Chinese Chemical Letters  2021, Vol. 32 Issue (1): 328-338   PDF    
Recent advance in synthesis and application of heteroatom zeolites
Tingting Panga,b,1, Xuanyu Yangb,1, Chenyi Yuanb, Ahmed A. Elzatahryc, Abdulaziz Alghamdid, Xing Hea, Xiaowei Chengb,*, Yonghui Dengb,e     
a School of Materials Science and Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China;
b Department of Chemistry, Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials, Fudan University, Shanghai 200433, China;
c Materials Science and Technology Program, College of Arts and Sciences, Qatar University, P.O. Box 2713, Doha, Qatar;
d Department of Chemistry, College of Science, King Saud University, P.O. Box 2455, Riyadh 11451, Saudi Arabia;
e State Key Lab of Transducer Technology, Shanghai Institute of Microsystem and Information Technology, Chinese Academy of Sciences, Shanghai 200050, China
Abstract: Incorporation of heteroatoms into the framework of zeolites has become a significant strategy to improve their performance in catalysis and adsorption, because the obtained heteroatom zeolites exhibit quite different properties from the conventional aluminosilicate zeolites in aspects of surface acidity, pore structures, particle size and so on. In this review, the progress on the heteroatom zeolites including their synthesis and application is highlighted. First, the recent advance on the design and synthesis of different heteroatom zeolites is summarized. Special emphasis is placed on the introduction and comparison of three typical methods, including the direct synthesis, post synthesis and improved direct synthesis, for the traditional heteroatom zeolites (such as TS-1, Sn-MFI, Sn-β) and newly-reported heteroatom zeolites (such as W-MFI, Mo-MFI). According to their intrinsic characteristics, the application of heteroatom zeolites in diverse fields, such as production of fine chemicals, air pollution control and biomass conversion is then discussed. Finally, the challenges and perspective on the future development of heteroatom zeolites in low-cost preparation and practical application are proposed.
Keywords: Heteroatom zeolite    Incorporation    Tetrahedral coordination    Research status    Synthesis    Application    Prospective    
1. Introduction

Zeolite is one of the most representative molecular sieve materials, which possess intrinsic characteristics including regular pore structure, high specific surface area, and ability of sieving substances at the molecular level. Stilbite (STI) zeolite which showed the boiling property in liquid, was discovered as the first natural zeolite by Cronstedt in 1756 [1], and until now more than 80 natural zeolites have been reported, such as analcite (ANA), chabazite (CHA), mordenite (MOR) and so on. Therefore, zeolite was mainly defined as a class of microporous aluminosilicate materials at first, containing three-dimensional regular micropores (pore size < 2 nm) and highly-crystallized framework. The zeolite framework is usually composed of TO4 tetrahedrons as the primary structural units, which are further connected with other four TO4 tetrahedrons through common oxygen bridges. The framework T atoms in the traditional zeolites generally contain two elements of silicon (Si) and aluminum (Al), and they connect with oxygen (O) atoms through sp3 orbit hybridization to form Si-O bonds (1.61 Å) and Al-O bonds (1.75 Å), respectively [2-4]. Through end-to-end connections of TO4 tetrahedrons, polycyclic rings are formed as the secondary structural units, which connect with each other in three dimensions to construct the zeolite cages. By further interconnection and arrangement of the cages in certain rules, a crystalline zeolite with 3-dimentional framework is established, showing specific topology structure, regular microporous channels/cavities and surface acidic properties [5]. The zeolites with the same topology are given one code by International Zeolite Association (IZA), and to date there have already been 248 framework type codes assigned for natural and artificial zeolites. Several artificial aluminosilicate zeolites, such as ZSM-5 (MFI), Y (FAU), β (BEA), mordenite (MOR), can be synthesized in low cost by a facile hydrothermal method, which are widely used in acid/shape selective catalysis, gas adsorption, ion exchange and so on.

Quite different from the amorphous walls of mesoporous silica materials, the highly-crystallized frameworks endow the aluminosilicate zeolites with excellent thermal and hydrothermal stability. The arrangement of AlO4 tetrahedrons strictly follows the Lowenstein rule, meaning that there are no Al-O-Al connections in zeolite framework. Thus, it means that Al atoms can only connect with four Si atoms through O bridges, and fourcoordinated aluminum makes the zeolite framework electronegative, requiring the equivalent numbers of positively charged ions out of the framework. The Brönsted acid sites are introduced onto the zeolite when hydrogen ions are exchanged on the framework as the equilibrium cations. The H-type zeolites can be extensively used as the solid acid catalysts in various acid-catalyzed reactions [6, 7], mainly attributed to their intrinsic properties of high framework stability, regular micropore structures, large specific surface area, and tunable surface acidity. However, the application of conventional aluminosilicate zeolites is greatly limited by their unchangeable framework components of Si and Al elements, leading to insufficient active sites (especially lack of Lewis acid sites) and sometimes low catalytic activity and short lifetime. Therefore, in order to improve the performance of aluminosilicate zeolites, a great number of modification strategies have been tried to precisely adjust the micropore structures, surface acidic/hydrophobic properties, ion-exchanging ability, and so on. For a typical example, the catalytic properties are greatly changed when TiO4 tetrahedrons replace AlO4 tetrahedrons in ZSM-5 by isomorphous substitution to receive TS-1 zeolite (titanium silicate-1, MFI type), which is an important milestone in the history of heteroatom zeolites and performs as an efficient and stable catalyst for the epoxidation of propene with hydrogen peroxide [8]. The PO4 tetrahedrons can also enter the framework to connect with SiO4 and AlO4 tetrahedrons, resulting in the formation of silicon-aluminum-phosphorous zeolites (SAPO), such as SAPO-34 (CHA type), a commercial catalyst in methanol-to-olefin (MTO) reaction [9, 10]. Therefore, in past several years the synthesis of heteroatom zeolites by isomorphous substitution, containing different main group metals (As, Ga, Sn, Ge, etc.), non-metals (B, C, F, etc.), and transition metals (Ti, Fe, Zr, Cu, Mn, V, W, Mo, etc.) incorporated into the zeolite framework in form of TO4 tetrahedrons, has been considered as an attractive route for the improvement of catalysis efficiency and adsorption selectivity [11, 12]. The introduction of heteroatoms in framework shows a significant modulation on the micropore structure, surface acidity, and particle size, which greatly changes the physical and chemical properties, increases the desired catalytic active sites, and improves the catalytic performance [13, 14]. To date there have already been numerous reports about the heteroatom zeolites, which mainly focus on the investigation about the synthesis methods and application as catalysts or adsorbents. But to the best of our knowledge, among these publications few reviews have been found to systematically depict heteroatom zeolites according to their synthesis strategies, structural characteristics and proposed applications. Herein, we reviewed the recent advance of several important heteroatom zeolites including the research status of synthesis methods, newly-reported heteroatom zeolites, and their applications in production of fine chemicals, air pollution control, biomass conversion and so on. Based on the research status, we proposed the perspective and outlook about the future development of heteroatom zeolites.

2. Research status of synthesis methods for heteroatom zeolites 2.1. Direct synthesis

The direct synthesis is one of the most convenient and universal methods for the preparation of heteroatom zeolites, which is usually conducted in a traditional hydrothermal system. Heteroatoms are introduced into zeolites through directly adding metal salt solutions or metal organics during hydrothermal synthesis. The hydrolysis of the heteroatom precursors occurs simultaneously with that of the silicon/aluminum sources, forming a homogeneous sol under alkaline (or neutral) conditions, which is then in situ crystallized into the heteroatom zeolite by the hydrothermal treatment at a certain temperature. This method is conducive to the uniform insertion of several kinds of heteroatoms in the zeolite framework during crystallization, including main group elements (e.g., Ga, Ge, B) and transition metal elements (e.g., Ti, Fe, V, Cr, Mn) [11-13]. The heteroatom elements can partially substitute silicon (or aluminum) in framework, and connect with the primary SiO4 tetrahedrons to form the new heteroatom framework. Otherwise, since the hydrothermal synthesis of zeolite is usually carried out under alkaline conditions, transition metal hydroxide precipitates are easily formed, resulting in the inhomogeneous distribution of the heteroatom species on the surface or in the channels of zeolite [15], which will seriously block the micropores. In this section we mainly discuss the recentlyreported direct hydrothermal synthesis for heteroatom zeolites through framework isomorphous substitution, in which the heteroatom sources of nitrate, acetate, acetylacetone chelate and chloride are commonly used.

