Nonclassical from-shell-to-core growth of hierarchically organized SAPO-11 with enhanced catalytic performance in hydroisomerization of n-heptane

Abstract

Integrating hierarchical porosity over microporous zeotype materials is an effective way to promote their mass transfer properties and catalytic performances. A combined synthetic strategy using small molecular growth inhibitor 1,2,3-hexanetriol and tumbling crystallization condition to generate hierarchically organized SAPO-11 is herein presented. The addition of 1,2,3-hexanetriol in the synthetic gel of SAPO-11 under agitating conditions significantly altered its crystallization behaviour, resulting in the formation of a hierarchically organized architecture. An underlying nonclassical from-shell-to-core crystallization has been disclosed by time-dependent observation of the formation process. The hierarchically self-organized structure has been characterized by a suite of characterization techniques, such as XRD, N₂ physisorption, SEM, TEM, mercury intrusion measurements, ²⁷Al, ²⁹Si, ³¹P MAS NMR and pyridine adsorption IR (Py-IR). The structure featuring barrel-shaped architecture is comprised of aligned 300–400 nm primary building blocks with voids in between, constructing an auxiliary macro-/meso-pore system open to external surfaces. The catalytic performance of Pt supported on hierarchical SAPO-11 in n-heptane hydroisomerization has been assessed, showing that both catalytic activity and isomer yield have been increased with respect to a conventional sample. As the acidity for the hierarchical SAPO-11 is comparable to the conventional sample, the enhancement in catalytic performance is attributed to the small primary crystal size and macro-/meso-pore-connectivity, that are important for mass transfer.

---

Introduction

Zeolites and zeotype materials consist of an important family of microporous (<2.0 nm) solids that find a broad scope of applications including catalysts, sorbents or ion-exchangers in petrochemical and chemical processes.^1,2 Zeotype materials are distinct in terms of their well-defined channel/cavity systems that can selectively accommodate molecules with respect to their size and shape, henceforth, they are also termed as molecular sieves.² Such a capability also endows them unique shape-selectivity in heterogeneous catalysis. Among them, silicoaluminophosphates (SAPOs) represented by SAPO-11 (AEL topology, mainly used as catalyst for n-paraffin hydroisomerization in petrochemical industry^3,4) and SAPO-34 (CHA topology, employed as methanol-to-olefins catalyst in practice⁵), constitute a subclass of zeotype materials that are important as industrial solid-acid catalysts.⁶ The acidity of SAPOs stems from the P or Al + P pairs substitution by Si into the otherwise electron-neutral aluminophosphate (AlPO) frameworks, which incurs a negatively charged Si site that needs to be charge balanced by a proton, thereby imparts the material with Brønsted acidity.^6,7 Si incorporation can occur via SM2, SM3 or their combined mechanism, leading to diversified Si incorporation and acidity.⁷ The acidic properties for SAPOs and their catalytic performances are highly dependent on the amount, location and distribution of Si atoms.^7,8 Synthetic conditions, such as structure-directing agents,^9,10 additives,^11,12 crystallization conditions,^13,14 can influence Si incorporation and affect the acidic properties accordingly.

