Ionothermal Synthesis of Triclinic SAPO-34 Zeolites

: A triclinic SAPO-34 molecular sieve was synthesized ionothermally. The as-synthesized products were characterized by powder X-ray diffraction (XRD), scanning electron microscope (SEM), nuclear magnetic resonance (NMR), fourier infrared spectrometer (FT-IR) and thermogravimetric (TG) analyses. The formation mechanism of the hierarchical porous triclinic SAPO-34 zeolites and the factors affecting the morphology of the SAPO-34 molecular sieve were investigated. The results show that the formation mechanism of the hierarchical pores is in accordance with Ostwald ripening theory, and the accumulation of grains constitutes the existence of mesopores and macropores. The crystallization temperature, ionic liquid type, and organic amines can effectively change the morphology and crystallinity of the SAPO-34 molecular sieve. The crystallization temperature, ionic liquid and template have great inﬂuence on the (111) crystal plane, thus affecting the morphology of the molecular sieve. Moreover, it can be proven through NMR and TG analyses that ionic liquids and organic amines can be used as structure directing agents together.


Introduction
Zeolites and zeotypes have emerged as some of the most significant crystalline microporous materials. Owing to their structural variability and widespread industrial applications, they have been extensively studied [1][2][3]. The silicoalumino-phosphate zeolite SAPO-34 with a CHA framework is an important microporous crystal with uniform and intricate channels, high specific surface area, adsorption capacity, and high thermal and hydrothermal stabilities, which have been widely used in catalysis [4,5], adsorption, ion-exchange, separation [6], etc. The performance in these fields is greatly influenced by the structure and properties of the SAPO-34 zeolite. The number and utilization of active sites, the ratio of weak acid to strong acid sites and the specific surface area can be changed by changing the morphology and size of zeolites [7][8][9]. The introduction of mesopores or macropores into microporous molecular sieves can greatly improve the mass transfer rate, especially in catalytic reactions, contributing to the diffusion of larger molecules, reduce the rate of formation of coke and prolong the catalytic lifetime [4]. Moreover, the synthesis methods and conditions have a substantial impact on the structure and properties of SAPO-34 zeolite.
Conventionally, SAPO-34 zeolite is mainly synthesized by a hydrothermal method, which produces high autogenous pressure and a long crystallization time. As the solvent and structure directing agent (SDA) are two different species, there is always competition between the solvent and SDA to interact with the growing framework [10]. Moreover, synthesis under high pressure must use an autoclave lined with Teflon, which limits

Effect of Crystallization Temperature
The powder X-ray diffraction (XRD) patterns are shown in Figure 1. The characteristic peaks at 9.68 • , 15.97 • , 20.2 • , 26.0 • , and 31.0 • confirm the structure of the triclinic SAPO-34 molecular sieve in these materials. The intensity of the CHA reflection peaks increased greatly with increasing temperature from 150 • C to 180 • C. The characteristic peaks of the AEI structure appeared at approximately 24.5 • and 25.4 • when the temperature rose to 200 • C. Figure 2 shows scanning electron microscope (SEM) images of the samples, and the morphology of the triclinic SAPO-34 molecular sieve changed greatly with increasing temperature. Sample S1 (150 • C) shows spherical agglomerates composed of many micrometer-sized flaky crystals, while sample S1 (180 • C) changes to a spherical aggregate composed of many cubic crystals, and each cubic crystal is approximately 1 µm. Increasing the reaction temperature from 150 • C to 180 • C promotes the growth of the (100) and (111) crystal planes. Sample S1 (200 • C) is a spherical aggregate composed of cubic crystals, but there are a few rhombus cubes. The growth of the (111) crystal plane is further promoted with increasing crystallization temperature, which favors the formation of cubic crystals.
The above results show that crystallization temperature is an important factor influencing the SAPO-34 molecular sieve. In the crystallization process of the triclinic SAPO-34 molecular sieve, a lower crystallization temperature is beneficial for the formation of flaky crystals, while a higher crystallization temperature is beneficial for the growth of the (111) crystal plane and the generation of large cubic crystals.  Scanning electron microscope (SEM) images of the samples synthesized at different crystallization temperatures ((a,b) for S1 (150 °C), (c,d) for S1 (180 °C), and (e,f) for S1 (200 °C)).

Synthesis of SAPO-34 in [EMIm]Cl
Cation [EMIm] + with smaller size is easier to form molecular sieves with an eight-membered ring CHA structure; therefore, [EMIm]Cl ionic liquid is firstly used to synthesize SAPO-34 molecular sieves.
A triclinic SAPO-34 molecular sieve was synthesized in [EMIm]Cl. The silicon contents and HF have a great influence on the synthesis, structure and morphology of the SAPO-34 molecular sieves. Figure S1 shows the XRD patterns of samples AlPO4-34, S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5. The SAPO-34 molecular sieve was prepared with a low content of silicon by using ionothermal synthesis. The detailed XRD data are shown in Table S1, and the decrease in interplanar spacing and the shift in the diffraction peak prove that silicon atoms enter the framework of the SAPO-34 molecular sieve. With increasing silicon molar amount, the morphology of the SAPO-34 molecular sieve has no obvious change, and only the overall sizes of cubic crystals and spherical aggregates decrease ( Figure S2). As shown in Figure 3, samples S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5 have absorption peaks at  The above results show that crystallization temperature is an important factor influencing the SAPO-34 molecular sieve. In the crystallization process of the triclinic SAPO-34 molecular sieve, a lower crystallization temperature is beneficial for the formation of flaky crystals, while a higher crystallization temperature is beneficial for the growth of the (111) crystal plane and the generation of large cubic crystals.  Scanning electron microscope (SEM) images of the samples synthesized at different crystallization temperatures ((a,b) for S1 (150 °C), (c,d) for S1 (180 °C), and (e,f) for S1 (200 °C)).