As early as 1991, Kosslick et al. firstly synthesized Ga-ZSM-5 zeolites as a recipe of Al-ZSM-5 through isomorphous substitution of silicon with gallium by the direct hydrothermal synthesis [16]. The gallium, located in the Group IIIA the same with aluminum, was proved to successfully enter the zeolite framework, and it could effectively reduce the decomposition temperature required for template removal. Recently, using the similar method, Jin et al. synthesized a series of heteroatom ZSM-5 zeolites (denoted as MAlMFI, M = Fe, Ga), inserting Ga or Fe species into Al-ZSM-5 [17], which greatly affected their acid properties and catalytic performance. The parent H-AlMFI showed the highest methanol conversion, but relatively poor selectivity to light olefins in MTO reaction. Additionally, the acid strength could be improved by incorporating both heteroatoms of Ga and Fe into the zeolite framework (FeGaAlMFI), resulting in much better reaction activity, increased yields of light olefins, and improved catalytic stability. The presence of Ga in framework (H-GaAlMFI) could accelerate the formation of C5+ and aromatic-rich heavy fractions. Moreover, the Fe incorporated Al-ZSM-5 (H-FeAlMFI) showed higher catalytic stability and better propylene selectivity than H-AlMFI and Fe3+-exchanged samples in MTO reaction. Hodoshima et al. also reported the synthesis of FeGaAl-ZSM-5 by simultaneously incorporating Fe and Ga heteroatoms into Al-ZSM-5 zeolite through the hydrothermal method, which was further evaluated as a catalyst in n-hexane cracking reaction [18]. Although the acid strength of zeolite was decreased by incorporation of iron atoms, the dehydrogenation activity for paraffinic hydrocarbons was improved by isomorphous substitution of gallium species in framework. The combination of Fe and Ga heteroatoms in zeolite framework, greatly changing the surface acidity, could prolong the lifetime of the catalyst to 80 h as well as increase the propylene yields (18–31 wt%) by suppressing the formation of aromatics under mild conditions.

As one of the most important commercial heteroatom zeolites, TS-1 (MFI type) has attracted a great interest for a long period since its discovery by Taramasso et al. in 1983 [19-22], which is highlighted as a mile stone in titanium-silicate zeolites. The Ti atoms were introduced in framework through the formation of peroxotitanate species, which can in situ substitute partial silicon atoms in tetrahedral coordination by a hydrothermal method (Fig. 1). The Ti-incorporated framework has the potential to receive electron pairs, resulting in the production of six-coordinated Ti, therefore, TS-1 zeolite exhibits special adsorption ability towards hydrogen peroxide, which can be activated on TS-1 and act as an oxidant in the selective catalytic oxidation of aliphatic and aromatic hydrocarbons [23]. Moreover, other titanium-silicate zeolites were also synthesized in subsequent reports, including Ti-silicalite-2 [24], Ti-β [25-30], Ti-MOR [31-34], etc. For example, by using a direct hydrothermal synthesis method containing F- medium as the mineralization agent, Ti heteroatom could be successfully incorporated into the framework of aluminum-free β zeolite [35], producing Ti-β zeolite with a high hydrophobicity and selective oxidation ability for organic substrates with H2O2. However, the incorporation of Ti into the framework is highly dependent on the pH values of the synthesis system. When pH value was close to 7.0, the amount of incorporated Ti showed an upper limit of about 2.3 Ti/unit cell, thus the excess Ti fell off the framework to form anatase phase. When pH value was around 11, the amount of incorporated Ti in framework was further increased without anatase formation, resulting in the increase of unit cell volume. Furthermore, since the Ti-β zeolite synthesized in F- medium showed high hydrophobicity, it could be used as the catalyst in the epoxidation of substrates containing a polar moiety, such as unsaturated fatty acids, esters and so on.

Download:
Fig. 1. Original synthesis procedures of TS-1 zeolite. Reproduced with permission [22]. Copyright 2017, Springer International Publishing.

Recently, Mintova et al. reported a completely new W-MFI heteroatom zeolite for the first time [36], which was synthesized through in situ introducing tungsten atoms into the framework of all-silica MFI zeolite by the one-pot hydrothermal method. The obtained W-MFI zeolite preserved the MFI topology structure and was composed of monodispersed single nanoparticles in size of 70–80 nm, showing relatively smooth surfaces compared to Si-MFI zeolite (Fig. 2). Since the newly-formed flexible W-O-Si bonds were more stable than the Si-O-Si bonds in all-silica MFI zeolite, they could effectively prevent the formation of silanol groups, resulting in a great improvement of surface hydrophobicity. In addition, the incorporation of W could also enhance the Lewis acid strength. Therefore, in comparison with W species loaded silicalite-1 and silica, the W-MFI zeolite displayed an excellent catalytic activity in styrene epoxidation (Fig. S1a in Supporting information). The W-MFI zeolite also showed high adsorption selectivity of CO2/CO and NO2/NO (Figs. S1b and c in Supporting information), which could be used as a promising adsorbent for CO2 and NO2 removal in engine exhaust and stationary gas. Following this work, Lv et al. incorporated W atoms into the framework of TS-1 zeolite by the one-pot direct hydrothermal synthesis, and successfully obtained the heteroatom WTS-1 zeolite [37]. Compared to W-MFI and TS-1 zeolites, the isomorphous substitution of both Ti and W atoms in framework endowed WTS-1 zeolite with stable Lewis acid sites and rough heterogeneous morphology, resulting in the improved catalytic efficiency and good reusability more than ten cycles in oxidative desulfurization of organosulfur compounds.

Download:
Fig. 2. HRTEM images of nanosized zeolites at different magnifications. (a–c) W-MFI nanosized zeolites. (d–f) Si-MFI nanosized zeolites. Reproduced with permission [36]. Copyright 2017, Springer Nature.

Quite recently, Mintova's group successfully introduced the well-dispersed molybdenum atoms into the MFI-type zeolite to obtain nanosized Mo-MFI heteroatom zeolite by the hydrothermal synthesis method [38]. Notably, the Mo atoms with the size of 0.05 nm could be uniformly distributed in framework, as shown in scanning transmission electron microscopy high-angle annular darkfield images (STEM-HAADF) (Fig. 3). The homogeneous incorporation of Mo atoms in framework could reduce the symmetry of the space groups of MFI structure from orthorhombic to monoclinic, and the unit cell volume increased accordingly. Since both Mo and W elements locate in Group VIB, the Mo-MFI zeolite has the similar properties with W-MFI. For example, the Mo insertion into the framework could also remarkably reduce the contents of silanol defects, resulting in high hydrophobicity of Mo-MFI zeolite. The Lewis acid sites were formed due to the existence of the Mo(-OSi)4 units by introducing atomically distributed Mo into SiO4 tetrahedrons of MFI framework.

Download:
Fig. 3. STEM-HAADF micrographs revealing the distribution of Mo species in (a) Mo-MFI-D and (b) Mo-MFI-P. Reproduced with permission [37]. Copyright 2019, American Chemical Society.

As one of the most common and efficient strategies for heteroatom zeolites, the direct hydrothermal synthesis method can be easily operated and highly repeated for the industrial scale-up production. The produced heteroatom zeolites generally show high crystallinity as well as controllable heteroatom content, topology structure, morphology, and particle size. However, due to the difference in the radii between transition metals and silicon/aluminum ions and their mismatching hydrolysis rates during the synthesis, there are still great difficulties and shortcomings in the direct synthesis method, such as the low content of heteroatoms incorporated into the zeolite framework in alkaline synthesis systems, long reaction time required for complete crystallization, easy formation of large crystalline domains and extra-framework heteroatoms, severe and narrow synthesis conditions, and sometimes occurring of metal agglomeration [39, 40]. Therefore, other methods, such as post-synthesis process have been desired for the synthesis of heteroatom zeolites [41].

2.2. Post synthesis

The post-synthesis method to obtain heteroatom zeolites is generally conducted through inserting heteroatoms into the framework of as-synthesized zeolites in liquid-solid, solid-solid, or gas-solid systems. This method is based on the framework dealumination or desilication combined with the isomorphous substitution of framework atoms by post treatment under certain conditions. Voids or defects are formed in the zeolite framework by removal of Si (or Al) species, which favor the incorporation of appropriate metal complexes into the zeolite framework [42-46]. In comparison with the direct synthesis method, the post-synthesis method is more efficient for doping transition metal heteroatoms with a much larger radius than Si or Al atoms (such as Ti, Sn, Zr) into the zeolite framework, resulting in the heteroatom zeolites with a high metal loading without the formation of extra-framework bulk metal oxides.

Rigutto et al. reported a new post-treatment method to synthesize Ti-β zeolite by gas-solid isomorphous substitution, which was then used as an efficient catalyst in the 1-hexene epoxidation reaction [47]. In this method, partial boron atoms were firstly removed from the framework of B-β zeolite by dilute acid treatment for generation of lattice vacancies in the framework. Then the deboronated H-B-β zeolite was treated in titanium chloride vapor at 300 ℃ combined with an efficient methanolysis treatment, in order to replace the boron atoms as well as insert Ti heteroatoms at tetrahedral sites into the zeolite framework. The Ti-β zeolite prepared by this method was stable and contained almost no extra-framework titanium, exhibiting high activity and selectivity during the reaction of alkenes epoxidation with hydrogen peroxide.