Besides acidity, the diffusion properties of SAPOs are also important for their catalytic performance. Synthetic SAPOs are micron-sized crystals that suffer from diffusion limitations in an analogous way to their aluminosilicate cousin, zeolites. Indeed, downsizing crystal size (i.e., generating nanozeolites^15,16), introducing auxiliary macro-/meso-porosity (i.e., synthesizing hierarchical zeolites^17,18), and controlling crystal shapes (for instance, preparing 2-dimensional zeolites^19–21) have been found to enhance the catalytic performances of zeotype materials as a result of promoted diffusion properties. Compared with protocols to generate hierarchical zeolites for most frequently employed aluminosilicates, only limited successful methods for preparing hierarchical SAPOs have been advanced. Hard template route using carbon blacks,²² post-synthetic acid/base treatment,^23,24 and soft-templating routes^{3,4,10,19,25–27} have recently been reported. Notably, Ryoo et al.¹⁹ have used bifunctional multiamines with amphiphilic structures to generate hierarchical AEL, ATO and AFI SAPOs, cobalt aluminophosphate and gallium phosphate. The SAPO-11 thus obtained has the precisely controlled 2 nm thickness along the 1D channels in the c-axis, resulting in an improved C₇ isomer yield in hydroisomerization of n-heptane. Most of the known synthetic strategies involve either the use of expensive or home-made reagents, as well as tedious procedures for their preparations. From a catalytic viewpoint, it is important to find ways that enable tuning the hierarchical pore structure while retaining the acidity, as both factors can cooperatively influence catalytic behaviours.³ Inappropriate preparation may lead to an excessive distribution of acid sites at the external crystal surfaces that causes a detrimental effect to the isomer selectivity, as found by Choi et al.³ Some synthetic strategies may adversely affect the integrity of SAPO framework and cause the loss of crystallinity and associated acidity when mesopores are introduced.

Among numerous catalytic materials applicable as an acidic catalyst for n-paraffin isomerization, SAPO-11 as a support for bifunctional catalyst exhibits outstanding performance.^28–30 SAPO-11 can attain high yields of mono-branched isomers, and simultaneously suppress multi-branched isomers that are susceptible to cracking due to its unique pore shape (10 member ring, one dimensional 4.0 × 6.5 Å pore) and mild acidity.^31–33 Although SAPO-11 has been deployed for catalytic isomerization of n-paraffins in industry, recently studies have demonstrated that there is still space to promote the isomer selectivity by using hierarchical or nanosized SAPO-11. Besides, the isomerization and cracking reaction are consecutive and highly branched alkanes are more susceptible to cracking.³⁴ Branched iso-paraffins have larger kinetic diameters than linear n-paraffins, which means that the diffusion of iso-paraffins should be slower than that of n-paraffins. Henceforth, the long retention time of iso-paraffins in the micropores of SAPO-11 can increase the possibility of further isomerization and consecutive cracking. Choi et al.³ observed that secondary mesoporosity can facilitate the diffusion of iso-paraffins out of the SAPO-11 micropores and can therefore suppress the consecutive cracking reaction. Bao et al.³⁵ reported a method to hydrothermally synthesize SAPO-11 with hierarchically porous structure by using a tetradecylphosphoric acid (TDPA) template and a conventional di-n-propylamine (DPA) template, and the hierarchical SAPO-11 synthesized can efficiently restrain the hydrocarbon cracking on medium and strong acid sites. Bértolo et al.³⁶ reported that carbon templated mesoporous SAPO-11 can improve the catalytic conversion of n-decane hydroisomerization. Xiao et al.³⁷ recently showed that a hierarchical SAPO-11 synthesized in the presence of polyhexamethylenebiguanidine (PHMB) soft template exhibits higher activity, better isomer selectivity, and lower cracking product selectivity in n-dodecane hydroisomerization. Overall, the benefits of hierarchical structured SAPO-11 are high activity and isomer selectivity, as well as low cracking probability.

Previously, hierarchical porosity has been mainly achieved by using polymeric type³⁷ or surfactant type poregens,³⁸ and less has been explored to introduce porosity with small molecular growth modifiers. In the present work, we report the synthesis, formation mechanism, and catalytic property of hierarchically organized SAPO-11 in the presence of a small molecular growth modifier, 1,2,3-hexanetriol under agitating conditions. 1,2,3-Hexanetriol has been recently proposed as a growth modifier for zeolite L (LTL topology) by Rimer et al.,³⁹ but no report on its effect over the growth of SAPO-11 is known, to the best of our knowledge. By revealing the details of its effect on structure features, texture properties, nucleation and crystallization process, acidic properties with a combination of characterization techniques (such as XRD, SEM, TEM/SAED, N₂-physisorption, mercury intrusion, ²⁷Al, ³¹P, ²⁹Si MAS NMR, pyridine adsorption IR (Py-IR)), we intend to understand the role of additive on crystallization process, texture and acidic properties. We will show that 1,2,3-hexanetriol can modify the crystallization process to a from-shell-to-core growth route under agitating, and can substantially influence Si incorporation and mesoscale structure of SAPO-11. The catalytic behaviour in hydroisomerization of n-heptane will also be explored and compared with a reference sample derived from conventional hydrothermal synthesis.

