Synthesis of Small-Sized SAPO-34 Crystals with Varying Template Combinations for the Conversion of Methanol to Olefins

SAPO-34 molecular sieves were synthesized under hydrothermal conditions using different combinations of tetraethyl ammonium hydroxide (TEAOH)/morpholine (Mor)/triethylamine (TEA) as templates, with different silicon:aluminum ratios. The physicochemical properties of the synthesized SAPO-34 were characterized using XRD, SEM, N2 adsorption–desorption, XRF, TG, NH3-TPD, FT-IR, and 29Si MAS NMR analyses. According to the SEM and the N2 adsorption–desorption of the catalysts produced by the ternary template exhibited a larger surface area and a smaller crystal size than those produced by the single or binary templates. The FT-IR analysis indicated the increased acidity of the catalyst prepared by the ternary template. A high activity and selectivity to olefins (C2 + C3) and an optimal silicon to aluminum ratio of 0.4 were obtained from the catalyst synthesized with the ternary template. At the reaction temperature of 450 ◦C, the methanol conversion approached 100% and the ethylene–propylene selectivity and the lifetime of the catalyst reached maximums of 89.15% and 690 min, respectively.


Introduction
The methanol-to-olefin (MTO) process has attracted much attention as the most hopeful way to convert coal and natural gas to chemicals via methanol [1].Various molecular sieves have been used to catalyze this reaction effectively.Among those molecular sieves, SAPO-34, with its CHA structure, small pore openings (about 3.8 Å), and gentle acidity [2,3] leads to high ethylene and propylene yields at high methanol conversions in the MTO reaction and could be the most promising catalyst for this process [4].However, the structural features of SAPO-34 cause the deposition of carbon, resulting in a short catalytic lifetime [5].In order to solve the rapid deactivation problem, various synthetic strategies have been developed in order to inhibit the coke deposition, such as decreasing the crystal size, modifying the acidity, and building hierarchical structures with the introduction of mesopores or macropores.
In recent decades, small sized SAPO-34 nanocrystals have attracted widespread attention.Several synthesis methods have been designed to produce SAPO-34 nanocrystals.Wu et al. [6] prepared nanosheet-like SAPO-34 crystals with a TEAOH template under microwave conditions.The use of microwaves leads to a high nucleation density and a slow crystal growth rate, resulting in the unique morphology of the sheet-like nano-SAPO-34 crystals.The catalyst lifetime was prolonged from 212 min to 380 min because the nanostructures shorten the diffusion path and decrease the coke formation rate.Yang et al. [7] synthesized SAPO-34 nanocrystals with a cube-like structure through post-synthesis treatment, which involved the milling of the micrometer-sized primary molecular sieves, followed by their recrystallization.Post-treatment, the formation of nanoscale particles occurred and there was a reduction in Si enrichment on the external surface.The nanosized catalysts exhibited a prolonged catalyst lifetime and improved their selectivity to light olefins.Mojtaba et al. [8] synthesized SAPO-34 crystals via ultrasound assisted hydrothermal methods.The ultrasound treatment not only reduced the SAPO-34 crystal formation time, but it also led to the creation of a nanosized product.The olefin selectivity and catalyst stability of the sonochemically synthesized SAPO-34 crystals were considerably higher than those of the conventional hydrothermally synthesized catalysts.
It is well known that during the synthesis of molecular sieves, the template plays important roles, such as being involved in structure-directing, space-filling, and charge-compensating [9,10].Various organic amines have been used as structure-directing agents (SDAs) to produce SAPO-34 crystals.Because of the high price and toxicity of some amines, combinations of different templates have been employed during the synthesis of SAPO-34.Wang et al. [11] studied the synthesis of SAPO-34 with a mixed template of triethyl amine (TEA) and tetraethyl ammonium hydroxide (TEAOH).SAPO-34 synthesized with a TEAOH/TEA value of 0.1 possessed a low Si content, a large surface area, and a mild acidity level, which are responsible for its higher olefin selectivity and lower carbon deposition rate in the MTO reaction.Rostami et al. [12] investigated the effects of the different types and combinations of templates on the structure and morphology of SAPO-34.The catalyst obtained using the binary template of TEAOH-Mor, which gave it a larger surface area, a smaller particle size, and allowed it to exhibit a good catalytic performance in the MTO reaction.Sedighi et al. [13] also studied the synthesis of SAPO-34 with the hydrothermal method using different combinations of Mor and TEAOH as templates.The catalyst synthesized with a TEAOH/Mor value of 1:1 showed the longest lifetime, owing to its optimal crystal size and acidity.Najafi et al. [5] used a mixture of TEAOH, Mor, diethylamine (DEA), and TEA as SDAs to synthesize SAPO-34.Compared to the single template synthesized SAPO-34, the SAPO-34 created using two, three, and four SDAs led to a crystallinity increase, a crystallite size reduction, and a morphology change to spherical aggregates of several cubic nanosized crystals.Although some other papers have reported the use of mixed templates as SDAs, systematic investigations on SAPO-34 synthesized with combinations of Mor/TEAOH/TEA and their performances on the catalytic test in MTO conversion are relatively immature.
In this work, SAPO-34 catalysts were synthesized using Mor, TEAOH, and TEA as templates, and the silicon to aluminum ratio was further tailored to improve the catalysts' performance.The synthesized catalysts were studied with XRD, SEM, N 2 adsorption-desorption, XRF, TG, NH 3 -TPD, FT-IR, and 29 Si MAS NMR characterization tests, and their catalytic performances were evaluated by taking MTO as the model reaction.