By using a two-step post-synthesis method, Tang et al. successfully incorporated Ti atoms into the framework of β zeolite to prepare the Ti-β zeolite (Fig. 4), which was used as the catalyst in the epoxidation of unsaturated ketones [48]. Firstly, the commercial H-β zeolite (Si/Al = 13.5) was dealuminated in concentrated nitric acid solution to receive Si-β (Si/Al > 1800), leading to the introduction of vacant T sites with associated silanol groups. Secondly, the dry impregnation method was utilized through mechanically grinding Si-β and organometallic precursor of titanocene dichloride (Cp2TiCl2), and the physical mixture was further calcinated at 800 ℃. Most of the incorporated Ti species were proved to enter the zeolite framework in the form of isolated tetrahedrally coordinated Ti(IV), whose number showed a quasi-linear correlation with the epoxidation rate of 2-cyclohexen-1-one using hydrogen peroxide as an oxidant. Additionally, the obtained Ti-β zeolite exhibited high epoxidation activity and selectivity toward epoxide in comparison with TS-1 and Ti-MCM-41 materials (Fig. S2 in Supporting information).

Download:
Fig. 4. Schematic representation of the incorporation of tetrahedrally coordinated Ti(IV) species into β zeolite. Reproduced with permission [48]. Copyright 2017, The Royal Society of Chemistry

A facile TiCl4 vapor-treatment method was taken by Yoo et al. for the synthesis of Ti-ZSM-5 zeolite [49]. In this work no pretreatment was done to the raw ZSM-5 zeolite with different Si/Al ratios, and Ti atoms in tetrahedral coordination could be successfully incorporated into the framework of ZSM-5. Among the precursors used in the synthesis, such as TiCl4, ammonium titanyl oxalate and Cp2TiCl2, the successful insertion of Ti heteroatoms into the framework could be obtained only by using TiCl4 as the precursor, which did not affect the structure and surface properties of ZSM-5 zeolite. The content of incorporated Ti in framework was closely dependent on the SiO2/Al2O3 ratios of ZSM-5 zeolite, and hydrophobic zeolite facilitated the insertion of Ti at appropriate temperature ranging from 400 ℃ to 500 ℃. The Ti-ZSM-5 zeolite was used as a catalyst in cyclohexanone ammoxidation, exhibiting higher conversion and selectivity than TS-1, due to the exposing of tetrahedrally coordinated Ti atoms on zeolite surface as the active sites. Guo et al. also used the gas-solid isomorphous substitution method to introduce Ti atoms into the ZSM-5 framework by replacing B atoms in B-ZSM-5, and the obtained Ti-ZSM-5 zeolite showed low contents of extra-framework Ti species [50]. The properties of Ti-ZSM-5 zeolite presented a close relationship with the raw B-ZSM-5 precursors. The B-ZSM-5 synthesized at lower crystallization temperature with smaller size, could be transformed into smaller crystals of Ti-ZSM-5 zeolite, which exhibited better catalytic performance and recyclability in phenol hydroxylation due to its higher adsorption capacity, more accessible active sites and shorter diffusion length.

The post-treatment method of gas-solid isomorphous substitution is also applicable for the synthesis of Sn-β zeolite [41], which owns large pores and Lewis acidity, and has received much attention in biomass transformation by green oxidation processes. The highly dealuminated β zeolite was firstly prepared by the treatment of commercial nanosized β in HNO3 solution, resulting in a great number of hydroxyl defects occupying tetrahedrally coordinated sites. Then Sn heteroatoms reacted with the generated silanol groups in high-temperature SnCl4 vapor, and through an atom-planting process, the Sn atoms could be introduced into the zeolite framework to obtain the Sn-β zeolite. The Sn content in the nanosized Sn-β zeolite was as high as 6.2 wt%, which was much higher than that obtained by the direct synthesis. Furthermore, the nanosized Sn-β possessed shorter in-pore diffusion length than the conventional microsized one synthesized in fluoride system, therefore, the former exhibited the enhanced mass diffusion to bulky molecules and much higher catalytic activity in Baeyer-Villiger oxidation of 2-adamantanone in the presence of hydrogen peroxide.

Wolf et al. prepared a series of Lewis acidic heteroatom zeolites of Sn-β, Zr-β and Ti-β by a two-step post-treatment method [51]. Similar with the post-synthesis procedures described above, the commercial Al-β zeolite needed to be dealuminated to form a high-silica β zeolite as the silanol-riched precursor, which was then treated by the solid-solid or solid-liquid isomorphic exchange with different metal precursors. The heteroatoms of Sn, Zr or Ti could be incorporated into the zeolite framework with a high metal loading, resulting in the increase of Lewis acid centers. The heteroatom zeolites of Sn-β, Zr-β and Ti-β exhibited comparable catalytic activity and selectivity in the isomerization of glyceraldehyde to dihydroxyacetone and the epoxidation of bulky olefins with hydrogen peroxide. For example, Ti-β zeolite performed best in the cyclooctene epoxidation reaction, and Sn-β zeolite was the best catalyst in the reaction of glyceraldehyde isomerization.

In conclusion, the post-synthesis method owns several distinct advantages compared to the direct synthesis method, such as short synthesis time, small crystal size, high heteroatom content, improved catalytic efficiency, and so on. However, there are still some shortcomings for the post-synthesis method. For example, the dealuminated zeolites are usually used as the raw materials, which are obtained by deep dealumination in strong acid solution, leading to introduction of hydroxyl nests but the decrease of crystallinity and collapse of zeolite framework. Moreover, most of the isomorphous substitution of heteroatoms in zeolite framework should be conducted in vapor of metal precursors at high temperature, which seems to be dangerous, complicated and energy cost. The post-treatment process is more complicated and expensive than the direct hydrothermal synthesis, therefore, it is not a universal method to realize industrial production of heteroatom zeolites in low cost [52].

2.3. Improved direct synthesis

In view of some shortcomings of both the direct hydrothermal synthesis and the post-synthesis methods described above, many researchers have put their attention on developing more convenient and effective green synthesis routes. At present, several improved methods based on the direct synthesis have been reported for different heteroatom zeolites, such as the improved hydrothermal method using a special ionic complex as both template and metal source, the dry gel conversion [53-61].

Li et al. prepared heteroatom zeolites by one-step hydrothermal method through making an ion complex containing heteroatom source in advance, which was also used as a structure directing agent during the synthesis [53-57]. The content of heteroatoms could be increased to be as high as 15 wt% with uniform dispersion in zeolite framework. Since this one-step hydrothermal method was very simple and highly efficient for heteroatom insertion into framework, it could also be extended to the synthesis of other heteroatom zeolites. For example, EDTA was used as a complexing agent to prepare the ionic complex of [(C4H9)4N]2+[Ni(EDTA)]2-, which then performed as both the template and nickel source (Fig. 5). The heteroatom zeolite of Ni-MFI with high nickel content of 5–15 wt% in framework was successfully synthesized. The characterization results showed that the crystal structure and morphology of the typical MFI zeolite were well preserved through incorporating high content of Ni species into the framework. The prepared Ni-MFI zeolite showed good thermal and hydrothermal stability, enhanced acidity, increased specific surface area and pore volume. In addition, the average pore size increased to about 2.5– 3.9 nm, in the range of mesopores [53]. The method was also suitable for synthesizing other transition metal-rich M-MFI zeolites (M = Fe, Co, Ni, Cu) with high metal content (10–15 wt%) and uniform pores [54], showing high crystallinity, large surface area, increased pore volume and average pore size. Using the similar strategy, Li et al. synthesized a composite heteroatom molecular sieve of Ni-MFI/Co-MCM-41, which contained two metal heteroatoms by introduction of two ionic complexes as the structure directing agents [55]. Notably, a great number of heteroatoms could be effectively incorporated into the framework of molecular sieves through isomorphous substitution, in which the Ni content reached as high as 15 wt%. The composite molecular sieve showed both the typical MFI orthogonal crystal phase and the MCM-41 hexagonal phase with ordered mesopores. The specific surface area, pore volume and average pore size of Ni-MFI/Co-MCM-41 were greatly increased. Due to the existence of two kinds of pore structures and numerous transitional metals, the composite materials exhibited the promising catalytic performance in hydrocracking of residual oil, which could improve the yield of gasoline and reduce the coke production rate [55].

Download:
Fig. 5. Schematic of the synthetic mechanism of new template for synthesizing the mesoporous zeolite Ni-MFI. Reproduced with permission [53]. Copyright 2009, Elsevier Inc.

Wang et al. synthesized the Fe-MFI zeolite through an improved direct hydrothermal method by using potassium hexacyanoferrate [K3Fe(CN)6] as the iron source and tetrabutylammonium bromide (TBAB) as the template [56]. It was proved that Fe atoms existed in the form of Fe3+ bound to the tetrahedrally-coordinated sites in the zeolite framework, demonstrating super micropores, highly-crystallized MFI zeolite structure, large surface area and pore volume. The Fe-MFI zeolite was further utilized as the catalyst in phenol hydroxylation by hydrogen peroxide, displaying higher phenol conversion of 38.4% and better selectivity to hydroxybenzene of 84.2% than silicalite-1, an all-silica MFI zeolite. This result well indicated that the catalytic activity came from the tetrahedrally coordinated Fe incorporated in the zeolite framework [56]. Moreover, lanthanide ion-containing (Ln = La3+, Ce3+) micro/mesoporous composite molecular sieves of Ln-ZSM-5/MCM-41 were also reported by using the same strategy [57], in which kaolin was used as silica and alumina sources, hexadecyl trimethylam-monium bromide (CTMAB) and TBAB were used as the cotemplates. The results showed that the lanthanide ions were incorporated into the framework of ZSM-5 zeolite in tetrahedral coordination, introducing numerous Brönsted acid sites on Ln-ZSM-5. The heteroatom composite molecular sieves of Ln-ZSM-5/MCM-41 possessed large specific surface area (710 m2/g) and uniform mesoporous diameters (3.5–4.0 nm), which exhibited high catalytic activity toward the esterification reaction of acetic acid with n-butanol.