Experimental section

Synthesis

Synthesis of SAPO-11 in the presence of 1,2,3-hexanetriol (BIOXTRA, ≥98.0 wt%) was conducted under agitation. In a typical synthesis, 1.36 g of pseudoboehmite (PB, 75 wt% Al₂O₃, Aluminium Corporation of China) was hydrolysed in 21.25 g of deionized water under stirring for 4 h to afford a mixture, to which 2.30 g of phosphoric acid (≥85.0 wt%, Shanghai Lingfeng Chemical Reagent Co. Ltd.) was added and homogenized under stirring for 8 h. 0.64 g of tetraethyl orthosilicate (TEOS, ≥28.0 wt% SiO₂, Shanghai Lingfeng Chemical Reagent Co. Ltd.) was subsequently added to the mixture, followed by the addition of 1.50 g of dipropylamine (DPA, >99 wt%, TCI) and 0.14 g of 1,2,3-hexanetriol. The mixture was further homogenized by stirring for an additional 5 h. The final molar composition of the synthesis gel was 1.2DPA : 1.0Al₂O₃ : 1.0P₂O₅ : 0.3SiO₂ : 40H₂O : 0.001 1,2,3-hexanetriol. The synthetic gel was transferred into a Teflon-lined stainless steel autoclave and hydrothermally crystallized at 453 K for 4 days under tumbling (50 rpm). The resultant product was filtered off, washed with deionized water, and dried at 393 K for 5 h. The as-synthesized SAPO-11 was calcined under a continuous flow of dry air at 823 K for 15 h. The product was denoted as SAPO-11-H, whereby H represents hierarchical.

A conventional sample for comparison was synthesized under static conditions as frequently used. The molar composition of the synthesis gel was 1.2DPA : 1.0Al₂O₃ :1.0P₂O₅ : 0.3SiO₂ : 40H₂O. A typical synthesis procedure was detailed as follows: 1.36 g of pseudoboehmite and 2.30 g of phosphoric acid were hydrolysed in 21.25 g of deionized water under stirring for 4 h. 0.64 g of TEOS was subsequently added to the mixture and stirred vigorously for 2 h. Then 1.50 g of DPA was slowly added and the mixture was stirred for 8 h. The synthetic gel was hydrothermally crystallized at 453 K for 4 days under static conditions. The as-synthesized SAPO-11 was filtered off, washed with deionized water, and dried at 393 K for 5 h. Subsequently, the sample was calcined under a continuous flow of dry air at 823 K for 15 h. The obtained product was denoted as SAPO-11-C, with suffix C standing for conventional.

Crystallization mechanism studies were carried out under identical crystallization conditions with the same recipe as for SAPO-11-H. The same synthetic gel for the time-dependent crystallization was divided into equal parts and charged into autoclaves with identical volume before crystallization started. Each distinct autoclave was opened at a designated crystallization time, solids out of which were sampled, separated via filtration and characterized to understand the structure evolution over crystallization time.

Catalyst preparation

The Pt/SAPO-11-C and Pt/SAPO-11-H catalysts were prepared by incipient wetness impregnation of SAPO-11 with an aqueous solution of H₂PtCl₆, followed by drying and then calcination in an air flow (100 cm³ min⁻¹) at 673 K for 3 h. The Pt content of each catalyst was 0.5 wt%. Previous reports by Maesen et al.⁴⁰ and Höchtl et al.⁴¹ have shown that such a loading will be sufficient to certify that the isomerization reaction is unaffected by dehydrogenation process on metal sites.