Synthesis of SAPO-34 in [EMIm]Cl
Cation [EMIm] + with smaller size is easier to form molecular sieves with an eight-membered ring CHA structure; therefore, [EMIm]Cl ionic liquid is firstly used to synthesize SAPO-34 molecular sieves.
A triclinic SAPO-34 molecular sieve was synthesized in [EMIm]Cl. The silicon contents and HF have a great influence on the synthesis, structure and morphology of the SAPO-34 molecular sieves. Figure S1 shows the XRD patterns of samples AlPO4-34, S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5. The SAPO-34 molecular sieve was prepared with a low content of silicon by using ionothermal synthesis. The detailed XRD data are shown in Table S1, and the decrease in interplanar spacing and the shift in the diffraction peak prove that silicon atoms enter the framework of the SAPO-34 molecular sieve. With increasing silicon molar amount, the morphology of the SAPO-34 molecular sieve has no obvious change, and only the overall sizes of cubic crystals and spherical aggregates decrease ( Figure S2). As shown in Figure 3, samples S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5 have absorption peaks at Scanning electron microscope (SEM) images of the samples synthesized at different crystallization temperatures ((a,b) for S1 (150 • C), (c,d) for S1 (180 • C), and (e,f) for S1 (200 • C)).
The above results show that crystallization temperature is an important factor influencing the SAPO-34 molecular sieve. In the crystallization process of the triclinic SAPO-34 molecular sieve, a lower crystallization temperature is beneficial for the formation of flaky crystals, while a higher crystallization temperature is beneficial for the growth of the (111) crystal plane and the generation of large cubic crystals.