The steam-assisted conversion (SAC) method is highly efficient for the synthesis of heteroatom zeolites, in which the dry gel precursor can be transformed into porous crystalline zeolites under steam treatment at certain temperature. Compared to the conventional hydrothermal synthesis for zeolites, the SAC method possesses several distinct advantages, such as rapid crystallization under mild conditions of temperature and pH, less template required and less waste exhaust [58-60]. Recently, Kang et al. successfully prepared a highly-crystallized Sn-β zeolite via the SAC route from an amorphous dry stannosilicate gel (Fig. S3 in Supporting information) [61]. The crystallization of Sn-β zeolite was performed under mild conditions for very short time of 5 h. Under high conversion of Sn-containing dry gel, Sn atoms were completely incorporated into the framework after the crystallization of Sn-β zeolite, in which the SnO2 content was about 3.8 wt% (Si/Sn = 83). The Sn-β zeolite prepared by the SAC method showed uniformly dispersed tetrahedrally-coordinated Sn atoms in framework, high Sn-substitution content, crystallinity and yield, compared to the samples by the traditional hydrothermal method. The Sn-β zeolite exhibited promising catalytic activity, high selectivity and good recyclability in Baeyer-Villiger oxidation of cyclohexanone to ε-caprolactone. Therefore, the SAC method is highly efficient for the synthesis of Sn-β zeolite, which is considered as one facile and low-cost strategy for the scale-up production of heteroatom zeolites [61].

Quite recently, another novel improved direct synthesis method, named as the solvent-free route, was reported by Han et al. for the synthesis of Fe-ZSM-5 zeolite [62]. In this method, sodium silicate, fumed silica, TPABr, ammonium chloride and ferric nitrate were firstly added into a mortar for mechanical grinding and homogeneous mixing. Then, the precursor mixture was placed in a Teflon-lined autoclave and the crystallization was conducted at certain temperature for short time. The obtained Fe-ZSM-5 zeolite possessed a large specific surface area and pore volume, resulting in the increased adsorption capacity for phenol and excellent catalytic performance in the phenol hydroxylation reaction. The solvent-free method can effectively enhance the crystallization rate, shorten the crystallization time of Fe-ZSM-5 zeolite, and reduce the waste of solvent, which can be extended for the synthesis of other iron-based zeolite catalysts.

Miyake et al. reported another effective solid transformation method, named as dry gel conversion (DGC) to obtain Fe-MFI zeolite with small grain size by using a single template [63]. This synthesis method was quite similar with the described SAC route for the synthesis of Sn-β [61]. Firstly, the uniform sol containing precursors was mixed and stirred, and the solvent was then evaporated to obtain a xerogel. Secondly, the xerogel was ground and put into an autoclave with water in the bottom. Crystallization of xerogel into Fe-MFI zeolite was conducted under the promotion of high-temperature steam. The crystallization rate was slower than that of the traditional hydrothermal synthesis method, resulting in smaller crystal size (< 100 nm). Through incorporation of Fe in the framework, the Fe-MFI zeolite exhibited weak Brönsted acid sites. Therefore, the nanosized Fe-MFI zeolite showed high propylene/ethylene ratio and low paraffins yield, as well as long lifetime in MTO reaction.

Recently, Wu et al. derived a new strategy to synthesize Sn-β zeolite through interzeolite transformation of all-silica ITQ-1 zeolite (MWW type) by hydrothermal synthesis [64]. This method required the as-synthesized ITQ-1 zeolite as a silicon source, the dealuminated H-β zeolite as the seeds, SnCl4·5H2O as the Sn source, and tetraethylammonium hydroxide (TEAOH) as the template. As illustrated in Fig. 6, the crystallization of Sn-β zeolite followed the procedure including dissolution of MWW crystals and recrystallization of degraded silicate fragments assisted by β seeds under the aid of TEAOH template. Compared with the conventional hydrothermal synthesis method, this interzeolite transformation approach could greatly shorten the crystallization time of Sn-β zeolite, increase the content of tetrahedral Sn4+ in framework to 3.03 wt%, and enhance the hydrophobicity and strength of Lewis acid sites. Therefore, the Sn-β zeolite could be used as a promising Lewis acid catalyst in the Baeyer-Villiger oxidation of cyclohexanone with hydrogen peroxide, exhibiting low molar ratio of H2O2/cyclohexanone, high catalytic activity and product selectivity. This interzeolite transformation is considered as a powerful protocol for the synthesis of other heteroatom zeolites with high content of framework heteroatoms in short crystallization time.

Download:
Fig. 6. Schematic representation of the proposed crystallization mechanism for seed-assisted transformation of parent ITQ-1 to Sn-β. The red arrows indicate the BEA* CBUs derived from the degradation of seeds. Reproduced with permission [64]. Copyright 2017, Elsevier Inc.

In summary, the improved synthesis method has some advantages and optimization in comparison with the direct synthesis and post synthesis methods, such as diverse synthesis routes, short crystallization time, controllable synthesis of heteroatom zeolites with multiple structures and components, high heteroatom content in framework, environmental friendliness and so on. However, the synthetic systems are generally complex, containing some assistant substances for zeolite crystallization and heteroatom insertion into the framework. Besides, it is still unclear about the crystallization mechanism in interzeolite transformation, and there are great difficulties in screening zeolite precursors for transformation.

3. Application status of heteroatom zeolites 3.1. Application in the production of fine chemicals

Light olefins and alkyl aromatic compounds (such as toluene and poly methylbenzene), as the important chemical intermediates, have been widely used in industrial production of fine chemicals, usually based on the appropriate solid catalysts. It is well known that the unique pore structure of zeolites allows the reactant and product molecules to selectively diffuse into and out of the pores during the catalytic process, thereby exhibiting excellent shape selectivity. Therefore, several kinds of zeolites have already been used as the promising shape-selective acid catalysts for production of fine chemicals especially in petrochemistry field. Due to the adjustable acid properties and pore structures, in recent years the heteroatom zeolites have also attracted much attention as the catalysts in this field, showing improved activity, good selectivity, and strong anti-coking ability [65, 66].

Yuan et al. prepared tin-containing MFI zeolite (Sn-MFI or Sn-Al-MFI) with controlled Al content by using EDTA-Sn complex as the tin source and citrate as a buffer through an improved direct synthesis method for the first time [67]. The Sn4+ releasing rate could be well controlled to connect with tetrahedral silicon species during the synthesis, resulting in the incorporation of isolated tetrahedral Sn heteroatoms in zeolite framework. Both Brönsted and Lewis acid sites were controllably introduced in Sn-Al-MFI, which could promote the dehydration and rearrangement reactions, respectively, therefore, the catalyst exhibited remarkable performance in the conversion of dihydroxyacetone (DHA) to methyl lactate (ML), which is a sustainable reaction for the conversion of renewable biomass-derived compounds into highvalue chemical products.

Using Triton X-100 (polyethylene glycol tert-octylphenyl ether) as a mesoporous template, Yu et al. successfully synthesized a hierarchical TS-1 zeolite (H-TS-1) with uniform intracrystalline mesopores by a direct hydrothermal method [68]. The oxidative desulfurization (ODS) reactions of thiophene (Th), benzothiophene (BT), dibenzothiophene (DBT) and 4, 6-dimethyldibenzothiophene (4, 6-DMDBT) were used to evaluate the catalytic performance of the hierarchical TS-1 and the conventional TS-1 without mesopores for comparison. In the presence of hydrogen peroxide or tert-butyl hydroperoxide, both the conventional TS-1 and the hierarchical TS-1 showed high removal rate for Th and BT, mainly due to their small molecular size and poor micropore limitation effect inside both TS-1 zeolite catalysts. However, the catalytic activity for DBT and 4, 6-DMDBT on the hierarchical TS-1 was much higher than that on the conventional TS-1, indicating that the mesoporous structure was conducive to the improvement of catalytic performance containing large organic chemicals. With the similar method by using Triton X-100 as mesoporous template, Yu et al. also successfully synthesized nanosized hierarchical TS-1 zeolite without extra-framework anatase TiO2 under rotational crystallization conditions [69]. The materials showed higher catalytic activity and stability in the epoxidation of large olefins (such as 1-heptene and cyclopentene) than the hierarchical TS-1 zeolite synthesized under the static conditions. This is mainly due to that all titanium ions are incorporated into the framework under the assistance of Triton X-100 and acceleration of rotational crystallization.