Characterization techniques

The phase structure of the SAPO-11 was characterized by powder X-ray diffraction (XRD) patterns recorded on a Rigaku D/Max 2550 VB/PC diffractometer operating at 40 kV and 100 mA with Cu-Kα radiation (λ = 1.5418 Å) as the X-ray source. The patterns were collected in the 2θ range from 3 to 50°, with a scanning speed of 8° min⁻¹. The size and morphological features of the samples were determined by means of scanning electron microscopy (SEM) images recorded with a NOVA Nano SEM450 microscope. Transmission electron microscope (TEM) images were obtained from a JEM-2011 (JEOL) electron microscope setup. A drop of the examined solution was placed on a TEM grid covered by a perforated carbon film. Nitrogen adsorption–desorption isotherms were measured on an ASAP 2020 (Micromeritics, USA) analyzer at 77 K, after samples were outgassed at 623 K under vacuum for 6 h. The surface areas and pore volumes were determined by the Brunauer–Emmett–Teller (BET) method, and non-local density functional theory (NLDFT) method, respectively. Micropore volumes were derived from a t-plot approach, and the total pore volume values were estimated by the adsorption at a relative pressure P/P₀ = 0.99. A Micromeritics Autochem 2920 was used for chemisorption measurements. The exposed active surface areas of the catalysts were determined via CO titration at room temperature. In each experiment, 0.10 g of catalyst was used. The catalyst was heated to 673 K in 10% H₂/Ar with a flow rate of 30.0 mL min⁻¹, ramped by a step of 5 K min⁻¹ and kept at 673 K for 2 h. After reduction, the catalyst was purged with an ultra-high-pure helium flow before being cooled to room temperature. CO pulses were charged over the reduced catalyst and the CO uptake in each pulse was monitored using a thermal conductivity detector (TCD). Solid-state NMR measurements were performed on an Agilent DD2-500 MHz spectrometer operated at a magnetic field strength of 11.7 T. ²⁷Al MAS NMR spectra were recorded at 130.2 MHz with a spinning rate of 13 kHz, 200 scans, and 2 s recycle delay. The chemical shifts were referenced to 1% Al(NO₃)₃ aqueous solution. ³¹P MAS NMR experiments with high power proton decoupling were conducted at 202.3 MHz with a spinning rate of 14 kHz, 20 scans and 10–60 s recycle delay. The chemical shifts were referenced to 85% H₃PO₄. ²⁹Si MAS NMR spectra with high power proton decoupling were recorded at 99.3 MHz with a spinning rate of 4 kHz, 600 scans and 10–300 s recycle delay. Elemental analyses were measured by a XRF-1800 Sequential X-ray Fluorescence Spectrometer (Shimadzu). The acidity of SAPO-11 was probed by pyridine absorption infrared (Py-IR) spectra on a Spectrum 100 FT-IR spectrometer (Nicolet Co., USA). All samples were pressed into self-supporting wafers (diameter: 1.4 cm, weight: 166 mg) and were pre-heated at 723 K for 4 h under vacuum (1.3 × 10⁻² Pa). The background baselines of the samples were collected and subtracted at room temperature. The amount of adsorbed probe molecules was calculated from the integrated area of given bands with their distinct molar extinction coefficients (ε_Brønsted(1540 cm⁻¹) = 6.8 cm μmol⁻¹, ε_Lewis(1450 cm⁻¹) = 4.4 cm μmol⁻¹) that are given in literature.⁴² Mercury intrusion tests were performed by using an AutoPore IV 9500 instrument. The intrusion volumes were measured at stepwise increasing pressures equilibrating at each pressure step. The pore size distribution was calculated according to the intrusion curves.