Synthesis of SAPO-34 in [EMIm]Cl
Cation [EMIm] + with smaller size is easier to form molecular sieves with an eightmembered ring CHA structure; therefore, [EMIm]Cl ionic liquid is firstly used to synthesize SAPO-34 molecular sieves.
A triclinic SAPO-34 molecular sieve was synthesized in [EMIm]Cl. The silicon contents and HF have a great influence on the synthesis, structure and morphology of the SAPO-34 molecular sieves. Figure S1 shows the XRD patterns of samples AlPO 4 -34, S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5. The SAPO-34 molecular sieve was prepared with a low content of silicon by using ionothermal synthesis. The detailed XRD data are shown in Table S1, and the decrease in interplanar spacing and the shift in the diffraction peak prove that silicon atoms enter the framework of the SAPO-34 molecular sieve. With increasing silicon molar amount, the morphology of the SAPO-34 molecular sieve has no obvious change, and only the overall sizes of cubic crystals and spherical aggregates decrease ( Figure S2). As shown in Figure 3, samples S1-Si-0.1, S1-Si-0.3 and S1-Si-0.5 have absorption peaks at 1215 cm −1 , 1100 cm −1 , 735 cm −1 , 645 cm −1 , 560 cm −1 , 525 cm −1 and 485 cm −1 , which prove the synthesis of SAPO-34 zeolites, and the peak at 480 cm −1 is the T-O bending vibration peak of SiO 4 [25], which confirms that silicon enters the skeleton of the crystal.
Catalysts 2021, 11, x FOR PEER REVIEW 4 of 13 1215 cm −1 , 1100 cm −1 , 735 cm −1 , 645 cm −1 , 560 cm −1 , 525 cm −1 and 485 cm −1 , which prove the synthesis of SAPO-34 zeolites, and the peak at 480 cm −1 is the T-O bending vibration peak of SiO4 [25], which confirms that silicon enters the skeleton of the crystal. Figure 4 depicts resonances at −107 and −111.8 ppm, which correspond to Si(OAl)(OSi)3 and Si(OSi)4, respectively [24,26], indicating that the substitution of Si follows the SM2 and SM3 mechanisms. The combination of SM2 and SM3 mechanisms leads to the formation of silica-rich domains that are commonly known as Si-islands. The EDS results (Table S2) point out that the Si/Al ratio is increased by increasing the amount of Si in the synthesized gel, which also proves the formation of Si-islands.   Figure S3. SAPO-34 molecular sieves cannot be prepared without the addition of HF, and S1-F-0 has an amorphous morphology. A proper amount of HF is beneficial for the reaction of phosphorus and aluminum sources with F − , which can generate more crystal nuclei and promote the formation of the molecular sieve, while excessive HF can inhibit the release rate of phosphorus and aluminum species in the system, reduce the pH value of the system, and is not conducive to crystal growth. A SAPO-34 molecular sieve    3 and Si(OSi) 4 , respectively [24,26], indicating that the substitution of Si follows the SM2 and SM3 mechanisms. The combination of SM2 and SM3 mechanisms leads to the formation of silica-rich domains that are commonly known as Si-islands. The EDS results (Table S2) point out that the Si/Al ratio is increased by increasing the amount of Si in the synthesized gel, which also proves the formation of Si-islands. 1215 cm −1 , 1100 cm −1 , 735 cm −1 , 645 cm −1 , 560 cm −1 , 525 cm −1 and 485 cm −1 , which prove the synthesis of SAPO-34 zeolites, and the peak at 480 cm −1 is the T-O bending vibration peak of SiO4 [25], which confirms that silicon enters the skeleton of the crystal. Figure 4 depicts resonances at −107 and −111.8 ppm, which correspond to Si(OAl)(OSi)3 and Si(OSi)4, respectively [24,26], indicating that the substitution of Si follows the SM2 and SM3 mechanisms. The combination of SM2 and SM3 mechanisms leads to the formation of silica-rich domains that are commonly known as Si-islands. The EDS results (Table S2) point out that the Si/Al ratio is increased by increasing the amount of Si in the synthesized gel, which also proves the formation of Si-islands.  The influence of HF was investigated by performing the synthesis in solution with different amounts of HF. XRD patterns of samples under different amounts of HF are shown in Figure S3. SAPO-34 molecular sieves cannot be prepared without the addition of HF, and S1-F-0 has an amorphous morphology. A proper amount of HF is beneficial for the reaction of phosphorus and aluminum sources with F − , which can generate more crystal nuclei and promote the formation of the molecular sieve, while excessive HF can inhibit the release rate of phosphorus and aluminum species in the system, reduce the pH value of the system, and is not conducive to crystal growth. A SAPO-34 molecular sieve  The influence of HF was investigated by performing the synthesis in solution with different amounts of HF. XRD patterns of samples under different amounts of HF are shown in Figure S3. SAPO-34 molecular sieves cannot be prepared without the addition of HF, and S1-F-0 has an amorphous morphology. A proper amount of HF is beneficial for the reaction of phosphorus and aluminum sources with F − , which can generate more crystal nuclei and promote the formation of the molecular sieve, while excessive HF can inhibit the release rate of phosphorus and aluminum species in the system, reduce the pH value of the system, and is not conducive to crystal growth. A SAPO-34 molecular sieve with uniform morphology and high crystallinity was obtained when the amount of HF was 0.7 ( Figure S4). Figure S5a shows the N 2 adsorption/desorption isotherms of the SAPO-34 samples with different silicon contents. The steep rise followed by flat curves at low partial pressure results from the filling of the micropores, while hysteresis loops were observed which are ascribed to the presence of mesopores. The uptakes near the saturation pressure in the isotherms indicate that macropores exist in the samples. The detailed data are shown in Table S2. The mesoporous size distribution of the sample is 3~30 nm, as shown in Figure S5b. The total surface areas of all samples are higher than 512 m 2 /g. With the increase in the silicon source in the synthesis system, the specific surface area and mesoporous volume of the molecular sieve increase, and sample S-Si-0.5 has the largest external surface area (57 m 2 /g) and mesoporous volume (0.17 cm 2 /g).
To study the formation process of hierarchical porous structures, comparative timedependent experiments were conducted. Figures S6 and S7 show the XRD patterns and SEM images. The results show that the complete SAPO-34 characteristic diffraction peak and porous structure appeared at 3 h. By extending the reaction time further, the relative crystallinity increased, and after 48 h, it resembled the maximum relative crystallinity. The triclinic SAPO-34 molecular sieve started to nucleate, and metastable small grains appeared slowly in the system after crystallization treatment for 3 h. After prolonging the crystallization time to 6 h, the triclinic SAPO-34 molecular sieve showed a spherical aggregate composed of 100~200 nm small grains and the crystal surface was not completely covered by small grains. While the crystal grows rapidly with the extension of crystallization time to 12 h, the overall size of the crystal gradually increases from 4 µm to 10 µm. The crystal grows completely after 48 h, and the aluminosilicate ions in the liquid phase cannot satisfy the further growth of the crystal, so the overall size of the crystal will not increase significantly.
The proposed process of forming triclinic SAPO-34 zeolite with a hierarchically porous structure is shown in Scheme 1. The formation mechanism was in accordance with the Ostwald ripening theory. During the synthesis of the molecular sieve, the smaller grains will gradually dissolve into the surrounding medium due to the larger curvature and higher energy content and then precipitate on the surface of the larger grains [12], forming spherical agglomerates composed of small grains. With the extension of crystallization time, the crystals continue to grow and finally form a hierarchical porous triclinic SAPO-34 molecular sieve with more thermodynamically stable cubic crystals. The accumulation of grains constitutes the existence of mesopores and macropores. with uniform morphology and high crystallinity was obtained when the amount of HF was 0.7 ( Figure S4). Figure S5a shows the N2 adsorption/desorption isotherms of the SAPO-34 samples with different silicon contents. The steep rise followed by flat curves at low partial pressure results from the filling of the micropores, while hysteresis loops were observed which are ascribed to the presence of mesopores. The uptakes near the saturation pressure in the isotherms indicate that macropores exist in the samples. The detailed data are shown in Table S2. The mesoporous size distribution of the sample is 3~30 nm, as shown in Figure S5b. The total surface areas of all samples are higher than 512 m 2 /g. With the increase in the silicon source in the synthesis system, the specific surface area and mesoporous volume of the molecular sieve increase, and sample S-Si-0.5 has the largest external surface area (57 m 2 /g) and mesoporous volume (0.17 cm 2 /g).
To study the formation process of hierarchical porous structures, comparative time-dependent experiments were conducted. Figures S6 and S7 show the XRD patterns and SEM images. The results show that the complete SAPO-34 characteristic diffraction peak and porous structure appeared at 3 h. By extending the reaction time further, the relative crystallinity increased, and after 48 h, it resembled the maximum relative crystallinity. The triclinic SAPO-34 molecular sieve started to nucleate, and metastable small grains appeared slowly in the system after crystallization treatment for 3 h. After prolonging the crystallization time to 6 h, the triclinic SAPO-34 molecular sieve showed a spherical aggregate composed of 100~200 nm small grains and the crystal surface was not completely covered by small grains. While the crystal grows rapidly with the extension of crystallization time to 12 h, the overall size of the crystal gradually increases from 4 µm to 10 µm. The crystal grows completely after 48 h, and the aluminosilicate ions in the liquid phase cannot satisfy the further growth of the crystal, so the overall size of the crystal will not increase significantly.
The proposed process of forming triclinic SAPO-34 zeolite with a hierarchically porous structure is shown in Scheme 1. The formation mechanism was in accordance with the Ostwald ripening theory. During the synthesis of the molecular sieve, the smaller grains will gradually dissolve into the surrounding medium due to the larger curvature and higher energy content and then precipitate on the surface of the larger grains [12], forming spherical agglomerates composed of small grains. With the extension of crystallization time, the crystals continue to grow and finally form a hierarchical porous triclinic SAPO-34 molecular sieve with more thermodynamically stable cubic crystals. The accumulation of grains constitutes the existence of mesopores and macropores. Scheme 1. Schematic drawing of the formation mechanism of hierarchical porous triclinic SAPO-34 molecular sieves in ionic liquids. Scheme 1. Schematic drawing of the formation mechanism of hierarchical porous triclinic SAPO-34 molecular sieves in ionic liquids.