Quite recently, Wang et al. reported the direct hydrothermal synthesis of aluminum-free hierarchical Ti-β zeolite and its application in the epoxidation of cyclohexene and 1-oectene [70]. The Ti-β zeolite was composed of plate-like nanocrystals in size of about 100 × 300 × 300 nm and intracrystalline mesopores in size of 10–40 nm, and most of the Ti heteroatoms were inserted into the zeolite framework through isomorphous substitution. Due to its unique structure, high thermal stability and improved accessible active sites, the hierarchical Ti-β zeolite exhibited the excellent catalytic performance in the epoxidation of cyclohexene and 1-oectene, and the turnover number (TON) was improved even more than 30% than the conventional reference zeolite. Therefore, the nanosized hierarchical Ti-β zeolite would be considered as a promising oxidation catalyst for the olefins in large size, showing better activity and long lifetime than the conventional Ti-β and TS-1 zeolites. The method can also be extended to prepare other aluminum-free heteroatom β zeolites with hierarchical structures, such as mesoporous Sn-β and Zr-β with Lewis acidity, which are usually used as catalysts in biomass transformation to useful fine chemicals.

The alkylation of benzene with alkenes such as methane is an important reaction for production of alkyl aromatic compounds in industry. Using the post-synthesis method of isomorphous substitution in vapor phase, Wang et al. synthesized the Zn-ZSM-5 zeolite, which exhibited Lewis acid sites, and was used as a catalyst for the alkylation reaction of benzene with methane in the temperature range of 298–623 K [71]. The in-situ solid state NMR spectroscopy and GC-MS analysis showed that toluene was selectively generated over 523 K by oxidization of O2 or N2O. The reaction pathway was proposed for the toluene formation by benzene alkylation with methane on Zn-ZSM-5 catalyst in oxidization atmosphere. Methane was first activated to generate the intermediates of zinc methyl species (Zn-CH3) and methoxy species at room temperature before the occurrence of alkylation reaction. The methyl groups of methoxy species were electrophilic, which could directly attack phenyl rings of benzene to produce toluene through the electrophilic substitution reaction (Fig. S4 in Supporting information). However, the Zn-CH3 species did not directly participate in the alkylation reaction due to their negative charge, which could be easily transformed into methoxy species at elevated temperature. Therefore, the Zn-ZMS-5 zeolite played an important role in activation of CH4, which could react with CH4 to produce active zinc methyl species (Zn-CH3) and methoxy species.

In recent years, the research about methanol to hydrocarbons (MTH) process has attracted much attention, which can produce fuel grade hydrocarbons from non-oil resources such natural gas, biomass, coal and so on. The acidic zeolites are usually employed as the catalyst in the MTH reaction for production of alkanes, alkenes, and aromatics. The selectivity toward diverse hydrocarbons can be adjusted by controlling the surface acidity and micropore structure, which is dependent on the incorporation of different heteroatoms in zeolites. Mentzel et al. synthesized conventional Ga-MFI and mesoporous Ga-MFI by the direct hydrothermal method and post-synthesis method, respectively, and they were then used as the MTH catalyst to compare their deactivation behavior with Al-MFI zeolite [72]. The Brönsted acid strength was reduced through incorporating Ga atoms in framework, resulting in insignificant coke of only 0.2 wt% in the deactivated mesoporous Ga-MFI zeolite. However, this deactivation was irreversible, quite different from that of the Al-MFI zeolite by coke deposition. The deactivation of Ga-MFI in MTH reaction was induced by the loss of weak Brönsted acid sites due to steaming of Ga-O bonds and formation of extra-framework gallium during the catalytic reaction and reactivation process.

The selective activation of functional groups is highly required in organic and biomass transformations, and the reaction is usually conducted under the assistance of Lewis acid catalysts. Luo et al. successfully synthesized stannosilicate zeolites with MFI topology and nanosheet morphology (denoted as Sn-MFI-ns) by using a dual porogenic surfactant as the organic structure-directing agent [73]. The Sn atoms were proved to enter the framework in tetrahedrally-coordinated form, leading to the introduction of Lewis acid sites in Sn-MFI-ns, which was then utilized as the catalyst in Baeyer-Villiger oxidation of 2-adamantanone in the presence of hydrogen peroxide. The catalyst showed higher conversion (92%) and selectivity to lactone (>98%) than the conventional bulk Sn-MFI, owing to much larger intersheet mesopore volume and external surface area (> 430 m2/g) of Sn-MFI-ns, composed of large number of nanosheets in thickness of ~2 nm. Moreover, the Sn-MFI-ns and Sn-MCM-41, a mesoporous material, showed comparable catalytic activity and efficiency, indicating both the catalysts owned open pore architectures and accessible active sites. However, the Sn-MFI-ns exhibited higher thermal and hydrothermal stability than Sn-MCM-41. After heating at 1273 K for 4 h, the surface area of Sn-MCM-41 was reduced from 900 m2/g to 53 m2/g, and the oxidation activity was reduced by 78%. For comparison, the surface area of the Sn-MFI-ns sample only decreased by 5% of its original surface area, and the oxidation activity was not reduced at all, revealing its great reusability and long lifetime in the high-temperature catalytic reactions.

Although the rapid development of industry brings huge economic benefits to human beings, it also produces the environmental pollution problems. The emission of SOx, NOx and other waste gases and the sewage generated during the production process seriously damage the ecological environment as well as threaten human health. Therefore, in recent years, researchers have put a lot of attention on developing new materials and techniques for purifying or reducing the exhaust gases [74]. The crystalline framework of zeolites is usually negatively charged, with positive ions as equilibrium charges. Besides, the intrinsic structural characteristics, such large specific surface area, low cost, excellent adsorption and catalytic performance, endow the zeolites with intensive applications in air purification and wastewater treatment [75].

Xiao et al. successfully synthesized Fe-β zeolite by the organic template-free and seed-directed route [76], showing high crystallinity, open micropores, large surface area, and tetrahedrally coordinated Fe3+ species in framework. The Fe-β zeolite was evaluated as a catalyst in the direct decomposition of N2O, a greenhouse gas with high warming potential. The steam treatment could induce the extraction of framework Fe3+ species, resulting in the formation of Fe3+ oligomeric clusters in the zeolite channel, which were proved as active sites for the direct decomposition of N2O. Therefore, steaming treatment could promote the redistribution of iron species, and the steam-treated Fe-β zeolite exhibited excellent catalytic activity for 100% N2O conversion at 776 K.

Heteroatom zeolites can also be used as the filter materials for separation and purification of the polluted gases. Martin-Gil et al. prepared titanosilicate zeolites including ETS-10 and TS-1 zeolites by hydrothermal synthesis method [77]. Then, the obtained titanosilicate zeolites were fabricated as fillers of the mixed matrix membranes to evaluate their performance in CO2/CH4 gas separation. The results indicated that the tetrahedron Ti atoms in zeolite framework played an important role in gas separation. Therefore, the TS-1 zeolite with higher Ti content in framework showed higher adsorption capacity for CO2, and the membrane using TS-1 (Si/Ti = 25) as the filler displayed the maximum increase of 89.1% of CO2 permeability and 23.9% increase in separation factor, which was the most suitable filler material in mixed matrix membranes for gas separation applications.

Mintova et al. reported a highly hydrophobic W-MFI zeolite [36], and assembled these nanosized crystals into membranes for selective detection of exhaust gases (CO, CO2, NO, and NO2), as shown in Fig. S5 (Supporting information) [78]. As described above, the hydrophobicity was greatly improved because the W incorporation into the framework could reduce the number of hydrophilic silanol groups. Using a spin coating method, the W-MFI zeolite nanocrystals were uniformly casted on silicon wafers to form the homogeneous films with controllable thickness of 600– 1600 nm and high mechanical stability, which were then used for detecting exhaust gases in low concentrations, including CO, CO2, NO, and NO2 (1–100 ppm) in the presence of water (100 ppm). Compared to pure Si-MFI film, the W-MFI film showed high sensitivity to the water-contained exhaust gases with low concentrations (1–30 ppm). Notably, ultra-low concentrations of NO2 (1 ppm) and CO2 (1–3 ppm) could be selectively detected by the W-MFI film, demonstrating its high molecular recognition capacity toward low concentrations of exhaust gases in the humid atmosphere.

Mintova et al. synthesized tin containing MFI zeolite nanocrystals (Sn-MFI) by the hydrothermal method, which were then casted into a self-supported thin film for the selective detection of automobile exhaust gases (CO, CO2, NOx) [79]. The NOx adsorption ability was improved when the Sn atoms were incorporated into the framework, showing higher sensitivity towards NO2 than Si-MFI zeolite. In addition, the Sn-MFI film exhibited high adsorption capacity and fast response (30 s) towards NO2, indicating that it could be considered as the selectively-sensing materials for NO2 especially at very low concentration in the humid atmosphere. Therefore, the advantages of nanosized heteroatom zeolites and their self-supported films, such as high thermal/hydrothermal stability, well-developed porosity, accessibility of active sites, tunable surface hydrophobicity/hydrophilicity endow them with a great potential for designing multi-functional sensors to selectively and accurately detect harmful gases or volatile organic compounds.