Characterization of the SAPO-34 Catalysts
SAPO-34 molecular sieves were synthesized under hydrothermal conditions and synthesis conditions for the different SAPO-34 materials were showed in Materials and Methods part.The X-ray diffraction patterns of the synthesized catalysts prepared by different gel compositions are shown in Figure 1.The strong peak at 9.5 • and double peaks at 26 • and 31 • are the characteristic diffraction peaks of SAPO-34 [14].The powder XRD patterns of all the prepared samples agreed well with the typical pattern, which indicates that both the single Mor template and mixed-template (Mor/TEA or Mor/TEA/TEAOH) can be used to prepare SAPO-34 with a CHA structure.However, there were some differences in the reflection intensity and peak width.The characteristic diffraction peak of S-1 was strong and sharp, indicating that the crystallinity and the phase purity of the sample prepared with a single template were better than those of the mixed template synthesized sample.In particular, sample S-3 exhibited the least intense reflections as a result of a loss of crystallinity due to the formation of lattice defects or crystal skeleton distortions and the reduction of crystal size with TEAOH incorporation.For samples S-4, S-5, and S-6 prepared with the ternary template, the peak intensity decreased and line broadening was observed with a decreasing silicon to aluminum ratio in the initial gel.These observations indicate that a loss of crystallinity occurred due to the high Al source incorporation.It should be noted that a weak diffraction peak of SAPO-18 at 2θ ≈ 21.6 • [15] was observed for S-5 and S-6, indicating the coexistence of SAPO-18 and SAPO-34.It may be concluded that high Si/Al favors the CHA structure (pure phase SAPO-34) formation using a mixed template, while the SAPO-18 impurity phase is formed when the mole ratio of Si/Al is lower than 0.4.
Catalysts 2018, 7, x FOR PEER REVIEW 3 of 4 TEAOH incorporation.For samples S-4, S-5, and S-6 prepared with the ternary template, the peak intensity decreased and line broadening was observed with a decreasing silicon to aluminum ratio in the initial gel.These observations indicate that a loss of crystallinity occurred due to the high Al source incorporation.It should be noted that a weak diffraction peak of SAPO-18 at 2θ ≈ 21.6° [15] was observed for S-5 and S-6, indicating the coexistence of SAPO-18 and SAPO-34.It may be concluded that high Si/Al favors the CHA structure (pure phase SAPO-34) formation using a mixed template, while the SAPO-18 impurity phase is formed when the mole ratio of Si/Al is lower than 0.4.Figure 2 shows the SEM images of the SAPO-34 catalysts.Their surface smoothness and crystallite sizes differed considerably, despite their similar cubic-like crystalline shapes [16].The approximate average crystal sizes of samples followed the order S-1 > S-2 > S-3 > S-4 > S-5.The addition of TEA and TEAOH facilitated the formation of small crystals, but the surface smoothness reduced.As the molar ratio of TEAOH decreased, molecular sieve size distributions became uniform and the equivalent diameters of spheres reduced after the small cubic particles became stacked into spherical crystals.S-5 was also shown to be a small bulk crystal with a small number of particle crystals, which may be a SAPO-18 heterocrystal detected by XRD.As the silica/alumina ratio decreased from 0.6 to 0.4, molecular sieves tended to form spherical crystals with a regular morphology, and the particle size gradually decreased.When the silica/alumina ratio continued to reduce to 0.2, the content of skeleton silica may be insufficiently matched, making SAPO-34 difficult to form and mostly amorphous.Only a small amount of the bar-shaped single crystal existed in S-6.This was confirmed by the XRD patterns, which expressed a reduction in crystallinity.Figure 2 shows the SEM images of the SAPO-34 catalysts.Their surface smoothness and crystallite sizes differed considerably, despite their similar cubic-like crystalline shapes [16].The approximate average crystal sizes of samples followed the order S-1 > S-2 > S-3 > S-4 > S-5.The addition of TEA and TEAOH facilitated the formation of small crystals, but the surface smoothness reduced.As the molar ratio of TEAOH decreased, molecular sieve size distributions became uniform and the equivalent diameters of spheres reduced after the small cubic particles became stacked into spherical crystals.S-5 was also shown to be a small bulk crystal with a small number of particle crystals, which may be a SAPO-18 heterocrystal detected by XRD.As the silica/alumina ratio decreased from 0.6 to 0.4, molecular sieves tended to form spherical crystals with a regular morphology, and the particle size gradually decreased.When the silica/alumina ratio continued to reduce to 0.2, the content of skeleton silica may be insufficiently matched, making SAPO-34 difficult to form and mostly amorphous.Only a small amount of the bar-shaped single crystal existed in S-6.This was confirmed by the XRD patterns, which expressed a reduction in crystallinity.
The surface area and volume distributions of the samples with different templates calculated from the nitrogen adsorption-desorption isotherms are listed in Table 1.Among samples S-1, S-2, and S-3, sample S-1 was shown to have the lowest surface area.This demonstrates that the surface area of SAPO-34 prepared by a mixed template improved obviously.When less TEAOH was used, there was an increase in crystallinity and a reduction in particle size which led to an enhancement of the surface area, and based on the pKa values of different templating agents shown in Table 9, this may be related to the pH of the sol-gel.
The specific surface areas of samples synthesized at different Si/Al ratios were quite different.The pore volume and the mean pore size distribution were in the order S-6 > S-5 > S-4.As the silica/alumina ratio increased, the molecular sieve pore volume and the mean pore diameter was reduced.However, the specific surface area followed the ranking S-5 > S-4 > S-6.These differences can be explained by the small particle size and the high crystallinity among the samples.Consequently, crystallinity has a higher effect on surface area than particle size [5].The surface area and volume distributions of the samples with different templates calculated from the nitrogen adsorption-desorption isotherms are listed in Table 1.Among samples S-1, S-2, and S-3, sample S-1 was shown to have the lowest surface area.This demonstrates that the surface area of SAPO-34 prepared by a mixed template improved obviously.When less TEAOH was used, there was an increase in crystallinity and a reduction in particle size which led to an enhancement of the surface area, and based on the pKa values of different templating agents shown in Table 9, this may be related to the pH of the sol-gel.
The specific surface areas of samples synthesized at different Si/Al ratios were quite different.The pore volume and the mean pore size distribution were in the order S-6 > S-5 > S-4.As the silica/alumina ratio increased, the molecular sieve pore volume and the mean pore diameter was reduced.However, the specific surface area followed the ranking S-5 > S-4 > S-6.These differences can be explained by the small particle size and the high crystallinity among the samples.Consequently, crystallinity has a higher effect on surface area than particle size [5].
Figure 3 shows the TG-DTG curves of the synthesized samples under different templating agents burning in an air atmosphere.The weight loss of all samples can be divided into three temperature regions.The first weight loss (200 °C) occurred due to adsorbed water desorption.The second weight loss (200~500 °C) was attributed to the decomposition of the template.Finally, the third weight loss (>500 °C), was related to the removal of the residual templating agent in the molecular sieve [17,18].Three absorption peaks were observed in different templating samples.Water decomposition occurred at a low temperature with a weak peak, indicating that the water content in the molecule was small and the interactions with the molecular sieve were weak.The second absorption peak represents the template removal peak intensity.In particular, sample S-3, which was obtained by the synthesis of the templating agent MOR-TEA-TEAOH, had a strong and sharp removal peak.Different templating agents also had different peak heights at this point (S-3 > S-1 > S-2), indicating that the interaction between TEAOH and the molecular sieve was the strongest.
Table 2 shows the results of the TG analysis of different templating agent synthetic samples.In the molecular weight loss study, the weight loss of water was relatively small, and the main weight Figure 3 shows the TG-DTG curves of the synthesized samples under different templating agents burning in an air atmosphere.The weight loss of all samples can be divided into three temperature regions.The first weight loss (200 • C) occurred due to adsorbed water desorption.The second weight loss (200~500 • C) was attributed to the decomposition of the template.Finally, the third weight loss (>500 • C), was related to the removal of the residual templating agent in the molecular sieve [17,18].Three absorption peaks were observed in different templating samples.Water decomposition occurred at a low temperature with a weak peak, indicating that the water content in the molecule was small and the interactions with the molecular sieve were weak.The second absorption peak represents the template removal peak intensity.In particular, sample S-3, which was obtained by the synthesis of the templating agent MOR-TEA-TEAOH, had a strong and sharp removal peak.Different templating agents also had different peak heights at this point (S-3 > S-1 > S-2), indicating that the interaction between TEAOH and the molecular sieve was the strongest.
Table 2 shows the results of the TG analysis of different templating agent synthetic samples.In the molecular weight loss study, the weight loss of water was relatively small, and the main weight loss was concentrated on the removal of the template, and the weight loss of the different samples was quite different.The molecular sieves prepared under different template synthesis systems were directly calculated using the thermogravimetric analysis results.For sample 1, which was synthesized with a single template (MOR), the average number of templating agents per cage in the sample molecular sieve was 1.76, and S-2 and S-3 had 1.3 and 1.44, respectively.It can be seen that MOR, as a template agent, more favorably entered the molecular sieve cage structure, followed by TEAOH, and the worst was TEA, which may be due to the molecular structure and volume of the template itself.TEA contains three ethyl branches.Its molecular volume is significantly larger than that of the cyclic MOR.For TEAOH, it can be seen from the DTG analysis of Figure 3 that the TEAOH templating agent and the molecular sieve had the strongest force, and easily bonded with the silicon source, phosphorus source, and aluminum source at the initial stage of crystallization, so adding TEAOH in the MOR-TEA system helped to increase the average number of templating agents in the molecular sieve cage.TEAOH, and the worst was TEA, which may be due to the molecular structure and volume of the template itself.TEA contains three ethyl branches.Its molecular volume is significantly larger than that of the cyclic MOR.For TEAOH, it can be seen from the DTG analysis of Figure 3 that the TEAOH templating agent and the molecular sieve had the strongest force, and easily bonded with the silicon source, phosphorus source, and aluminum source at the initial stage of crystallization, so adding TEAOH in the MOR-TEA system helped to increase the average number of templating agents in the molecular sieve cage.Acidity plays a significant role in the catalytic properties of solid acid catalysts.Figure 4 shows the TPD profile of the prepared SAPO-34 sample.Two NH3 desorption peaks on SAPO-34 molecular sieve can be observed, centered around about 250-300 °C, representing low temperature desorption at the weak acidic sites, and 450-520 °C as high-temperature desorption sites attributed to the strong acidic sites generated by the incorporation of silicon into the framework of the SAPO-34 molecular sieve [19].As shown in Figure 4 and Table 3, the peak positions of the strong and weak acid peaks of Acidity plays a significant role in the catalytic properties of solid acid catalysts.Figure 4 shows the TPD profile of the prepared SAPO-34 sample.Two NH 3 desorption peaks on SAPO-34 molecular sieve can be observed, centered around about 250-300 • C, representing low temperature desorption at the weak acidic sites, and 450-520 • C as high-temperature desorption sites attributed to the strong acidic sites generated by the incorporation of silicon into the framework of the SAPO-34 molecular sieve [19].As shown in Figure 4 and Table 3, the peak positions of the strong and weak acid peaks of the SAPO-34 samples synthesized by different templating agents were different than those synthesized with the MOR single templating agent-under the action of double template, the strong and weak acid adsorption peak positions move to the low temperature direction.That is, the acidity of the strong acid decreased and the effect of the three template agents (MOR/TEA/TEAOH) on the peak position was not a simple "addition" effect.Compared with the single template agent, weak acid desorption moved towards a high temperature, and the strong acid desorption peak shifted to a low temperature.Compared with the double template, the strong and weak acid peaks were biased toward the high-temperature region.
compared with that of the single template agent distinctly.
With different silica/alumina ratios, the molecular sieves exhibited different distributions of strong and weak acids.In the case of the S-6 samples, the number of weak acid sites was greater than that of strong acids.As the silica/alumina ratio increased, the number of strong acid sites in the sample gradually increased, from 45.61% to 57.69%.When SiO2/Al2O3 = 0.4, the strong and weak acid centers were evenly distributed, and the ratio between them was close to 1.When the SiO2/Al2O3 ratio increased to 0.6, the strong acid sites in sample S-4 dominated the acidic distribution.The TPD results were in good agreement with the XRD patterns.Crystallinity significantly affected the weak acids of molecular sieves.Figure 5 shows the FT-IR spectra of different templating agent synthetic samples.The f-line indicates that the sample has not absorbed NH3, and the infrared spectrum at 323 K (shown in a~e) indicates the infrared spectrum at different temperatures after the sample has adsorbed ammonia gas.It can be seen from the figure that in the sample synthesized by different template agents, the absorption peak of the hydroxyl group structure appeared at 3610 cm −1 , as compared to the f and a  Under the action of the single template MOR, the strong acid played a dominant role in acid distribution.With the addition of TEA and TEAOH, the weak acid content was gradually enhanced, and TEOH more favorably increased the amount of weak acid than TEA.In particular, when the MOR/TEA/TEAOH tri-template was used in combination, the relative content of weak acid was the highest, increasing by 15.7% compared with that of the double template agent, and by 20.01%compared with that of the single template agent distinctly.
With different silica/alumina ratios, the molecular sieves exhibited different distributions of strong and weak acids.In the case of the S-6 samples, the number of weak acid sites was greater than that of strong acids.As the silica/alumina ratio increased, the number of strong acid sites in the sample gradually increased, from 45.61% to 57.69%.When SiO 2 /Al 2 O 3 = 0.4, the strong and weak acid centers were evenly distributed, and the ratio between them was close to 1.When the SiO 2 /Al 2 O 3 ratio increased to 0.6, the strong acid sites in sample S-4 dominated the acidic distribution.The TPD results were in good agreement with the XRD patterns.Crystallinity significantly affected the weak acids of molecular sieves.