Effect of Types of Ionic Liquid
To study the influence of cations on the triclinic SAPO-34 molecular sieve, ionic liquids with the same anions but different cations were used. The structure of the used ionic liquids is shown in Figure S8. The SAPO-34 molecular sieves synthesized in either [EMIm]Cl (S-EM-C) or [EMIm]Br (S-EM-B) were compared to those obtained in either [BMIm]Cl (S-BM-C) or [BMIm]Br (S-BM-B). The morphology of the sample changed from spherical to cubic as the cation of the ionic liquid was changed from 1-ethyl-3-methyl imidazolium to 1-butyl-3-methyl imidazolium, while the size of the sample decreased. Cations in short carbon chain substituents have a higher charge density and strong interaction with the molecular sieve skeleton, which leads to strong structural guidance and stabilizes molecular sieve crystals [27]. Combined with the XRD patterns ( Figure 5), the growth of the (111) crystal plane was promoted because the alkyl chain changed from ethyl to butyl, thereby causing a change in the morphology of the SAPO-34 molecular sieve. When the cation content of the ionic liquid was increased to 1-ethyl-2,3-dimethyl imidazolium or 1-butyl-2,3-dimethyl imidazolium, the framework structure of the molecular sieve was changed, and the characteristic peaks of the AEL and AFI structures appeared.
the growth of the (111) crystal plane was promoted because the alkyl chain changed from ethyl to butyl, thereby causing a change in the morphology of the SAPO-34 molecular sieve. When the cation content of the ionic liquid was increased to 1-ethyl-2,3-dimethyl imidazolium or 1-butyl-2,3-dimethyl imidazolium, the framework structure of the molecular sieve was changed, and the characteristic peaks of the AEL and AFI structures appeared.
Since the study discussed above has shown that changing the cation can alter the morphology of the SAPO-34 molecular sieve, the next step was to investigate the effect of the anion. For this purpose, ionic liquids with the same cation but different anions were used. Hence, the SAPO-34 molecular sieves synthesized in either [EMIm]Cl (S-EM-C) or [BMIm]Cl (S-BM-C) were compared to those obtained in either [EMIm]Br (S-EM-B) or [BMIm]Br (S-BM-B). The overall morphology of S-EM-C and S-EM-B is spherical, and both S-BM-C and S-BM-B show cubic crystal morphology ( Figure 6). However, because the anion of the ionic liquid determines its polarity and coordination ability, the coordination ability of Cl − is weaker than that of Br − [28]. Therefore, [EMIm]Cl and [BMIm]Cl systems provide an environment of weak coordination and strong polarity, and the reaction in which ionic liquid cations participate is accelerated. In [EMIm]Cl, the nucleation rate is accelerated, so sample S-EM-C is stacked by small cubes, and in the [BMIm]Cl system, the generation rate of aluminosilicate ions cannot meet the needs of crystal growth, so S-BM-C is scratched and the apex angle is defective while the surface of sample S-BM-B is smooth.
Therefore, ionic liquids have a great influence on the morphology and framework type of molecular sieves, so to a certain extent, ionic liquids can be used as SDAs.  Since the study discussed above has shown that changing the cation can alter the morphology of the SAPO-34 molecular sieve, the next step was to investigate the effect of the anion. For this purpose, ionic liquids with the same cation but different anions were used. Hence, the SAPO-34 molecular sieves synthesized in either [EMIm]Cl (S-EM-C) or [BMIm]Cl (S-BM-C) were compared to those obtained in either [EMIm]Br (S-EM-B) or [BMIm]Br (S-BM-B). The overall morphology of S-EM-C and S-EM-B is spherical, and both S-BM-C and S-BM-B show cubic crystal morphology ( Figure 6). However, because the anion of the ionic liquid determines its polarity and coordination ability, the coordination ability of Cl − is weaker than that of Br − [28]. Therefore, [EMIm]Cl and [BMIm]Cl systems provide an environment of weak coordination and strong polarity, and the reaction in which ionic liquid cations participate is accelerated. In [EMIm]Cl, the nucleation rate is accelerated, so sample S-EM-C is stacked by small cubes, and in the [BMIm]Cl system, the generation rate of aluminosilicate ions cannot meet the needs of crystal growth, so S-BM-C is scratched and the apex angle is defective while the surface of sample S-BM-B is smooth.
Therefore, ionic liquids have a great influence on the morphology and framework type of molecular sieves, so to a certain extent, ionic liquids can be used as SDAs. The solid-state 13 C nuclear magnetic resonance (NMR) spectra of sample S-EM-C were characterized to further study the role of [EMIm]Cl in the synthesis of triclinic SAPO-34 molecular sieves. As we can see in Figure 7, peaks at 132 ppm and 125 ppm belong to the carbon atoms on the imidazole ring of the ionic liquid. The two peaks of the N-C-C-N bond in [EMIm]Cl overlap into one peak (125 ppm) in the solid 13 C NMR spectrum. The 39 ppm peak belongs to the substituted methyl group on the imidazole ring [23]. Peaks at 11 ppm and 44 ppm are attributed to substituted ethyl groups on the imidazole ring, which indicates that [EMIm]Cl exists in the framework structure of the molecular sieve.