Peng et al. prepared a Sn-MFI zeolite with a Si/Sn molar ratio of 33–133, by incorporating Sn into the MFI zeolite with a direct hydrothermal synthesis method [80], which was then evaluated as the catalyst in selective reduction of NOx with propane. The Sn atoms were proved to enter the zeolite framework in tetrahedral coordination, introducing rich Lewis acid sites, large number of mesopores and active surface oxygen species. The Lewis acid sites could facilitate the adsorption of NOx and active the NOx adsorbates as well. Moreover, the mesoporous structure is favorable for the in-pore diffusion of reactants and products. Therefore, the Sn-MFI zeolite showed an improved activity of NOx selective catalytic reduction, a strong resistance to deactivation in sulfur dioxide and steam. The intrinsic characteristics endow the Sn-MFI zeolite with a potential application as an industrial catalyst in exhaust abatement.

3.3. Application in the field of biomass conversion

In recent years, highly-efficient transformation of renewable resources of biomass has received more and more attention. Biomass is an easily available and widespread renewable carbon source, which can be used to produce a variety of useful chemical intermediates and fuels. Among them, the effective catalytic conversion of lignocellulosic biomass to platform chemicals has attracted more curious interests [81-83]. For example, lactic acid is often used as a monomer for production of biodegradable polymers (such as polylactic acid), which are environmentally friendly alternatives to common plastics. Therefore, lactic acid is considered as one of the most representative biomass platform molecules, which can be further transformed into various useful chemical intermediates [82-86]. In comparison with the complicated biological methods to produce these substances, the catalytic methods from biomass/carbohydrates to lactic acid or alkyl lactate have many advantages and receive much attention [85-87]. The ketone-aldehyde intermediates are rearranged by the substance transfer mechanism [88, 89], in which zeolites containing Lewis acid sites can be used as the promising catalysts for carbohydrate isomerization. To date, many reports have been published about the framework Sn-doped zeolites, exhibiting high activity and selectivity in this type of reactions [90, 91].

Davis et al. synthesized large-pore Sn-β zeolite through incorporating Sn into the framework of β zeolite by a direct hydrothermal synthesis method [92]. The Sn-β zeolite was used as the catalyst in isomerization of glucose into fructose, which is an important reaction for production of high-fructose corn syrup (HFCS) and key intermediate step for biomass transformation to fuels and other chemicals. The incorporation of Sn in framework could introduce large number of Lewis acid sites in the zeolite, and they were active in the isomerization of aldoses, such as glucose, because a hydride could easily transfer from the hydroxyl groups of the alcohol to the carbonyl groups of the ketone, similar with the Meerwein-Ponndorf-Verley (MPV) reduction of carbonyl compounds. The reaction performance was found to be greatly affected by the framework Sn atoms and the micropore structures. Moreover, the isomerization of glucose did not occur on the medium-pore zeolite (such as Sn-MFI), and the activity was lower when using Sn-MCM-41 as the catalyst. Most importantly, the Sn-β zeolite could be reused for several cycles, and it still exhibited high activity in acid solution or at high temperature, demonstrating high stability of the Lewis acid sites in the framework of Sn-β zeolite.

In recent years, the catalytic reactions of thiol coupling or disulfide bond formation have received much attention, because the reactions are usually enclosed in protein chemistry or biological functionalities of proteins. Patra et al. synthesized mesoporous Mn-ZSM-5 zeolite through a two-step hydrothermal method [93], incorporating Mn4+ atoms into the zeolite framework, which were stable in the hot liquid and provided active sites for the cleavage of sulfur-hydrogen bonds in air and formation of disulfide bonds from thiols. The Mn-ZSM-5 zeolite was used as a catalyst for aerobic oxidation of thiols to disulfides in air as the oxidizing agent, showing high catalytic activity in comparison with ZSM-5 and mesoporous MnO2. The catalyst was also demonstrated to possess very high recycling efficiency after five consecutive cycles, indicating that Mn sites were firmly bound in the zeolite framework and the oxidation state of Mn was well maintained after the catalytic reaction. The superior catalytic performance made the mesoporous Mn-ZSM-5 zeolite have a new opportunity in heterogeneous catalysis for the synthesis of value-added disulfides in aqueous medium.

The zeolite materials with Lewis acid sites, such as Sn-β, have been generally reported to convert sugars, such as glucose, fructose and sucrose into racemic methyl lactate. Holm et al. prepared Sn-β zeolite containing Sn in the framework, and used it in the esterification reaction of hemicellulose pentose and hexose sugars into methyl lactate, glycolaldehyde dimethyl acetal and methyl vinylglycolate in the presence of methanol [94]. The total yield in pentose conversion could reach 53%–55%. No differences in the product distribution were observed for the monosaccharides, indicating they underwent the similar isomerization within their respective families such as aldose, ketose, aldose. The pentoses and hexoses were converted into methyl lactate with moderate yields, therefore, the crude sugar mixture could be used for the esterification reaction using Sn-β zeolite as the catalyst.

Ren et al. synthesized a single-unit-cell Sn-MFI zeolite by a one-pot hydrothermal method for the first time [95], based on the rotational intergrowths of single-unit-cell lamellae to form the repetitive branching. The Sn atoms were demonstrated to be uniformly distributed within the framework in tetrahedral coordination, introducing strong Lewis acid sites in the zeolite framework. In comparison with the conventional micro- and mesoporous materials, the single-unit-cell Sn-MFI zeolite showed promising catalytic performance in the isomerization of glucose and lactose, mainly due to their fully accessible active sites and combination of both Lewis acid and Brönsted acid in the framework.

Guo et al. successfully obtained Sn-MWW zeolite by incorporating Sn into the deboronized B-MWW zeolite framework by a post-treatment method [96], and then used it in the conversion of sugars to methyl lactate and lactic acid. The MWW zeolite comprises a 10-membered ring (MR) interlayer pore opening, which connects with a 12-MR supercage and an independent intralayer sinusoidal 10-MR channel. The characterizations showed that the tetrahedrally-coordinated Sn atoms in the framework introduced large number of Lewis acid sites in Sn-MWW zeolite, exhibiting high DHA conversion and high selectivity towards methyl lactate, in comparison with B-MWW which only possessed strong Brönsted acidity. Since the highly-crystallized Sn-MWW zeolite exhibited high stability under hydrothermal conditions, the catalytic activity in sugar conversion remained at 100% after repeated recycling for 3 times. Therefore, the Sn-MWW zeolite was considered as a highly-efficient and selective catalyst for the conversion of mono- or disaccharides into methyl lactate.

The commercial TS-1 zeolite was treated in tetrapropylammonium hydroxyde for the introduction of mesoporosity inside the crystals [97], and the mesoporous TS-1 zeolite was then used as the catalyst to produce di-acids from biomass. The yield of maleic acid could be greatly improved, which was produced from lignocellulosic biomass in the presence of hydrogen peroxide on TS-1 catalyst in combination with a suitable solvent. The use of a renewable solvent of γ-valerolactone could effectively alleviate the inactivation of TS-1 zeolite caused by the deposition of insoluble substances and the adhesion of by-products inside the zeolite micropores during the reaction. Therefore, the service life of TS-1 zeolite was extended, and the number of repeated uses was up to 17 times. After 6 h reaction at 70 ℃, the yield of maleic acid increased from 54% to 70%. Therefore, the combination of suitable catalysts and solvents could well accelerate the conversion of biomass to renewable chemicals.

In addition to the catalytic reactions mentioned above, heteroatom zeolites were also reported in other catalytic reactions. For example, Ce-Ni-MFI was used in low-temperature water-gas shift reactions [98], nanosized-Ti-β zeolite could catalyze the oxidative conversion of benzothiophenes [99], and V-MFI zeolite was used as the catalyst in the oxidative dehydrogenation of propane [100]. For the large family of zeolites, their application range in catalytic reactions is very wide. In addition to the three main types of applications we mentioned in this review, the zeolites can also be used in the fields of water pollution treatment, adsorption and recovery of heavy metal ions in soil and so on. But for the heteroatom aluminosilicate zeolites, their applications are mainly concentrated in the three areas of production of fine chemicals, air pollution control and biomass conversion. There are few reports on applications in other fields. At present, only zeolite containing W heteroatoms shows good hydrophobicity, and it was also applied in the field of water pollution control [101]. We strongly believe that the applications of heteroatom zeolites may be developed in the future.

4. Summary and outlook

The research on the synthesis and application of heteroatom zeolites has attracted great interest of many researchers. For the synthesis methods of heteroatom zeolites, three typical routes including direct hydrothermal synthesis, post synthesis and improved direct synthesis, are discussed in detail, which show different features and have been used for the synthesis of several traditional heteroatom zeolites (TS-1, Sn-MFI, Sn-β, etc.) and new heteroatom zeolites (W-MFI, Mo-MFI, Sn-MFI nanosheets, etc.), respectively. In term of the structural characteristics and surface properties, the main application of heteroatom zeolites is summarized in different fields including production of fine chemicals, air pollution control and biomass conversion. However, to extensively realize the fundamental and application study about the heteroatom zeolites, several challenges should be overcome in the future research.

In the case of the synthesis methods, there are still several shortcomings in each route. Therefore, more efficient, low-cost and environmentally friendly synthesis strategies should be developed, which combine the advantages of the reported methods and are suitable for the general synthesis and scale-up production of various heteroatom zeolites. Additionally, the content and coordination state of heteroatoms as well as their thermal/hydrothermal stability in framework should be controlled precisely, which dominate the performance and industrial application in catalysis and adsorption. Moreover, much more efforts should be contributed to the research on heteroatom zeolites with new structures (topology, morphology, acidity, etc.) and new incorporated heteroatoms in framework as well as hierarchical heteroatom zeolites with well-developed mesopores.