Catalytic assessments

The hydroisomerization of n-heptane was carried out in a flow fixed-bed reactor at atmospheric pressure. The catalyst loading was 0.3 g. Prior to testing, the catalyst was reduced in H₂ flow at 673 K for 2 h, then cooled down to the reaction temperature under H₂ flow. n-Heptane was introduced to the reactor by a carrier gas of H₂ (30 cm³ min⁻¹) flowing through a saturator maintained at 288 K, which gave a H₂/n-heptane molar ratio of 27. The weight hourly space velocity (WHSV) of n-heptane was 0.9 h⁻¹. The product gas mixture was analyzed periodically on-line with a gas chromatograph (GC, Shanghai Huaai Chromatograph Analysis Co. Ltd.) equipped with a flame ionization detector (FID) and a HP-PONA capillary column (50 m × 0.2 mm × 0.5 μm).

Results and discussion

Structural features of SAPO-11-H

The XRD pattern for SAPO-11-H is shown in Fig. 1, along with reference SAPO-11-C. Samples synthesized with or without 1,2,3-hexanetriol present the same diffraction patterns with reflection peaks at 2θ = 8.1, 9.4, 13.1, 15.6, 20.3, and 22.1–23.2°, which can be attributed to the typical SPAO-11 with AEL structure (JCPDS no. 42-0428).^43,44 The absence of additional diffraction lines indicates that the samples are free of impurities, while no lumps from amorphous raw materials suggest that all the starting materials have been converted. The sharp diffraction peaks also manifest that SAPO-11-H is highly crystalline in nature, i.e., the introduction of 1,2,3-hexanetriol has negligible effect on the integrity of the AEL framework.

Fig. 1 XRD patterns of SAPO-11-H and SAPO-11-C.

N₂ adsorption–desorption isotherms for SAPO-11-C and SAPO-11-H are displayed in Fig. 2a and b, respectively. The adsorption–desorption isotherm for SAPO-11-C can be classified as type I, according to IUPAC classifications, from which the micropore-filling at low relative pressures (P/P₀ < 0.1) can be clearly seen. In addition, an uptake of N₂ at P/P₀ from 0.9 to 1.0 is ascribed to the presence of voids as a result of zeolitic particulates aggregation,³ i.e., SAPO-11-C contains mainly micropores. For SAPO-11-H, in addition to the micropore-filling process, an obvious jump in N₂ uptake out of capillary condensation effect at P/P₀ > 0.85 can be distinguished, which is caused by the presence of additional mesopores (Fig. 2b). Therefore, isotherm for SAPO-11-H is categorized as a hybrid of type I and type IV, that is often found for hierarchical SAPO-11.^35,37,38,45 The corresponding pore size distribution for SAPO-11-H deduced from the NLDFT method is exhibited in the inset of Fig. 2b, showing a rather broad pore-size distribution ranging from 10 to 60 nm. The BET, external and internal surface area values and pore volume data are tabulated in Table 1. The micropore volumes (0.06 and 0.08 cm³ g⁻¹, for SAPO-11-C and SAPO-11-H, respectively) for the two samples are comparable, indicating that the introduction of mesopores has no obvious effect on the inherent microporosity of AEL topology. The external surface area and mesopore volume for SAPO-11-C are 218 m² g⁻¹ and 0.29 cm³ g⁻¹, respectively, whereas SAPO-11-H shows a much higher external surface area of 316 m² g⁻¹ and a larger mesopore volume (0.44 cm³ g⁻¹). Complementary mercury intrusion experiments were carried out to detect the existence of even larger pores that is beyond the applicability of N₂ physisorption.⁴⁶ The obtained pore size distributions are plotted in Fig. 2c and d, for SAPO-11-C and SAPO-11-H, respectively. The large macropores within micron scale (>1 μm) are attributed to suspended particles in mercury during the measurements.⁴⁷ For SAPO-11-C, the mean large pore size appears at ca. 10 μm, in addition to a rather diverse pore size range from meso- to macro-pore scope, implying that the SAPO-11-C is far less ordered in terms of aggregation caused pore-size-distribution. On the other hand, the mercury intrusion measurements show that SAPO-11-H possesses only two major types of pores, a mean pore-size centered at ca. 1 μm and a small peak centered at ∼10 nm. The former is attributed to the pores between organized crystals of SAPO-11-H building blocks (cf. SEM and TEM images below), while the latter belongs to mesopores region that is consistent with N₂ physisorption data. Texture property measurements clearly show the hierarchically porous structure of SAPO-11-H, and the auxiliary pores consist of both macro- (>50 nm) and meso- (2–50 nm) pores.

