefins over H-ZSM-5 Zeolite:Reaction pathway is related to the framework aluminum siting[J]. ACS Catal, 2016, 6(11):  7311-7325. doi: 10.1021/acscatal.6b01771

  7. [7]

     SKLENAK S, DEDECEK J, LI C, WICHTERLOVA B, GABOVA V, SIERKA M, SAUER J. Aluminum siting in silicon-rich zeolite frameworks:A combined high-resolution 27Al NMR spectroscopy and quantum mechanics/molecular mechanics study of ZSM-5[J]. Angew Chem Int Ed, 2007, 46(38):  7286-7289. doi: 10.1002/(ISSN)1521-3773

  8. [8]

     KHARE R, LIU Z, HAN Y, BHAN A. A mechanistic basis for the effect of aluminum content on ethene selectivity in methanol-to-hydrocarbons conversion on H-ZSM-5[J]. J Catal, 2017, 348(4):  300-305.

  9. [9]

     ZHANG H, NING Z, LIU H, HAN S, TAO X, SEHN L, JIANG Y, GUO Y. Effect of silica sources on the properties of ZSM-5 and their catalytic performance for methanol conversion to propene[J]. Acta Pet Sin(Pet Process Sect), 2017, 33(4):  724-729.

  10. [10]

      ZHANG H, NING Z, LIU H, HAN S, TAO X, SEHN L, JIANG Y, GUO Y, DOU T. Effect of SiO₂/Al₂O₃ ratio on the properties of ZSM-5 and its catalytic performance for methanol to propene[J]. Mod Chem Ind, 2017, 2017(6):  79-83.

  11. [11]

      LIU J, ZHANG C, SHEN Z, HUA W, TANG Y, SHEN W, YUE Y, XU H. Methanol to propene:Effect of phosphorus on a high silica H-ZSM-5 catalyst[J]. Catal Commun, 2009, 10(11):  1506-1509. doi: 10.1016/j.catcom.2009.04.004

  12. [12]

      WEN P, MEI C, LIU H, YANG W, CHEN Q. Influence of methanol partial pressure and ZSM-5 particl size on distribution of products for methanol conversion to propene[J]. Chem React Eng Technol, 2007, 23(6):  481-486.

  13. [13]

      ZHAO T S, TAKEMOTO T, TSUBAKI N. Direct synthesis of propene and light olefins from dimethyl ether catalyzed by modified H-ZSM-5[J]. Catal Commun, 2006, 7(9):  647-650. doi: 10.1016/j.catcom.2005.11.009

  14. [14]

      ZHAO T S, TAKEMOTO T, YONEYAMA Y, TSUBAKI N. Selective conversion of dimethyl ether to propene and light olefins over modified H-ZSM-5[J]. Chem Lett, 2005, 34(7):  970-971. doi: 10.1246/cl.2005.970

  15. [15]

      FIROOZI M, BAGHALHA M, ASADI M. The effect of micro and nano particle sizes of H-ZSM-5 on the selectivity of MTP reaction[J]. Catal Commun, 2009, 10(12):  1582-1585. doi: 10.1016/j.catcom.2009.04.021

  16. [16]

      MEI C, WEN P, LIU Z, LIU H, WANG Y, YANG W, XIE Z, HUA W, GAO Z. Selective production of propene from methanol:Mesoporosity development in high silica H-ZSM-5[J]. J Catal, 2008, 258(1):  243-249. doi: 10.1016/j.jcat.2008.06.019

  17. [17]

      ZHANG H R, NING Z X, LIU H Y, SHANG J P, HAN S H, JIANG D D, JIANG Y, GUO Y. Bi₂O₃ modification of H-ZSM-5 for methanol-to-propene conversion:Evidence of olefin-based cycle[J]. RSC Adv, 2017, 7(27):  16602-16607. doi: 10.1039/C6RA27849C

  18. [18]

      ZHANG H R, NING Z X, SHANG J P, LIU H Y, HAN S H, QU W S, JIANG Y, GUO Y. A durable and highly selective PbO/H-ZSM-5 catalyst for methanol to propene (MTP) conversion[J]. Microporous Mesoporous Mater, 2017, 248(11):  173-178.

  19. [19]

      EGEBLAD K, KUSTOVA M, ZHU K, CHRISTENSEN C H. Mesoporous zeolite and zeotype single crystals synthesized in fluoride media[J]. Microporous Mesoporous Mater, 2007, 101(1):  214-223.