For the application of heteroatom zeolites, the rational design and synthesis of new heteroatom zeolites with large number of active sites and stable framework can promise them with robust performance in catalysis and adsorption for production of fine chemicals, air pollution control and biomass conversion. But to date only a few heteroatom zeolites have been used as the industrial catalysts in real, such as TS-1 zeolite for olefin epoxidation. In order to broaden their practical application in the future, the low-cost, highly efficient and stable heteroatom zeolites should be prepared as the candidate catalysts. In addition, the internal relationship between their intrinsic structure properties and catalytic performance as well as the definite catalytic mechanism should be intensively exploited, which can point out the direction of catalyst optimization and screen of catalytic conditions. Moreover, the heteroatom zeolites can also be applied in other emerging fields, such as CO2 adsorption and conversion to highly-valuable fuels, degradation of organics in industrial wastewater, and so on.

Declaration of competing interest

The authors declare that there are no conflicts of interest.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (Nos. 21875044, 21673048), Key Basic Research Program of Science and Technology Commission of Shanghai Municipality (No. 17JC1400100). The authors would like extend their thanks to the support by the state key laboratory of Transducer Technology of China (No. SKT1904), Program of Shanghai Academic Research Leader (No. 19XD1420300) and Research Supporting Project (No. RSP-2019/155) by King Saud University.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2020.04.018.