Effect of Organic Amines
First, the influence of TEA dosage was investigated in [EMIm]Cl and [BMIm]Cl. The XRD patterns (Figure 8) show that [EMIm]Cl can be used as an SDA. With the addition of TEA, the relative crystallinity of the samples gradually increased. With increasing molar ratio of TEA to 1.5, a diffraction peak of the AEL molecular sieve appears. In [BMIm]Cl, when TEA is not added or the amount of TEA is too small, the samples (S2-T-0 and S2-T-0.5) show the characteristic peak of the aluminum phosphate dense phase, and there is no CHA topological structure characteristic peak. With increasing molar amount of TEA, the characteristic diffraction peaks of triclinic SAPO-34 molecular sieves with high crystallinity appeared in samples S2-T-1.0 and S2-T-1.5. The morphology of samples with different molar amounts of TEA is shown in Figures S9 and S10. The solid-state 13 C nuclear magnetic resonance (NMR) spectra of sample S-EM-C were characterized to further study the role of [EMIm]Cl in the synthesis of triclinic SAPO-34 molecular sieves. As we can see in Figure 7, peaks at 132 ppm and 125 ppm belong to the carbon atoms on the imidazole ring of the ionic liquid. The two peaks of the N-C-C-N bond in [EMIm]Cl overlap into one peak (125 ppm) in the solid 13 C NMR spectrum. The 39 ppm peak belongs to the substituted methyl group on the imidazole ring [23]. Peaks at 11 ppm and 44 ppm are attributed to substituted ethyl groups on the imidazole ring, which indicates that [EMIm]Cl exists in the framework structure of the molecular sieve. The solid-state 13 C nuclear magnetic resonance (NMR) spectra of sample S-EM-C were characterized to further study the role of [EMIm]Cl in the synthesis of triclinic SAPO-34 molecular sieves. As we can see in Figure 7, peaks at 132 ppm and 125 ppm belong to the carbon atoms on the imidazole ring of the ionic liquid. The two peaks of the N-C-C-N bond in [EMIm]Cl overlap into one peak (125 ppm) in the solid 13 C NMR spectrum. The 39 ppm peak belongs to the substituted methyl group on the imidazole ring [23]. Peaks at 11 ppm and 44 ppm are attributed to substituted ethyl groups on the imidazole ring, which indicates that [EMIm]Cl exists in the framework structure of the molecular sieve.