Fig. 2 N₂ adsorption/desorption isotherms of SAPO-11-C (a) and SAPO-11-H (b), and the corresponding pore size distribution for SAPO-11-H (inset of b). Mercury intrusion pore size distribution of SAPO-11-C (c) and SAPO-11-H (d).

Table 1 Texture properties of the calcined SAPO-11-C and SAPO-11-H derived from N₂ physisorption measurements

Sample	SBET^a/m^2 g^−1	Smicro^b/m^2 g^−1	Sext^c/m^2 g^−1	Vtotal/cm^3 g^−1	Vmicro^c/cm^3 g^−1	Vmeso^c/cm^3 g^−1	Pt dispersion^d (%)	Molar composition^e
a Calculated by the BET method in the P/P0 range of 0.05–0.25.b Inferred using the t-plot method.c Deduced using the t-plot method.d Pt dispersion was determined by a pulse chemisorption after supporting 0.5 wt% Pt.e Molar composition is obtained by XRF.
SAPO-11-C	218	117	101	0.35	0.06	0.29	7.7	Si0.19Al1.16P0.79O2
SAPO-11-H	316	168	148	0.52	0.08	0.44	13.5	Si0.19Al1.09P0.96O2

---

SEM images for SAPO-11-C and SAPO-11-H are shown in Fig. 3a–c and d–f, respectively. The hydrothermally synthesized SAPO-11-C consists of spherical particulates of 10–30 μm (Fig. 3a–c), and surfaces of the spheres are rough. With high magnification images, it is found that the large spherical particulates are made up of irregularly shaped subunits between 300 and 500 nm. The grain boundaries between these building blocks are hardly discernible, as they intergrow into neighbouring ones. The same morphology has been observed by other groups for DPA derived SAPO-11.⁴⁷ The close packing of irregularly shaped building blocks for SAPO-11-C is in line with the N₂ physisorption data, as no obvious mesopores have been measured (Fig. 2a–c). It is also speculated that the diverse pore-size-distribution for smaller pores (below 10 μm) detected by mercury intrusion can be caused by the stacking of building blocks, whereas the mean pore size appeared at ca. 10 μm can be attributed to the voids between large spherical particulates. SAPO-11-H is composed of much smaller (5–6 μm) barrel-shaped particulates with uniform sizes (Fig. 3d–f), in addition to some detached debris with smaller sizes. When we zoom in to observe the surface of the big particulates, it is intriguing to find that the sides of the barrel-shaped particulates are built up of 100 × 300 nm to 200 × 400 nm brick-shaped building blocks. These building units are not irregularly stacked with neighbouring ones, but are aligned along the same crystallographic axis. Edges, corners and shapes of these building blocks are well defined. Between these bricks, clear void spaces can be intuitively seen, which could constitute the porous regions that have been observed by N₂ physisorption and mercury intrusion measurements. It is noteworthy that the voids are directly connected to the external surfaces, as only this type of auxiliary pores can bring about improved mass transfer.^48,49 As the AEL structure has an orthorhombic Imma structure, the unidimensional channel runs along the [100] direction that is normal to the barrel top. The aspect ratio for each brick can be inferred as ca. 1.2, indicating that the channel path along the [100] axis has been substantially shortened in the brick-shaped building units. Another advantage of such low aspect ratio is that most external surfaces are covered by pore openings, which is important for pore entrance of guest molecules.