Figure 5 shows the FT-IR spectra of different templating agent synthetic samples.The f-line indicates that the sample has not absorbed NH 3 , and the infrared spectrum at 323 K (shown in a~e) indicates the infrared spectrum at different temperatures after the sample has adsorbed ammonia gas.It can be seen from the figure that in the sample synthesized by different template agents, the absorption peak of the hydroxyl group structure appeared at 3610 cm −1 , as compared to the f and a lines where the double peaks appearing near 3287 cm −1 were only new peaks derived from the probe molecules; this may be the Fermi resonance band, where the absorption peak is not the skeleton vibration structure of the molecular sieve and is negligible.The absorption peak appearing in the infrared spectrum at 1460 cm −1 is the absorption peak of the Bronsted acid (B acid) center, which is generated by the reaction of ammonia molecules with protonic acids H + to form protonated NH 4 + ions.The 1620 cm −1 is the absorption peak of the Lewis acid (L acid) center, which is produced because the ammonia gas molecule has an independent electron pair and can be coordinated with L acid to form L: NH 3 .As the temperature rose, the strength of the L acid gradually weakened until it completely disappeared, and the strength of the B acid slowly weakened, indicating that the L acid is a weak acid and B acid is a strong acid.From Figure 5c, it can be found that as the temperature rose, the position of the B acid absorption peak gradually shifted towards the low-frequency direction.
A possible reason for this is that the adsorption solvation effect of the probe molecules is different at different temperatures, resulting in the shift of the absorption peak.This shift is negligible when studying the acidity of the SAPO-34 molecule.
molecules; this may be the Fermi resonance band, where the absorption peak is not the skeleton vibration structure of the molecular sieve and is negligible.The absorption peak appearing in the infrared spectrum at 1460 cm −1 is the absorption peak of the Bronsted acid (B acid) center, which is generated by the reaction of ammonia molecules with protonic acids H + to form protonated NH4 + ions.The 1620 cm −1 is the absorption peak of the Lewis acid (L acid) center, which is produced because the ammonia gas molecule has an independent electron pair and can be coordinated with L acid to form L: NH3.As the temperature rose, the strength of the L acid gradually weakened until it completely disappeared, and the strength of the B acid slowly weakened, indicating that the L acid is a weak acid and B acid is a strong acid.From Figure 5c, it can be found that as the temperature rose, the position of the B acid absorption peak gradually shifted towards the low-frequency direction.A possible reason for this is that the adsorption solvation effect of the probe molecules is different at different temperatures, resulting in the shift of the absorption peak.This shift is negligible when studying the acidity of the SAPO-34 molecule.Figure 6a shows the infrared spectrum of the non-adsorbed NH3 sample in the hydroxyl region.In the infrared spectrum, the molecular sieve hydroxyl appeared in the range of 3200 cm −1 to 4000 cm −1 .It can be seen that all the samples showed absorption peaks at 3623 cm −1 and 3600 cm −1 , which represents two different types of skeletons.The hydroxyl corresponding to 3623 cm −1 points to the center of the ellipsoid cage and the hydroxyl corresponding to 3600 cm −1 is located within the hexagonal prism [20].The intensity of the bridged hydroxyl absorption peak at this point followed the order S-1 > S-2 > S-3.Since the strong acid center was generated by Si-OH-Al, it corresponded to the results of the distribution of strong acid in the molecular sieves in Table 3.The distribution of the remaining hydroxyl groups of the different template agent synthesis samples was also slightly different.Compared with the single template MOR synthesis sample, the sample under the composite template agent showed smaller absorption peaks at 3740 cm −1 and 3676 cm −1 .The 3740 cm −1 peak is the Si-OH absorption peak at the ends of the lattice in the framework or impurities, the band at 3676 cm −1 is the P-OH absorption peak, and the absorption peaks at 3740 cm −1 and 3676 cm −1 indicate that the SAPO-34 molecular sieve crystal had defects and that the crystallinity decreased [21], which is consistent with the XRD analysis results.The acidity of the weak acid was caused by the phosphorus hydroxyl group, and the P-OH absorption peak intensity in sample S-3 was greater than that in S-2, so the number of weak acid sites would have been greater in S-3 than in S-2, in accordance with the results of the distribution of weak acid in different template agents in the TPD analysis results.
Figure 6b and Table 4 show the distribution of the B acid and L acid on the molecular sieve sample.Absorption peaks at 1460 cm −1 and 1640 cm −1 for the samples synthesized from different templating agents characterized the B acid and L acid centers on the molecular sieve [22,23], respectively.Because their absorption coefficients in the infrared spectra are approximately the same, the intensity of the absorption peak in the infrared spectrum was used as the reference value to calculate the relative contents of the B acid and L acid directly, and Gaussian fitting was used to Figure 6a shows the infrared spectrum of the non-adsorbed NH 3 sample in the hydroxyl region.In the infrared spectrum, the molecular sieve hydroxyl appeared in the range of 3200 cm −1 to 4000 cm −1 .It can be seen that all the samples showed absorption peaks at 3623 cm −1 and 3600 cm −1 , which represents two different types of skeletons.The hydroxyl corresponding to 3623 cm −1 points to the center of the ellipsoid cage and the hydroxyl corresponding to 3600 cm −1 is located within the hexagonal prism [20].The intensity of the bridged hydroxyl absorption peak at this point followed the order S-1 > S-2 > S-3.Since the strong acid center was generated by Si-OH-Al, it corresponded to the results of the distribution of strong acid in the molecular sieves in Table 3.The distribution of the remaining hydroxyl groups of the different template agent synthesis samples was also slightly different.Compared with the single template MOR synthesis sample, the sample under the composite template agent showed smaller absorption peaks at 3740 cm −1 and 3676 cm −1 .The 3740 cm −1 peak is the Si-OH absorption peak at the ends of the lattice in the framework or impurities, the band at 3676 cm −1 is the P-OH absorption peak, and the absorption peaks at 3740 cm −1 and 3676 cm −1 indicate that the SAPO-34 molecular sieve crystal had defects and that the crystallinity decreased [21], which is consistent with the XRD analysis results.The acidity of the weak acid was caused by the phosphorus hydroxyl group, and the P-OH absorption peak intensity in sample S-3 was greater than that in S-2, so the number of weak acid sites would have been greater in S-3 than in S-2, in accordance with the results of the distribution of weak acid in different template agents in the TPD analysis results.
Figure 6b and Table 4 show the distribution of the B acid and L acid on the molecular sieve sample.Absorption peaks at 1460 cm −1 and 1640 cm −1 for the samples synthesized from different templating agents characterized the B acid and L acid centers on the molecular sieve [22,23], respectively.Because their absorption coefficients in the infrared spectra are approximately the same, the intensity of the absorption peak in the infrared spectrum was used as the reference value to calculate the relative contents of the B acid and L acid directly, and Gaussian fitting was used to obtain the relative area of each peak [24].The results are shown in Table 4.The acidity of B was much greater than that of the L acid.The B acid was dominant in terms of the acidity of the molecular sieve.The distribution of the B acid in the samples synthesized by different templating agents followed the order S-3 > S-2 > S-1, indicating that the composite templating agent is beneficial to the improvement of molecular sieves.When the MOR/TEA/TEAOH tri-template was used, the amount of B acid reached a maximum of 1.789 mmol/g.The total acid amount of the molecular sieve also varied with the type of templating agent, and the total acid content of S-3 was the largest, followed by S-2, and S-1 was the smallest and the use of a composite templating agent was beneficial to increase the amount of acid in the molecular sieve.
obtain the relative area of each peak [24].The results are shown in Table 4.The acidity of B was much greater than that of the L acid.The B acid was dominant in terms of the acidity of the molecular sieve.The distribution of the B acid in the samples synthesized by different templating agents followed the order S-3 > S-2 > S-1, indicating that the composite templating agent is beneficial to the improvement of molecular sieves.When the MOR/TEA/TEAOH tri-template was used, the amount of B acid reached a maximum of 1.789 mmol/g.The total acid amount of the molecular sieve also varied with the type of templating agent, and the total acid content of S-3 was the largest, followed by S-2, and S-1 was the smallest and the use of a composite templating agent was beneficial to increase the amount of acid in the molecular sieve.Figure 7 shows the Si environment in the SAPO-34 framework studied by 29 Si MAS NMR.From the 29 Si MAS NMR spectrum, it can be seen that S-1 and S-3 all showed strong resonance peaks at δ = −90.6 [24], corresponding to the Si(4Al) structure.This is due to the substitution of P atoms by Si atoms alone (SM2) [25].Small formants appeared at δ = −93.6,−99.2, −105.9, and −108.9, corresponding to the Si(3Al), Si(2Al), Si 1Al), and Si(0Al) structures, which occurred as the result of the joint action of SM2 and SM3 (two Si atoms replaced a pair of adjacent Al + P), forming a silicon-rich region (silicon island) whose center was replaced by SM3 and surrounded by the SM2 method.
A large number of previous studies have shown how the movement and quantity of silicon in the molecular sieve framework directly affected the properties of the SAPO-34 molecular sieve.The Si distributions of the absorption peaks of each structure using Gaussian fitting are shown in Table 5.Both S-1 and S-3 samples were shown to have a large content of Si(4Al) in their structures.This coordination environment led to SAPO-34 molecular sieves forming an anion skeleton with a net negative charge that requires protons to balance the charge to produce the B acid center; therefore, there was a large number of B acid sites in different template agent synthesis samples (Table 4).Compared to a single template, the use of a composite tri-templating agent resulted in an increase in the relative contents of Si(3Al), Si(2Al), Si(1Al), and Si(0Al) structures.Combined with the distribution of the acid properties on the molecular sieves of different template agents synthesized in Table 4 as the size of the silica islands increased, the acid strength of the molecular sieves increased  Figure 7 shows the Si environment in the SAPO-34 framework studied by 29 Si MAS NMR.From the 29 Si MAS NMR spectrum, it can be seen that S-1 and S-3 all showed strong resonance peaks at δ = −90.6 [24], corresponding to the Si(4Al) structure.This is due to the substitution of P atoms by Si atoms alone (SM2) [25].Small formants appeared at δ = −93.6,−99.2, −105.9, and −108.9, corresponding to the Si(3Al), Si(2Al), Si 1Al), and Si(0Al) structures, which occurred as the result of the joint action of SM2 and SM3 (two Si atoms replaced a pair of adjacent Al + P), forming a silicon-rich region (silicon island) whose center was replaced by SM3 and surrounded by the SM2 method.
Catalysts 2018, 7, x FOR PEER REVIEW 9 of 4 gradually [26].The formation of silicon islands helped to increase the acidity of the SAPO-34 molecular sieves, and the strongest acid sites were distributed on the edges of the silicon islands.According to Liu [27], in the study of the performance of diethylamine-oriented synthesis of SAPO-34 proposed in their study of the number of skeleton charges, the role of the template in the cage and the maximum amount of Si (4Al) in the calculation method, based on the distribution of silicon environments (Table 5) and the elemental composition of the molecular sieve (Table 6), the number of templates used to balance the charge of the skeleton in each cage Ma, the number of templates that do not participate in the balance of the skeleton charge in each cage Mb ,and the maximum amount of Si (4Al) can be calculated as shown in Table 7.
Although the total amount of templating agent in the molecular sieve cage followed the order S1 > S3 (Table 2), as the content of silicon in the skeleton increased, the value of Ma increased.The Mb of sample S-3 was 0, indicating that the templating agent in the sample cage of the molecular sieve was used to balance the skeleton charge.The proportions of the different template agents used to balance the charge of the framework followed the order S3 > S1, using the SM2 or SM3 substitution mechanism [26,28,29].The negative ion of the skeleton was generated at the same time as the formation of the silicon island structure [28,30], so the more templating agents used to balance the skeleton charge and the greater the relative content of the Si structure distribution in the silicon island structure, the stronger the acidity exhibited by the molecular sieve was, which is consistent with the results presented in Tables 4 and 5.When the silicon content in the framework of the zeolite exceeded the maximum Si(4Al) content, other silicon structures began to appear.It can be seen from Table 6 that the maximum Si(4Al) content of the samples synthesized by the composite template agent was significantly higher than that of the samples synthesized by the single template agent.The amount of framework Si(4Al) per mole (Al + Si + P) was determined to be 0.0928 and because the Si(4Al) structure generates the B acid center, the amount of B acid of the composite template agent in the synthetic sample S-3 was also greater than that in S-1.A large number of previous studies have shown how the movement and quantity of silicon in the molecular sieve framework directly affected the acid properties of the SAPO-34 molecular sieve.The Si distributions of the absorption peaks of each structure using Gaussian fitting are shown in Table 5.Both S-1 and S-3 samples were shown to have a large content of Si(4Al) in their structures.This coordination environment led to SAPO-34 molecular sieves forming an anion skeleton with a net negative charge that requires protons to balance the charge to produce the B acid center; therefore, there was a large number of B acid sites in different template agent synthesis samples (Table 4).Compared to a single template, the use of a composite tri-templating agent resulted in an increase in the relative contents of Si(3Al), Si(2Al), Si(1Al), and Si(0Al) structures.Combined with the distribution of the acid properties on the molecular sieves of different template agents synthesized in Table 4 as the size of the silica islands increased, the acid strength of the molecular sieves increased gradually [26].The formation of silicon islands helped to increase the acidity of the SAPO-34 molecular sieves, and the strongest acid sites were distributed on the edges of the silicon islands.According to Liu [27], in the study of the performance of diethylamine-oriented synthesis of SAPO-34 proposed in their study of the number of skeleton charges, the role of the template in the cage and the maximum amount of Si (4Al) in the calculation method, based on the distribution of silicon environments (Table 5) and the elemental composition of the molecular sieve (Table 6), the number of templates used to balance the charge of the skeleton in each cage Ma, the number of templates that do not participate in the balance of the skeleton charge in each cage Mb, and the maximum amount of Si (4Al) can be calculated as shown in Table 7. a The level of silicon incorporation is defined as the molar ratio of (Si/(Si + Al + P)solid)/(Si/(Si + Al + P)gel).