Effect of Organic Amines
First, the influence of TEA dosage was investigated in [EMIm]Cl and [BMIm]Cl. The XRD patterns (Figure 8) show that [EMIm]Cl can be used as an SDA. With the addition of TEA, the relative crystallinity of the samples gradually increased. With increasing molar ratio of TEA to 1.5, a diffraction peak of the AEL molecular sieve appears. In [BMIm]Cl, when TEA is not added or the amount of TEA is too small, the samples (S2-T-0 and S2-T-0.5) show the characteristic peak of the aluminum phosphate dense phase, and there is no CHA topological structure characteristic peak. With increasing molar amount of TEA, the characteristic diffraction peaks of triclinic SAPO-34 molecular sieves with high crystallinity appeared in samples S2-T-1.0 and S2-T-1.5. The morphology of samples with different molar amounts of TEA is shown in Figures S9 and S10.

Effect of Organic Amines
First, the influence of TEA dosage was investigated in [EMIm]Cl and [BMIm]Cl. The XRD patterns (Figure 8) show that [EMIm]Cl can be used as an SDA. With the addition of TEA, the relative crystallinity of the samples gradually increased. With increasing molar ratio of TEA to 1.5, a diffraction peak of the AEL molecular sieve appears. In [BMIm]Cl, when TEA is not added or the amount of TEA is too small, the samples (S2-T-0 and S2-T-0.5) show the characteristic peak of the aluminum phosphate dense phase, and there is no CHA topological structure characteristic peak. With increasing molar amount of TEA, the characteristic diffraction peaks of triclinic SAPO-34 molecular sieves with high crystallinity appeared in samples S2-T-1.0 and S2-T-1.5. The morphology of samples with different molar amounts of TEA is shown in Figures S9 and S10. The amount of TEA has a significant influence on the synthesis of the SAPO-34 molecular sieve in [BMIm]Cl, so the influence of organic amines in [BMIm]Cl was investigated. Different kinds of organic amines (TEA, DEA, MOR, N-MIM, TEAOH) were added to explore the influence of organic amine types on the SAPO-34 molecular sieve. The structure of different organic amines is shown in Figure S11. Figure S12 shows the XRD patterns of samples with different types of organic amines. Except for sample S2-TEAOH, the characteristic diffraction peaks of CHA topology appear at 2θ = 9.68°, 15.97°, 20.2°, 26.0° and 31.0°. However, the characteristic peak of the dense phase of aluminum phosphate appears in S2-DEA. The peaks at 2θ = 9.68° and 15.97° change in most of the samples synthesized by adding different organic amines, which proves the growth of the (100) and (111) crystal planes of the triclinic SAPO-34 molecular sieve. Figure S13 shows SEM images of samples with different types of organic amines. Due to the unsuitable molar amount of TEAOH in sample S2-TEAOH, the pH environment of the synthesis system (less than 4.6) is unsuitable for the formation of the triclinic SAPO-34 molecular sieve, so the dense phase of aluminum phosphate is formed, showing an irregular sphere or cubic morphology. Sample S2-TEA shows a cubic morphology with a crystal size of approximately 6 µm, which exhibits nicks on the surface of the cubic crystal along the diagonal. Samples S2-DEA and S2-MOR are nanometer-scale cubic crystals. Sample S2-N-MIM is a scaly crystal with an average crystal size of 1 µm. 13 C NMR spectroscopy was performed to investigate the organic templates in the channels. Figure 9 shows the 13 C NMR spectra of samples with different types of templates. All samples have a strong peak at 37 ppm attributed to the methyl carbon; however, the characteristic peaks of four carbons on the butyl substituent of the ionic liquid do not exist. The peaks at 125 ppm and 130 ppm correspond to the carbon atoms on the imidazole ring [12]. Therefore, it can be clearly confirmed that the ionic liquid in the triclinic SAPO-34 molecular sieve is a 1,3-dimethylimidazole cation, which proves that [BMIm]Cl decomposes and reacts to generate 1,3-dimethylimidazole cation because of the addition of HF. In addition, the solid 13 C NMR spectra of all samples also show the resonance peaks of the organic amines [29], and the carbon atoms corresponding to the resonance peaks at different positions are shown in Figure 9.
The above results show that organic amines and ionic liquid cations fill the pores of the triclinic SAPO-34 molecular sieve together, forming a good mixed template system, which leads to the formation of products and inhibits the transition from a metastable molecular sieve structure to a thermodynamically stable inorganic dense phase. The template will significantly affect the particle size and physical and chemical properties of zeolite, and the addition of organic amine can adjust the crystal size, morphology and growth direction of the molecular sieve. The amount of TEA has a significant influence on the synthesis of the SAPO-34 molecular sieve in [BMIm]Cl, so the influence of organic amines in [BMIm]Cl was investigated. Different kinds of organic amines (TEA, DEA, MOR, N-MIM, TEAOH) were added to explore the influence of organic amine types on the SAPO-34 molecular sieve. The structure of different organic amines is shown in Figure S11. Figure S12 shows the XRD patterns of samples with different types of organic amines. Except for sample S2-TEAOH, the characteristic diffraction peaks of CHA topology appear at 2θ = 9.68 • , 15.97 • , 20.2 • , 26.0 • and 31.0 • . However, the characteristic peak of the dense phase of aluminum phosphate appears in S2-DEA. The peaks at 2θ = 9.68 • and 15.97 • change in most of the samples synthesized by adding different organic amines, which proves the growth of the (100) and (111) crystal planes of the triclinic SAPO-34 molecular sieve. Figure S13 shows SEM images of samples with different types of organic amines. Due to the unsuitable molar amount of TEAOH in sample S2-TEAOH, the pH environment of the synthesis system (less than 4.6) is unsuitable for the formation of the triclinic SAPO-34 molecular sieve, so the dense phase of aluminum phosphate is formed, showing an irregular sphere or cubic morphology. Sample S2-TEA shows a cubic morphology with a crystal size of approximately 6 µm, which exhibits nicks on the surface of the cubic crystal along the diagonal. Samples S2-DEA and S2-MOR are nanometer-scale cubic crystals. Sample S2-N-MIM is a scaly crystal with an average crystal size of 1 µm. 13 C NMR spectroscopy was performed to investigate the organic templates in the channels. Figure 9 shows the 13 C NMR spectra of samples with different types of templates. All samples have a strong peak at 37 ppm attributed to the methyl carbon; however, the characteristic peaks of four carbons on the butyl substituent of the ionic liquid do not exist. The peaks at 125 ppm and 130 ppm correspond to the carbon atoms on the imidazole ring [12]. Therefore, it can be clearly confirmed that the ionic liquid in the triclinic SAPO-34 molecular sieve is a 1,3-dimethylimidazole cation, which proves that [BMIm]Cl decomposes and reacts to generate 1,3-dimethylimidazole cation because of the addition of HF. In addition, the solid 13 C NMR spectra of all samples also show the resonance peaks of the organic amines [29], and the carbon atoms corresponding to the resonance peaks at different positions are shown in Figure 9.
The above results show that organic amines and ionic liquid cations fill the pores of the triclinic SAPO-34 molecular sieve together, forming a good mixed template system, which leads to the formation of products and inhibits the transition from a metastable molecular sieve structure to a thermodynamically stable inorganic dense phase. The template will significantly affect the particle size and physical and chemical properties of zeolite, and the addition of organic amine can adjust the crystal size, morphology and growth direction of the molecular sieve.
ically adsorbed water (near 100 °C) and the removal of organic matter in the region of 250~500 °C [12]. It is obvious that the total weight loss rate of sample S1-T-1.0 (32.38%) is greater than that of sample S1-T-0 (24.44%). In the region of 200~500 °C, the weight loss rate of sample S1-T-1.0 is 26.60%, while the weight loss rate of sample S1-T-0 is 23.29%, and the extra weight loss is attributed to the combustion of TEA [30].
Based on the above results, TEA and ionic liquids play a common structural guiding role in the synthesis of molecular sieves.