Fig. 3 SEM images for SAPO-11-C (a–c) and SAPO-11-H (d–f).

The intergrowth, particle size and pore structure of the inner part of the SAPO-11-H crystal can be probed by slicing crystals open, which permits us to observe the interior of the sample. The cryo-TEM images thus obtained are displayed in Fig. 4a and b for SAPO-11-C, in which it is seen that the primary particles of building blocks appear as dark field are found to be 300 to 500 nm which are close to the sizes of the external building blocks. For SAPO-11-H (Fig. 4c and d), dark field of building blocks with sizes vary from 100 × 200 nm to 200 × 400 nm can be visualized, corroborating that the inner part of SAPO-11-H are also composed of similar building blocks as for the external parts (Fig. 3f). The selected area electron diffraction (SAED) patterns (insets of Fig. 4b and d) show that both samples are single crystal-like, indicating that their building blocks are mutually aligned with neighbouring ones. Between the dark building blocks, bright areas constituting the auxiliary macropores are found all over the sample, showing the presence of large numbers of porous regions. In contrast, there are no obvious bright fields that are visible in the SAPO-11-C sample (Fig. 4a and b). Overall, SEM and cryo-TEM images disclose that SAPO-11-H is made up of organized nanobuilding blocks, between which porous regions are found, in good agreement with N₂ physisorption and mercury intrusion measurements.

Fig. 4 TEM images for SAPO-11-C (a and b) and SAPO-11-H (c and d) after calcination.

XRF measurements were conducted to compare the chemical composition of samples, as compiled in Table 1. The Si content in our samples are ca. 5.64 wt% and 5.55 wt%, for SAPO-11-C and SAPO-11-H, respectively. The theoretical Si content of SAPO-11 deduced from batch composition is 6.18%. Given the error of semi-quantitative XRF technique, the result of element composition is within the acceptable range. The yield of the present method is above 90% on inorganic basis, suggesting that most inorganic raw materials have been converted into the framework of the product.

NMR measurements were carried out to analyse chemical environments for framework atoms. The results for calcined SAPO-11-C and SAPO-11-H are shown in Fig. 5. In ²⁷Al MAS NMR spectra (Fig. 5a and b), strong peaks at 39 and 30 ppm are present in both samples, and are assignable to tetrahedrally coordinated framework aluminium atoms.⁵⁰ Octahedrally coordinated aluminium atoms appearing at −14 ppm are formed by an additional coordination of two water molecules to tetrahedrally coordinated framework aluminium atoms.⁵⁰ The weak signal at 8 ppm indicates the presence of a small amount of pentacoordinated aluminium atoms formed by an additional coordination of one water molecule to the tetrahedrally coordinated aluminium species.⁵⁰ In ³¹P MAS NMR spectra (Fig. 5c and d), the strong peak at −30 ppm is assigned to tetrahedrally coordinated phosphorus atoms bound to four aluminium atoms in the first coordination sphere of T atoms (P(OAl)₄).⁵⁰ Meanwhile, the significantly weaker signal at −23 ppm is attributed to phosphorus atoms which are additionally coordinated to a number of y water molecules (P(OAl)_x(H₂O)_y, with x = 4)⁵¹ or y water molecules instead of aluminium atoms (P(OAl)_x(H₂O)_y, with x = 4 − y).⁵²

Fig. 5 ²⁷Al MAS NMR spectra of SAPO-11-C (a) and SAPO-11-H (b), ³¹P MAS NMR spectra of SAPO-11-C (c) and SAPO-11-H (d), ²⁹Si MAS NMR spectra of SAPO-11-C (e) and SAPO-11-H (f).