Ma:
The number of templates used to balance the charge of the skeleton in each cage.Mb: The number of templates that do not participate in the balance of the skeleton charge in each cage.
Although the total amount of templating agent in the molecular sieve cage followed the order S1 > S3 (Table 2), as the content of silicon in the skeleton increased, the value of Ma increased.The Mb of sample S-3 was 0, indicating that the templating agent in the sample cage of the molecular sieve was used to balance the skeleton charge.The proportions of the different template agents used to balance the charge of the framework followed the order S3 > S1, using the SM2 or SM3 substitution mechanism [26,28,29].The negative ion of the skeleton was generated at the same time as the formation of the silicon island structure [28,30], so the more templating agents used to balance the skeleton charge and the greater the relative content of the Si structure distribution in the silicon island structure, the stronger the acidity exhibited by the molecular sieve was, which is consistent with the results presented in Tables 4 and 5.When the silicon content in the framework of the zeolite exceeded the maximum Si(4Al) content, other silicon structures began to appear.It can be seen from Table 6 that the maximum Si(4Al) content of the samples synthesized by the composite template agent was significantly higher than that of the samples synthesized by the single template agent.The amount of framework Si(4Al) per mole (Al + Si + P) was determined to be 0.0928 and because the Si(4Al) structure generates the B acid center, the amount of B acid of the composite template agent in the synthetic sample S-3 was also greater than that in S-1.