  20. [20]

      LOUIS B, KIWI-MINSKER L. Synthesis of ZSM-5 zeolite in fluoride media:An innovative approach to tailor both crystal size and acidity[J]. Microporous Mesoporous Mater, 2004, 74(1):  171-178.

  21. [21]

      CAULLET P, PAILLAUD J L, SIMON-MASSERON A, SOULARD M, PATARIN J. The fluoride route:A strategy to crystalline porous materials[J]. Comptes Rendus Chim, 2005, 8(3):  245-266.

  22. [22]

      PINAR A B, MÁRQUEZ-ÁLVAREZ C, GRANDE-CASAS M, PÉREZ-PARIENTE J. Template-controlled acidity and catalytic activity of ferrierite crystals[J]. J Catal, 2009, 263(2):  258-265. doi: 10.1016/j.jcat.2009.02.017

  23. [23]

      KESSLER H, PATARIN J, SCHOTT-DARIE C. The opportunities of the fluoride route in the synthesis of microporous materials[J]. Stud Surf Sci Catal, 1995, 26(14):  75-113.

  24. [24]

      AIELLO R, CREA F, NIGRO E, TESTA F, MOSTOWICZ R, FONSECA A, NAGY J B. The influence of alkali cations on the synthesis of ZSM-5 in fluoride medium[J]. Microporous Mesoporous Mater, 1999, 28(2):  241-259. doi: 10.1016/S1387-1811(98)00241-8

  25. [25]

      KONINGSVELD H V, JANSEN J C, BEKKUM H V. The orthorhombic/monoclinic transition in single crystals of zeolite ZSM-5[J]. Zeolites, 1987, 7(6):  564-568. doi: 10.1016/0144-2449(87)90099-6

  26. [26]

      SUN Y J, FU Y B, CHEN Q, ZHANG C M, SANG L J, ZHANG Y F. Silicon dioxide coating deposited by PDPs on PET films and influence on oxygen transmission rate[J]. Chin Phys Lett, 2008, 25(5):  1753-1756. doi: 10.1088/0256-307X/25/5/063

  27. [27]

      IVANOVA S, LEBRUN C, VANHAECKE E, PHAM-HUU C, LOUIS B. Influence of the zeolite synthesis route on its catalytic properties in the methanol to olefin reaction[J]. J Catal, 2009, 265(1):  1-7. doi: 10.1016/j.jcat.2009.03.016

  28. [28]

      IVANOVA S, LOUIS B, LEDOUX M J, PHAM-HUU C. Autoassembly of nanofibrous zeolite crystals via silicon carbide substrate self-transformation[J]. J Am Chem Soc, 2007, 129(11):  3383-3391. doi: 10.1021/ja0686209

  29. [29]

      BARRETT P A, CAMBLOR M A, CORMA A, JONES R H, VILLAESCUSA L A. Synthesis and structure of as-prepared ITQ-4, A large pore pure silica zeolite:The role and location of fluoride snions and organic cations[J]. J Phys Chem B, 1998, 102(21):  4147-4155. doi: 10.1021/jp980735e

  30. [30]

      AXON S A, KLINOWSKI J. Nuclear magnetic resonance studies of the synthesis of silicalite by the "fluoride method"[J]. J Chem Soc Fara Trans, 1993, 89(23):  4245-4248. doi: 10.1039/FT9938904245

  31. [31]

      GUTH J L, KESSLER H, WEY R. New route to pentasil-type zeolites using a non alkaline medium in the presence of fluoride Ions[J]. Stud Surf Sci Catal, 1986, 28:  121-128. doi: 10.1016/S0167-2991(09)60864-8

  32. [32]

      JI W P, LEE J Y, KIM K S, HONG S B, SEO G. Effects of cage shape and size of 8-membered ring molecular sieves on their deactivation in methanol-to-olefin (MTO) reactions[J]. Appl Catal A:Gen, 2008, 339(1):  36-44. doi: 10.1016/j.apcata.2008.01.005

  33. [33]

      NISHI K, YOKOMORI Y, KAMIYA N. Single-crystal structure of a pyridine sorption complex of zeolite H-ZSM-5(H-MFI)[J]. Microporous Mesoporous Mater, 2007, 101(1/2):  83-89.

  34. [34]

      ZHANG L, GAO J, HU J, LI W, WANG J. Lanthanum oxides-improved catalytic performance of ZSM-5 in toluene alkylation with methanol[J]. Catal Lett, 2009, 130(3/4):  355-361.