Catalyst Test
The product distribution and selectivity are related to the acidity of the catalysts in the methanol conversion reaction [31]. It can be seen from Figure S14 that there is a strong acid peak near 430 °C in S0, and the amount of strong acid in S0 is more than that of S1-Si-0.3. As shown in Table 1, the dominant products are markedly different in the methanol conversion reaction over both S1-Si-0.3 and S0. The main products over S1-Si-0.3 and S0 are dimethyl ether (DME, 72.2%) and light olefins (69%), respectively.  Figure 10 shows Thermogravimetric-Derivative thermogravimetry (TG-DTG) curves of triclinic SAPO-34 molecular sieves with (sample S1-T-1.0) and without TEA (sample S1-T-0). Both S1-T-0 and S1-T-1.0 have two distinct stages: the removal of physically adsorbed water (near 100 • C) and the removal of organic matter in the region of 250~500 • C [12]. It is obvious that the total weight loss rate of sample S1-T-1.0 (32.38%) is greater than that of sample S1-T-0 (24.44%). In the region of 200~500 • C, the weight loss rate of sample S1-T-1.0 is 26.60%, while the weight loss rate of sample S1-T-0 is 23.29%, and the extra weight loss is attributed to the combustion of TEA [30].
Catalysts 2021, 11, x FOR PEER REVIEW 9 of 13 Figure 10 shows Thermogravimetric-Derivative thermogravimetry (TG-DTG) curves of triclinic SAPO-34 molecular sieves with (sample S1-T-1.0) and without TEA (sample S1-T-0). Both S1-T-0 and S1-T-1.0 have two distinct stages: the removal of physically adsorbed water (near 100 °C) and the removal of organic matter in the region of 250~500 °C [12]. It is obvious that the total weight loss rate of sample S1-T-1.0 (32.38%) is greater than that of sample S1-T-0 (24.44%). In the region of 200~500 °C, the weight loss rate of sample S1-T-1.0 is 26.60%, while the weight loss rate of sample S1-T-0 is 23.29%, and the extra weight loss is attributed to the combustion of TEA [30].
Based on the above results, TEA and ionic liquids play a common structural guiding role in the synthesis of molecular sieves.

Catalyst Test
The product distribution and selectivity are related to the acidity of the catalysts in the methanol conversion reaction [31]. It can be seen from Figure S14 that there is a strong acid peak near 430 °C in S0, and the amount of strong acid in S0 is more than that of S1-Si-0.3. As shown in Table 1, the dominant products are markedly different in the methanol conversion reaction over both S1-Si-0.3 and S0. The main products over S1-Si-0.3 and S0 are dimethyl ether (DME, 72.2%) and light olefins (69%), respectively. Based on the above results, TEA and ionic liquids play a common structural guiding role in the synthesis of molecular sieves.