The Catalytic Performance of the Catalysts
The catalytic performance of these SAPO-34s synthesized was evaluated by taking the MTO as the model reaction under the conditions of 723 K and 3 hr −1 WHSV, as was described in detail in Section 3. At the initial stage of the reaction, methanol conversion reached 100% and short chain olefin selectivity was high.
As is well known, the main problem for the MTO process in the SAPO-34 sieve is the deactivation of carbon deposits.In this study, samples S-3, S-4, and S-5 synthesized with the TEA/Mor/TEAOH system showed improved stability in this reaction.However, the S-6 synthesized by the three templating agents had a short lifetime.This phenomenon indicates that the use of a composite templating agent is beneficial to prolong the lifetime of the catalyst, but the relatively low silicon to aluminum ratio is detrimental to the stability of the catalyst.This can be attributed to three different factors.First, the small particle molecular sieve pores are short, which promotes the participation of all active centers in the molecular sieve and delays the production of inert carbon deposits.Second, the specific surface area of the molecular sieve is S-5 > S-4 > S-6, and the specific surface area and pore volume of the molecular sieve are large.Thus, it is easy for it to fully utilize its own active center.Third, as indicated by 29 Si MAS NMR and FT-IR characterization, there is a stronger acid center in the S-5 sample.MTO side reactions tend to occur at the acidic center of the molecular sieve.Due to the mismatch with the pore size of the SAPO-34 molecular sieve, the macromolecular product formed by the side reaction could not escape the pore structure in time, and the reaction site of the acid layer accumulated in the pore channel, causing the catalyst to be deactivated within a short time period.Furthermore, Sample S-5 had a much higher selectivity to short chain olefins (ethylene and propylene) throughout the reaction test, giving values above 89.15%.S-1 and S-2 had poor selectivity to light olefins during the reaction because the number of B acid centers was small.It is known from the MTO reaction mechanism that the first step of methanol molecule loss to form dimethyl ether occurs in the B acid.The amount of B acid distributed was S-3 > S-2 > S-1; with the increase in the B acid center of the molecular sieve, more surface oxygen groups were formed, which promoted the selectivity of the low-carbon olefins (Figures 8 and 9).