  35. [35]

      CAMPO A E S, GAYUBO A G, AGUAYO A T, TARRÍO A, BILBAO J. Acidity, surface species, and mechanism of methanol transformation into olefins on a SAPO-34[J]. Ind Eng Chem Res, 1998, 37(6):  2336-2340. doi: 10.1021/ie970748z

  36. [36]

      WU W, GUO W, XIAO W, LUO M. Dominant reaction pathway for methanol conversion to propene over high silicon H-ZSM-5[J]. Chem Eng Sci, 2011, 66(20):  4722-4732. doi: 10.1016/j.ces.2011.06.036

  37. [37]

      BORGES P, PINTO R R, LEMOS M A N D A, LEMOS F, VÉDRINE J C, DEROUANE E G, RIBEIRO F R. Light olefin transformation over ZSM-5 zeolites:A kinetic model for olefin consumption[J]. Appl Catal A:Gen, 2007, 324(1):  20-29.

  38. [38]

      WANG C M, WANG Y D, XIE Z K, LIU Z P. Methanol to olefin conversion on HSAPO-34 zeolite from periodic density functional theory calculations:A complete cycle of side chain hydrocarbon pool mechanism[J]. J Phys Chem C, 2009, 113(11):  4584-4591. doi: 10.1021/jp810350x

  39. [39]

      SVELLE S, JOENSEN F, NERLOV J, OLSBYE U, LILLERUD K-P, KOLBOE S, BJØRGEN M. Conversion of methanol into hydrocarbons over zeolite H-ZSM-5:Ethene formation is mechanistically separated from the formation of higher alkenes[J]. J Am Chem Soc, 2006, 128(46):  14770-14771. doi: 10.1021/ja065810a

  40. [40]

      BJØRGEN M, BONINO F, KOLBOE S, LILLERUD K, ZECCHINA A, BORDIGA S. Spectroscopic evidence for a persistent benzenium cation in zeolite H-Beta[J]. J Am Chem Soc, 2003, 125(51):  15863-15868. doi: 10.1021/ja037073d

  41. [41]

      BJØRGEN M, OLSBYE U, SVELLE S, KOLBOE S. Conversion of methanol to hydrocarbons:The reactions of the heptamethylbenzenium cation over zeolite H-Beta[J]. Catal Lett, 2004, 93(1/2):  37-40. doi: 10.1023/B:CATL.0000016945.28495.f0

  42. [42]

      BJØRGEN M, OLSBYE U, PETERSEN D, KOLBOE S. The methanol-to-hydrocarbons reaction:Insight into the reaction mechanism from[¹²C]benzene and[¹³C]methanol coreactions over zeolite H-beta[J]. J Catal, 2004, 221(1):  1-10. doi: 10.1016/S0021-9517(03)00284-7

  43. [43]

      BJORGEN M, OLSBYE U, KOLBOE S. Coke precursor formation and zeolite deactivation:Mechanistic insights from hexamethylbenzene conversion[J]. J Catal, 2003, 215(1):  30-44. doi: 10.1016/S0021-9517(02)00050-7

  44. [44]

      SASSI A, WILDMAN M A, AHN H J, PRASAD P, HAW J F. Methylbenzene chemistry on zeolite HBeta:Multiple insights into methanol-to-olefin catalysis[J]. J Phys Chem B, 2002, 106(9):  2294-2303. doi: 10.1021/jp013392k

  45. [45]

      ARSTAD B, STEIN K. The reactivity of molecules trapped within the SAPO-34 cavities in the methanol-to-hydrocarbons reaction[J]. J Am Chem Soc, 2001, 123(33):  8137-8138. doi: 10.1021/ja010668t

  46. [46]

      ARSTAD B, KOLBOE S. Methanol-to-hydrocarbons reaction over SAPO-34 molecules confined in the catalyst cavities at short time on stream[J]. Catal Lett, 2001, 71(3/4):  209-212. doi: 10.1023/A:1009034600533

  47. [47]

      SONG W, HAW J F, HENEGHAN C S. Methylbenzenes are the organic reaction centers for methanol-to-olefin catalysis on HSAPO-34[J]. J Am Chem Soc, 2000, 122(43):  10726-10727. doi: 10.1021/ja002195g

  48. [48]

      SONG W, NICHOLAS J B, HAW J F. Acid-base chemistry of a carbenium ion in a zeolite under equilibrium conditions:Verification of a theoretical explanation of carbenium ion stability[J]. J Am Chem Soc, 2001, 123(1):  121-129. doi: 10.1021/ja002775d

  49. [49]