References
[1]
J. Cejka, A. Corma, S. Zones, Zeolites and Catalysis: Synthesis, Reactions and Applications, Wiley, Weinheim, 2010.
[2]
J. Cejka, G. Centi, J. Perez-Pariente, et al., Catal. Today 179 (2012) 2-15. DOI:10.1016/j.cattod.2011.10.006
[3]
R.R. Xu, W.Q. Pang, J.H. Yu, et al., Chemistry of Molecular Sieves and Porous Materials. Beijing: Science Press,, 20043.
[4]
J.H. Yu, W.F. Yan, Chemistry of Nanoporous Materials: Synthesis and Preparation ò. Beijing: Science Press.
[5]
N. Liu, Study on Catalytic Decomposition of N2O by Zeolite Molecular Sieves, Beijing University of Chemical Technology, Beijing, 2013.
[6]
A.A. Ismail, R.M. Mohamed, O.A. Fouad, et al., Cryst. Res. Technol. 41 (2006) 145-149. DOI:10.1002/crat.200510546
[7]
E. Zhao, S.E. Hardcastle, G. Pacheco, Microporous Mesoporous Mater. 31 (1999) 9-21. DOI:10.1016/S1387-1811(99)00053-0
[8]
G.Q. Wu, Y.Q. Wang, L.N. Wang, et al., Chem. Eng. J. 215-216 (2013) 306-314.
[9]
J.W. Zhong, J.F. Han, Y.X. Wei, et al., Catal. Sci. Technol. 7 (2017) 4905-4923. DOI:10.1039/C7CY01466J
[10]
H.G. Jang, H.K. Min, J.K. Lee, et al., Appl. Catal. A-Gen. 437-438 (2012) 120-130.
[11]
R.M. Barrer, Hydrothermal Chemistry of Zeolites. London: Academic Press, 1982.
[12]
P.A. Jacobs, K.U. Leuven, Catal. Rev. Sci. Eng. 24 (1982) 415-440. DOI:10.1080/03602458208079659
[13]
L. Xu, Study on Synthesis, Functionalization and Catalytic Performance of Novel Heteroatom Molecular Sieves, East China Normal University, Shanghai, 2015.
[14]
N. Kosinov, C. Liu, E.J.M. Hensen, et al., Chem. Mater. 30 (2018) 3177-3198. DOI:10.1021/acs.chemmater.8b01311
[15]
C. Baerlocher, L.B. Mc Cusker, D.H. Olson, Atlas of Zeolite Framework Types, 6th ed., Elsevier, Netherlands, 2007.
[16]
H. Kosslick, V.A. Tuan, R. Fricke, Cryst. Res. Technol. 26 (1991) 64-67. DOI:10.1002/crat.2170260327
[17]
Y.J. Jin, S. Asaoka, S.D. Zhang, et al., Fuel Process. Tech. 115 (2013) 34-41. DOI:10.1016/j.fuproc.2013.03.047
[18]
S. Hodoshima, A. Motomiya, S. Wakamatsu, et al., Res. Chem. Intermed. 41 (2015) 9615-9626. DOI:10.1007/s11164-015-2023-4
[19]
M. Taramasso, S.D. Milanese, G. Perego, et al., US Patent (1983) 4410501.
[20]
M. Taramasso, G. Manara, V. Fattore, et al., US Patent (1987) 4666692.
[21]
M. Moliner, Dalton Trans. 43 (2014) 4197-4208. DOI:10.1039/C3DT52293H
[22]
G. Bellussi, R. Millini, Struct. Bond 178 (2018) 1-52.
[23]
G. Perego, G. Bellussi, C. Corno, et al., Stud. Surf. Sci. Catal. 28 (1986) 129-135.
[24]
J.S. Reddy, R. Kumar, P. Ratnasamy, Appl. Catal. 58 (1990) L1-L4. DOI:10.1016/S0166-9834(00)82273-3
[25]
M.A. Camblor, A. Corma, J.P. Pariente, Zeolites 13 (1993) 82-87. DOI:10.1016/0144-2449(93)90064-A
[26]
M.A. Camblor, A. Corma, A. Martnez, et al., Chem. Commun. 8 (1992) 589-590.
[27]
M.A. Camblor, M. Costantini, A. Corma, et al., Chem. Commun. 11 (1996) 1339-1340.
[28]
T. Blasco, M.A. Camblor, A. Corma, et al., Chem. Commun. 20 (1996) 2367-2368.
[29]
N. Jappar, Q.H. Xia, T. Tatsumi, J. Catal. 180 (1998) 132-141. DOI:10.1006/jcat.1998.2266
[30]
T. Tatsumi, N. Japper, J. Phys. Chem. B 102 (1998) 7126-7131. DOI:10.1021/jp9816216
[31]
P. Wu, T. Komatsu, T. Yashima, J. Phys. Chem. 100 (1996) 10316-10322. DOI:10.1021/jp960307d
[32]
P. Wu, T. Komatsu, T. Yashima, J. Catal. 168 (1997) 400-411. DOI:10.1006/jcat.1997.1679
[33]
P. Wu, T. Komatsu, T. Yashima, Stud. Surf. Sci. Catal. 105 (1997) 663-670.
[34]
P. Wu, T. Komatsu, T. Yashima, J. Phys, Chem. B 102 (1998) 9297-9303. DOI:10.1021/jp982951t
[35]
T. Blasco, M.A. Camblor, A. Corma, et al., J. Phys. Chem. B 102 (1998) 75-88. DOI:10.1021/jp973288w
[36]
J. Grand, S.N. Talapaneni, A. Vicente, et al., Nat. Mater. 16 (2017) 1010-1015. DOI:10.1038/nmat4941
[37]
G.J. Lv, S.L. Deng, Z. Yi, et al., Chem. Commun. 55 (2019) 4885-4888. DOI:10.1039/C9CC00715F
[38]
F. Dubray, S. Moldovan, C. Kouvatas, et al., J. Am. Chem. Soc. 141 (2019) 8689-8693. DOI:10.1021/jacs.9b02589
[39]
Q. Guo, F.T. Fan, E.A. Pidko, et al., ChemSusChem 6 (2013) 1352-1356. DOI:10.1002/cssc.201300160
[40]
C.C. Chang, Z.P. Wang, P. Dornath, et al., RSC Adv. 2 (2012) 10475-10477. DOI:10.1039/c2ra21381h
[41]
P. Li, G.Q. Liu, H.H. Wu, et al., J. Phys. Chem. C 115 (2011) 3663-3670. DOI:10.1021/jp1076966
[42]
J. Dijkmans, D. Gabriëls, M. Dusselier, et al., Green Chem. 15 (2013) 2777-2785. DOI:10.1039/c3gc41239c
[43]
C.D. Chang, C.T.W. Chu, J.N. Miale, et al., J. Am. Chem. Soc. 106 (1984) 8143-8146. DOI:10.1021/ja00338a023
[44]
B. Kraushaar, J.H.C. Van Hooff, Catal. Lett. 1 (1988) 81-84. DOI:10.1007/BF00772769
[45]
K. Yamagishi, S. Namba, T. Yashima, J. Catal. 121 (1990) 47-55. DOI:10.1016/0021-9517(90)90215-6
[46]
K. Yamagishi, S. Namba, T. Yashima, J. Phys. Chem. 95 (1991) 872-877.
[47]
M.S. Rigutto, R.D. Ruiter, J.P.M. Niederer, et al., Stud. Surf. Sci. Catal. 84 (1994) 2245-2252.
[48]
B. Tang, W.L. Dai, X.M. Sun, et al., Green Chem. 16 (2014) 2281-2291. DOI:10.1039/C3GC42534G
[49]
J.W. Yoo, C.W. Lee, J.S. Chang, et al., Catal. Lett. 66 (2000) 169-173. DOI:10.1023/A:1019059925685
[50]
J. Gao, M. Liu, X.S. Wang, et al., Ind. Eng. Chem. Res. 49 (2010) 2194-2199. DOI:10.1021/ie901360y
[51]
P. Wolf, C. Hammond, S. Conrad, et al., Dalton Trans. 43 (2014) 4514-4519. DOI:10.1039/c3dt52972j
[52]
R. Xu, W. Pang, Chemistry-Zeolites and Porous Materials. Beijing: Science Press, 2004.
[53]
X. Li, B.S. Li, H.H. Mao, et al., J. Colloid Interface Sci. 332 (2009) 444-450. DOI:10.1016/j.jcis.2009.01.006
[54]
X. Li, B.S. Li, J.Q. Xu, Colloids Surf. A Physicochem. Eng. Asp. 434 (2013) 287-295. DOI:10.1016/j.colsurfa.2013.05.031
[55]
B.S. Li, X. Li, J.Q. Xu, et al., J. Colloid Interface Sci. 346 (2010) 199-207. DOI:10.1016/j.jcis.2010.02.021
[56]
J. Wang, J.Q. Xu, B.S. Li, et al., Mater. Lett. 124 (2014) 54-56. DOI:10.1016/j.matlet.2014.03.085
[57]
X. Li, B.S. Li, J.Q. Xu, et al., Appl. Clay Sci. 50 (2010) 81-86. DOI:10.1016/j.clay.2010.07.006
[58]
X.W. Cheng, J. Wang, J. Guo, et al., ChemPhysChem 7 (2006) 1198-1202. DOI:10.1002/cphc.200600002
[59]
X.W. Cheng, J.J. Mao, X.C. Lv, et al., J. Mater. Chem. A 2 (2014) 1247-1251. DOI:10.1039/C3TA14235C
[60]
W. Luo, X.Y. Yang, Z.R. Wang, et al., Microporous Mesoporous Mater. 243 (2017) 112-118. DOI:10.1016/j.micromeso.2017.01.040
[61]
Z.H. Kang, X.F. Zhang, H.O. Liu, et al., Chem. Eng. J. 218 (2013) 425-432. DOI:10.1016/j.cej.2012.12.019
[62]
Z.Y. Han, F. Zhang, X.H. Zhao, Microporous Mesoporous Mater. 290 (2019) 109679-109687. DOI:10.1016/j.micromeso.2019.109679
[63]
K. Miyake, Y. Hirota, K. Ono, et al., New J. Chem. 41 (2017) 2235-2240. DOI:10.1039/C6NJ03538H
[64]
Z.G. Zhu, H. Xu, J.G. Jiang, et al., J. Catal. 352 (2017) 1-12. DOI:10.1016/j.jcat.2017.04.031
[65]
G. Bellussi, G. Pazzuconi, C. Perego, et al., J. Catal. 157 (1995) 227-234. DOI:10.1006/jcat.1995.1283
[66]
P. Ratnasamy, D. Srinivas, H. Knözinger, J. Adv. Catal. Sci. Technol. 48 (2004) 1-169.
[67]
E.H. Yuan, W.L. Dai, G.J. Wu, et al., Microporous Mesoporous Mater. 270 (2018) 265-273. DOI:10.1016/j.micromeso.2018.05.032
[68]
S.T. Du, F. Li, Q.M. Sun, et al., Chem. Commun. 52 (2016) 3368-3371. DOI:10.1039/C5CC08441E
[69]
T.J. Zhang, X.X. Chen, G.R. Chen, et al., J. Mater. Chem. A 6 (2018) 9473-9479. DOI:10.1039/C8TA01439F
[70]
B.R. Wang, X.X. Peng, J.M. Yang, et al., Microporous Mesoporous Mater. 278 (2019) 30-34. DOI:10.1016/j.micromeso.2018.11.019
[71]
X.M. Wang, J. Xu, G.D. Qi, et al., J. Phys. Chem. C 117 (2013) 4018-4023. DOI:10.1021/jp310872a
[72]
U.V. Mentzel, K.T. HØjholt, M.S. Holm, et al., Appl. Catal. A-Gen. 417-418 (2012) 290-297.
[73]
H.Y. Luo, L. Bui, W.R. Gunther, et al., ACS Catal. 2 (2012) 2695-2699. DOI:10.1021/cs300543z
[74]
J. Despres, M. Koebel, O. Kröcher, et al., Microporous Mesoporous Mater. 58 (2003) 175-183. DOI:10.1016/S1387-1811(02)00627-3
[75]
M. Sakizci, B.E. Alver, E. Yörükoğullari, Adsorp. Soc. 17 (2011) 739-745. DOI:10.1007/s10450-011-9352-4
[76]
H.Y. Zhang, L.L. Chu, Q. Xiao, et al., J. Mater. Chem. A 1 (2013) 3254-3257. DOI:10.1039/c3ta01238g
[77]
V. Martin-Gil, A. López, P. Hrabanek, et al., J. Membr. Sci. 523 (2017) 24-35. DOI:10.1016/j.memsci.2016.09.041
[78]
J. Grand, S.N. Talapaneni, H.A. Aleksandrov, et al., ACS Appl. Mater. Interface 11 (2019) 12914-12919. DOI:10.1021/acsami.8b17626
[79]
S.N. Talapaneni, J. Grand, S. Thomas, et al., Mater. Design 16 (2016) 30317-30340.
[80]
Y. Peng, H.G. Peng, W.M. Liu, et al., RSC Adv. 5 (2015) 42789-42797. DOI:10.1039/C5RA05306D
[81]
P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538-1558. DOI:10.1039/C1CS15147A
[82]
J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, et al., Chem. Rev. 110 (2010) 3552-3599. DOI:10.1021/cr900354u
[83]
R.L. Huang, W. Qi, R.X. Su, et al., Chem. Commun. 46 (2010) 1115-1117. DOI:10.1039/B921306F
[84]
G. Epane, J.C. Laguerre, A. Wadouachi, et al., Green Chem. 12 (2010) 502-506. DOI:10.1039/b922286c
[85]
C. Gao, C.Q. Ma, P. Xu, Biotechnol. Adv. 29 (2011) 930-939. DOI:10.1016/j.biotechadv.2011.07.022
[86]
R.M. West, M.S. Holm, S. Saravanamurugan, et al., J. Catal. 269 (2010) 122-130. DOI:10.1016/j.jcat.2009.10.023
[87]
C.B. Rasrendra, B.A. Fachri, I.G.B.N. Makertihartha, et al., ChemSusChem 4 (2011) 768-777. DOI:10.1002/cssc.201000457
[88]
L. Li, C. Stroobants, K. Lin, et al., Green Chem. 13 (2011) 1175-1181. DOI:10.1039/c0gc00923g
[89]
P.P. Pescarmona, K.P.F. Janssen, C. Delaet, et al., Green Chem. 12 (2010) 1083-1089. DOI:10.1039/b921284a
[90]
E. Taarning, S. Saravanamurugan, M.S. Holm, et al., ChemSusChem 2 (2009) 625-627. DOI:10.1002/cssc.200900099
[91]
M.S. Holm, S. Saravanamurugan, E. Taarning, Science 328 (2010) 602-605. DOI:10.1126/science.1183990
[92]
M. Moliner, Y.R. Leshkov, M.E. Davis, PNAS 107 (2010) 6164-6168. DOI:10.1073/pnas.1002358107
[93]
A.K. Patra, A. Dutta, M. Pramanik, et al., ChemCatChem 6 (2014) 220-229. DOI:10.1002/cctc.201300850
[94]
M.S. Holm, Y.J.P. Torres, S. Saravanamurugan, et al., Green Chem. 14 (2012) 702-706. DOI:10.1039/c2gc16202d
[95]
L.M. Ren, Q. Guo, P. Kumar, et al., Angew. Chem. Int. Ed. 54 (2015) 10848-10851. DOI:10.1002/anie.201505334
[96]
Q. Guo, F.T. Fan, E.A. Pidko, et al., ChemSusChem 6 (2013) 1352-1356. DOI:10.1002/cssc.201300160
[97]
Y. Rodenas, R. Mariscal, J.L.G. Fierro, et al., Green Chem. 20 (2018) 2845-2856. DOI:10.1039/C8GC00857D
[98]
S. Alamolhoda, Novel Combination of Cerium and Nickel in Ce-Ni-MFI Catalysts in Low-Temperature Water Gas Shift Reaction, University of Calgary, Alberta, 2019.
[99]
U. Maity, J.K. Basu, S. Sengupta, Fuel 113 (2013) 180-186. DOI:10.1016/j.fuel.2013.05.079
[100]
A. Julbe, D. Farrusseng, J.C. Jalibert, et al., Catal. Today 56 (2000) 199-209. DOI:10.1016/S0920-5861(99)00277-1
[101]
Y. Dai, Y. Li, B.Q. Zhang, Cryst. Growth Des. 19 (2019) 4521-4525. DOI:10.1021/acs.cgd.9b00281