Catalyst Test
The product distribution and selectivity are related to the acidity of the catalysts in the methanol conversion reaction [31]. It can be seen from Figure S14 that there is a strong acid peak near 430 • C in S0, and the amount of strong acid in S0 is more than that of S1-Si-0.3. As shown in Table 1, the dominant products are markedly different in the methanol conversion reaction over both S1-Si-0.3 and S0. The main products over S1-Si-0.3 and S0 are dimethyl ether (DME, 72.2%) and light olefins (69%), respectively. S1-Si-0.3, which was prepared by the ionothermal method, is suitable for high-selective DME production because of its low acidity [31]. ratio of Si, and samples synthesized with different TEA amounts were denoted S1-T-y, where y represents the molar ratio of TEA. Samples synthesized with different HF amounts were denoted S1-F-z, where z represents the molar ratio of HF.
In the [BMIm]Cl ionic liquid, samples synthesized with different TEA amounts were denoted S2-T-y, where y represents the molar ratio of TEA, and samples synthesized in different kinds of templates were denoted S2-TEA, S2-DEA, S2-MOR, S2-N-MIM, and S2-TEAOH. Samples

Characterization
Powder X-ray diffraction (XRD) patterns were obtained on a D8 ADVANCE X-ray diffractometer. The crystal size and structural characteristics were observed by scanning electron microscope (SEM, JSM-7500F) equipped with energy dispersive spectrometry (EDS). The surface element content of the sample was determined by EDS. The functional group structure of the sample was determined on a Nicolet iS50 Fourier infrared spectrometer (FT-IR) from Thermo Company in the United States. Nitrogen adsorption/desorption experiments were performed on a Micromeritics ASAP-2460 automatic adsorption instrument. The zeolites were all outgassed at 150 • C for 6 h before measurement. An American PerkinElmer STA 8000 synchronous analyzer was used for thermogravimetric analysis of the samples, and the temperature was increased from 25 • C to 800 • C at a rate of 10 • C/min in an argon atmosphere. 13 C NMR and 29 Si NMR measurements were performed by a Swiss AVANCE (3) 400 WB nuclear magnetic resonance instrument.
Temperature-programmed desorption of ammonia (NH 3 -TPD) was performed on a multifunction adsorption instrument. The NH 3 -TPD process was determined using N 2 as the carrier gas with a flow rate of 30 mL·min −1 , and the activation and desorption temperature were 500 • C and 100-600 • C.
The methanol conversion reaction was tested on a fixed-bed reactor. The amount of calcined catalyst (40-60 mesh) was 0.3 g and the reaction temperature was 450 • C. The N 2 flow rate was 30 mL·min −1 , and the mass space velocity (WHSV) was 3 h −1 . On-line analysis of product components was performed using a hydrogen ion flame detector (FID) and a Plot-Q column. Methanol conversion and product selectivity were calculated by the area correction normalization method.

Conclusions
Triclinic SAPO-34 molecular sieves were prepared using ionothermal synthesis. The morphology and crystallinity of the SAPO-34 molecular sieve can be controlled by the crystallization temperature, organic amines and the type of ionic liquids. A lower crystallization temperature is beneficial for the formation of flaky crystals, while a higher crystallization temperature is beneficial for the generation of large-size cubic crystals. The type of ionic liquid has a great influence on the synthesis of SAPO-34 molecular sieves. The morphology of the sample changed from spherical to cubic as the cation of the ionic liquid was changed from 1-ethyl-3-methyl imidazolium to 1-butyl-3-methyl imidazolium. Continuing to increase the size of the cation is beneficial for the formation of large pore size AEL and AFI molecular sieves. Different templates have different effects on the growth of the (100) and (111) crystal planes of SAPO-34 zeolite. The template agent with a smaller main structure contributes to the development of a smaller grain size. In addition, it can be proven through NMR and TG analyses that ionic liquids and organic amines can be used as SDAs together.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/catal11050616/s1, Figure S1: XRD patterns of samples under different molar amounts of Si. Figure S2: SEM images of samples under different molar amounts of Si. Figure S3: XRD patterns of samples under different molar amounts of HF. Figure S4: SEM images of samples under different molar amounts of HF. Figure S5: (a) N 2 adsorption-desorption isotherms and (b) BJH pore size distributions of triclinic SAPO-34 molecular sieves. Figure S6: XRD patterns of samples at different crystallization times. Figure S7: SEM images of samples at different crystallization times. Figure S8: Structure of ionic liquids. Figure S9: SEM images of samples under different molar amounts of TEA. Figure S10: SEM images of samples at different molar amounts of TEA. Figure S11: Structure of different templating agents. Figure S12: XRD patterns of samples under different types of template. Figure S13: SEM images of samples under different types of template. Figure S14: NH3-TPD patterns of SAPO-34 molecular sieves. Table S1: XRD date of different samples. Table S2: Structural parameters of SAPO-34 molecular sieves.