Ma:
The number of templates used to balance the charge of the skeleton in each cage.Mb: The number of templates that do not participate in the balance of the skeleton charge in each cage.

The Catalytic Performance of the Catalysts
The catalytic performance of these SAPO-34s synthesized was evaluated by taking the MTO as the model reaction under the conditions of 723 K and 3 hr −1 WHSV, as was described in detail in Section 3. At the initial stage of the reaction, methanol conversion reached 100% and short chain olefin selectivity was high.
As is well known, the main problem for the MTO process in the SAPO-34 sieve is the deactivation of carbon deposits.In this study, samples S-3, S-4, and S-5 synthesized with the TEA/Mor/TEAOH system showed improved stability in this reaction.However, the S-6 synthesized by the three templating agents had a short lifetime.This phenomenon indicates that the use of a composite templating agent is beneficial to prolong the lifetime of the catalyst, but the relatively low silicon to aluminum ratio is detrimental to the stability of the catalyst.This can be attributed to three different factors.First, the small particle molecular sieve pores are short, which promotes the participation of all active centers in the molecular sieve and delays the production of inert carbon deposits.Second, the specific surface area of the molecular sieve is S-5 > S-4 > S-6, and the specific surface area and pore volume of the molecular sieve are large.Thus, it is easy for it to fully utilize its own active center.Third, as indicated by 29 Si MAS NMR and FT-IR characterization, there is a stronger acid center in the S-5 sample.MTO side reactions tend to occur at the acidic center of the molecular sieve.Due to the mismatch with the pore size of the SAPO-34 molecular sieve, the macromolecular product formed by the side reaction could not escape the pore structure in time, and the reaction site of the acid layer accumulated in the pore channel, causing the catalyst to be deactivated within a short time period.Furthermore, Sample S-5 had a much higher selectivity to short chain olefins (ethylene and propylene) throughout the reaction test, giving values above 89.15%.S-1 and S-2 had poor selectivity to light olefins during the reaction because the number of B acid centers was small.It is known from the MTO reaction mechanism that the first step of methanol molecule loss to form dimethyl ether occurs in the B acid.The amount of B acid distributed was S-3 > S-2 > S-1; with the increase in the B acid center of the molecular sieve, more surface oxygen groups were formed, which promoted the selectivity of the low-carbon olefins (Figures 8 and 9).