      BJORGEN M, SVELLE S, JOENSEN F, NERLOV J, KOLBOE S, BONINO F, PALUMBO L, BORDIGA S, OLSBYE U. Conversion of methanol to hydrocarbons over zeolite H-ZSM-5:On the origin of the olefinic species[J]. J Catal, 2007, 249(2):  195-207. doi: 10.1016/j.jcat.2007.04.006

  50. [50]

      GOGUEN P W, TENG X, BARICH D H, SKLOSS T W, SONG W, WANG Z, NICHOLAS J B, HAW J F. Pulse-quench catalytic reactor studies reveal a carbon-pool mechanism in methanol-to-gasoline chemistry on zeolite H-ZSM-5[J]. J Am Chem Soc, 1998, 120(11):  2650-2651. doi: 10.1021/ja973920z

  51. [51]

      SVELLE S, JOENSEN F, NERLOV J, OLSBYE U, LILLERUDK K P, KOLBOE S, BJØRGEN M. Conversion of methanol into hydrocarbons over zeolite H-ZSM-5:Ethene formation is mechanistically separated from the formation of higher alkenes[J]. J Am Chem Soc, 2006, 128(46):  14770-14771. doi: 10.1021/ja065810a

  52. [52]

      HUANG X, AIHEMAITIJIANG D, XIAO W D. Co-reaction of methanol and olefins on the high silicon H-ZSM-5 catalyst:A kinetic study[J]. Chem Eng J, 2016, 286(2):  150-164.

  53. [53]

      WANG C, CHU Y, ZHENG A, XU J, WANG Q, GAO P, QI G, GONG Y, DENG F. New insight into the hydrocarbon-pool chemistry of the methanol-to-olefins conversion over zeolite H-ZSM-5 from GC-MS, solid-state NMR spectroscopy, and DFT calculations[J]. Chem, 2014, 20(39):  12432-12443. doi: 10.1002/chem.v20.39

  54. [54]

      CHUAY T, CSTAIR P, NICHOLASJ B, SONGW , HAWJ F. UV Raman spectrum of 1, 3-dimethylcyclopentenyl cation adsorbed in zeolite H-MFI[J]. J Am Chem Soc, 2003, 125(4):  866-867. doi: 10.1021/ja028439+

  55. [55]

      CUI Z M, LIU Q, SONG W G, WAN L J. Insights into the mechanism of methanol-to-olefin conversion at zeolites with systematically selected framework structures[J]. Angew Chem Int Ed, 2006, 45(39):  6512-6515. doi: 10.1002/(ISSN)1521-3773

  56. [56]

      ILIAS S, BHAN A. Mechanism of the catalytic conversion of methanol to hydrocarbons[J]. ACS Catal, 2013, 44(3):  18-31.

• 

  Figure 1  XRD patterns of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片
  

  Figure 2  SEM images of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片
  

  Figure 3  NH₃-TPD profiles of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Py-FTIR spectra of various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片
  

  Figure 5  Methanol conversion (a), selectivity to ethene (b), selectivity to propene (c), and the P/E ratio (d) for MTP over various H-ZSM-5 zeolites: S-1, S-2, S-3, and S-4

  下载: 全尺寸图片 幻灯片
  

  Figure 6  Possible reaction pathways from mesitylene to ethene and propene

  下载: 全尺寸图片 幻灯片

  Table 1.  Relative crystallinity, SiO₂/Al₂O₃ ratio, and textural properties of H-ZSM-5 prepared with different F^-/Al₂O₃ ratios

  Catalyst	Relative crystallinitya/%	SiO2/Al2O3b	Surface area c A/(m2·g-1)				Pore volume v/(cm3·g-1)
  			total	micro	meso		totalf	microg
  S-1	100	285	388.0	356.0	32.0		0.2422	0.1774
  S-2	64	295	386.7	348.7	37.9		0.2277	0.1740
  S-3	63	359	154.8	126.8	27.9		0.1238	0.0657
  S-4	23	395	109.7	92.6	17.1		0.0797	0.0476
  note: a: the relative crystallinity of ZSM-5 was calculated based on the intensity of the diffraction peaks at 2θ of 22°-25°, using S-1 as the reference; b: the SiO2/Al2O3 ratio was determined by ICP-AES; c: total BET surface area was obtained by the BET method using adsorption data in p/p0 ranging from 0.05 to 0.25, whereas micro and meso-surface areas were determined by the t-plot method; d: the total pore volume was estimated from the adsorbed amount at p/p0 = 0.99, whereas the micro pore volume was determined by the t-plot method

  下载: 导出CSV
•