Synthesis of SAPO-34 Molecular Sieve
Crystalline SAPO-34 was synthesized hydrothermally using different combinations of TEAOH, TEA, and Mor as the template.The sources of Al, P, and Si were boehmite, phosphoric acid, and tetraethyl silicate, respectively.The molar composition of the reaction solution and the synthesis conditions for different SAPO-34 samples are given in Table 8.The structure and pKa of the templating agent used in the experiment are shown in Table 9.In a typical synthesis, the source of Al, P, or Si was added into a 250 mL Teflon-lined stainless-steel autoclave.After mixing, the templates and deionized water were added, and the synthesis solution was stirred for 2 h at 30 °C.Then the autoclave was heated at 200 °C for 48 h.The solid product was separated by centrifugation, washed with distilled water several times until neutral, and dried at 100 °C overnight.Finally, the catalyst sample was calcined at 550 °C for 6 h to remove the organic template.

Synthesis of SAPO-34 Molecular Sieve
Crystalline SAPO-34 was synthesized hydrothermally using different combinations of TEAOH, TEA, and Mor as the template.The sources of Al, P, and Si were boehmite, phosphoric acid, and tetraethyl silicate, respectively.The molar composition of the reaction solution and the synthesis conditions for different SAPO-34 samples are given in Table 8.The structure and pKa of the templating agent used in the experiment are shown in Table 9.In a typical synthesis, the source of Al, P, or Si was added into a 250 mL Teflon-lined stainless-steel autoclave.After mixing, the templates and deionized water were added, and the synthesis solution was stirred for 2 h at 30 • C. Then the autoclave was heated at 200 • C for 48 h.The solid product was separated by centrifugation, washed with distilled water several times until neutral, and dried at 100 • C overnight.Finally, the catalyst sample was calcined at 550 • C for 6 h to remove the organic template.

Synthesis of SAPO-34 Molecular Sieve
Crystalline SAPO-34 was synthesized hydrothermally using different combinations of TEAOH, TEA, and Mor as the template.The sources of Al, P, and Si were boehmite, phosphoric acid, and tetraethyl silicate, respectively.The molar composition of the reaction solution and the synthesis conditions for different SAPO-34 samples are given in Table 8.The structure and pKa of the templating agent used in the experiment are shown in Table 9.In a typical synthesis, the source of Al, P, or Si was added into a 250 mL Teflon-lined stainless-steel autoclave.After mixing, the templates and deionized water were added, and the synthesis solution was stirred for 2 h at 30 °C.Then the autoclave was heated at 200 °C for 48 h.The solid product was separated by centrifugation, washed with distilled water several times until neutral, and dried at 100 °C overnight.Finally, the catalyst sample was calcined at 550 °C for 6 h to remove the organic template.

Synthesis of SAPO-34 Molecular Sieve
Crystalline SAPO-34 was synthesized hydrothermally using different combinations of TEAOH, TEA, and Mor as the template.The sources of Al, P, and Si were boehmite, phosphoric acid, and tetraethyl silicate, respectively.The molar composition of the reaction solution and the synthesis conditions for different SAPO-34 samples are given in Table 8.The structure and pKa of the templating agent used in the experiment are shown in Table 9.In a typical synthesis, the source of Al, P, or Si was added into a 250 mL Teflon-lined stainless-steel autoclave.After mixing, the templates and deionized water were added, and the synthesis solution was stirred for 2 h at 30 °C.Then the autoclave was heated at 200 °C for 48 h.The solid product was separated by centrifugation, washed with distilled water several times until neutral, and dried at 100 °C overnight.Finally, the catalyst sample was calcined at 550 °C for 6 h to remove the organic template.

Synthesis of SAPO-34 Molecular Sieve
Crystalline SAPO-34 was synthesized hydrothermally using different combinations of TEAOH, TEA, and Mor as the template.The sources of Al, P, and Si were boehmite, phosphoric acid, and tetraethyl silicate, respectively.The molar composition of the reaction solution and the synthesis conditions for different SAPO-34 samples are given in Table 8.The structure and pKa of the templating agent used in the experiment are shown in Table 9.In a typical synthesis, the source of Al, P, or Si was added into a 250 mL Teflon-lined stainless-steel autoclave.After mixing, the templates and deionized water were added, and the synthesis solution was stirred for 2 h at 30 °C.Then the autoclave was heated at 200 °C for 48 h.The solid product was separated by centrifugation, washed with distilled water several times until neutral, and dried at 100 °C overnight.Finally, the catalyst sample was calcined at 550 °C for 6 h to remove the organic template.

Characterization of Samples
The X-ray diffraction (XRD) patterns of catalysts were obtained on a powder X-ray diffractometer (D8FOCUS, Bruker, Ettlingen, Germany) using Cu Kα radiation.The crystallinity was calculated

Figure 2 .
Figure 2. The SEM images of the prepared SAPO-34 samples.

Figure 2 .
Figure 2. The SEM images of the prepared SAPO-34 samples.

Figure 3 .
Figure 3.The TG-DTG curves for samples with different templates, a: TG curve, b: DTG curve.

Figure 6 .
Figure 6.The FT-IR images for different samples at 300 K: (a) before NH 3 absorption, (b) after NH 3 absorption.

Figure 7 .
Figure 7.The 29 Si MAS NMR spectra for samples synthesized by different templates.

Figure 7 .
Figure 7.The 29 Si MAS NMR spectra for samples synthesized by different templates.

Figure 8 .
Figure 8.The ethylene and propylene selectivity of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Figure 8 .
Figure 8.The ethylene and propylene selectivity of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Figure 9 .
Figure 9.The product distribution of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Table 9 .Figure 9 .
Figure 9.The product distribution of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Figure 9 .
Figure 9.The product distribution of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Table 9 .Figure 9 .
Figure 9.The product distribution of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Table 9 .Figure 9 .
Figure 9.The product distribution of the MTO reaction on SAPO-34 samples at 723 K and a WHSV of 3 h −1 .

Table 1 .
The structural properties and surface compositions of the catalysts.

Table 1 .
The structural properties and surface compositions of the catalysts.

Table 2 .
The TG analysis results for samples synthesized by different templates.

Table 2 .
The TG analysis results for samples synthesized by different templates.

Table 3 .
The acid distribution for samples synthesized with different gel compositions.

Table 3 .
The acid distribution for samples synthesized with different gel compositions.

Table 4 .
The B and L acid distribution of zeolite samples.

Table 4 .
The B and L acid distribution of zeolite samples.

Table 5 .
The distribution of silicon environments for samples synthesized by different templates.

Table 6 .
The element composition and relative crystallinity of the sample with different templates.

Table 5 .
The distribution of silicon environments for samples synthesized by different templates.

Table 6 .
The element composition and relative crystallinity of the sample with different templates.

Table 7 .
The net charge, the template number per cage and the maximum Si (4Al) contents.

Table 8 .
The gel composition and experimental synthesis conditions for the different SAPO-34 materials.

Table 8 .
The gel composition and experimental synthesis conditions for the different SAPO-34 materials.

Table 9 .
[31]structure of the three templating agents and the values of pKa[31].

Table 8 .
The gel composition and experimental synthesis conditions for the different SAPO-34 materials.

Table 8 .
The gel composition and experimental synthesis conditions for the different SAPO-34 materials.

Table 8 .
The gel composition and experimental synthesis conditions for the different SAPO-34 materials.