Unlike aluminosilicates (zeolites) whose acid sites density and strength can be directly correlated to their chemical composition (Si/Al ratios), the acid strength and sites density for SAPOs is dictated by the way in which Si is incorporated into the aluminophosphates framework.^{26,53–55 29}Si MAS NMR has been proven to be a powerful technique to discriminate the chemical environment for Si in SAPOs. As illustrated in Fig. 5e and f, for calcined SAPO-11-C, three major peaks can be determined after deconvolution (Table 2). The signal at −92 ppm represents isolated Si(4Al) generated by the substitution of a P atom by a Si atom (SM2 mechanism), which is associated with a Brønsted acid with weak strength.^26,54,56 The −112 ppm line can be assigned to Si(4Si) that locates in the inner part of Si patches, whereby a combined SM2 + SM3 substitution mechanism produces Si domains. Alongside, a peak at ca. −97 ppm attributable to Si(3Al, 1Si) can be discerned, giving rise to Brønsted acid sites with stronger strength.^26,53–56 The contribution from other lines cannot be completely precluded, but these signals appear weak in intensity and may overlap with other major ones or vanish into the baseline. The Si(3Al, 1Si) is mainly from Si atoms that situate at the borders of Si patches. The prominent intensity for signals from Si(4Si) and Si(3Al, 1Si) suggests the presence of Si patches in addition to isolated Si(4Al), implying two distinct acid strength sites for SAPO-11-C.

Table 2 Deconvolution results (%) of the ²⁹Si MAS NMR spectra of SAPO-11-C and SAPO-11-H based on the normalized peak areas of the different Si species

Sample	Si(4Al)	Si(4Al)	Si(3Al)	Si(1Al)	Si(0Al)
ppm	−85	−92	−97	−108	−112
SAPO-11-C	0	45.4	36.8	10.7	7.1
SAPO-11-H	3.4	11.0	30.5	29.6	25.5

---

For SAPO-11-H, on the other hand, a profile of five peaks can be simulated by using individual Gaussian lines (Table 2), i.e., −85, −92, −97, −108, and −112 ppm, respectively. In addition to Si(4Al) (−92 ppm) and Si(4Si) (−112 ppm), peaks appearing at −97 and −108 ascribed to Si(3Al, 1Si) and Si(Al, 3Si) can be detected. The peak appearing at −85 ppm represents insignificant aluminosilicate (SA) type silicon in Si(4Al) environment, the SA domain formed by combined SM3 and SM2 mechanism does not contain P atoms.⁵⁵ Following models proposed for Si incorporation in aluminophosphate frameworks,^26,53,55 these lines stem from Si patches of varied sizes. It is anticipated that the formation of small Si patches will lead to high concentration of Si(nAl, 4-nSi, 0 < n < 4), and broader environments.²⁶ Conversely, large Si patches predominate when the Si(4Si) peak grows in intensity among others. It is noteworthy that a much lower Si(4Si) concentration has been inferred for SAPO-11-H (for quantitative analyses, see Table 2), indicating a better Si dispersion with respect to SAPO-11-C.²⁶ As a consequence, an increase in acid sites number can be anticipated. The improved dispersion for Si in SAPO-11-H is ascribed to the tumbling crystallization condition, which favours the mixing of ingredients.

To sum up, ²⁷Al, ³¹P, ²⁹Si MAS NMR show consistently that all the three elements have been incorporated into the aluminophosphate framework, i.e., the addition of growth inhibitor in the synthesis does not affect the integrity of the AEL structure. Moreover, SAPO-11-H comprises more small Si patches, implying an increase in strong acid sites number. The results also highlight that the generation of auxiliary macro-/meso-pores does not adversely affect the intrinsic acidity.

Formation mechanism of SAPO-11-H

To shed light on the formation mechanism of SAPO-11-H, the entire crystallization process has been tracked using XRD and SEM techniques to understand the evolution of order. The time-dependent XRD patterns are described in Fig. 6, crystallization takes place very rapidly when compared to the crystallization of SAPO-11-C (Fig. S1,† time-dependent XRD patterns) and diffraction peaks for SAPO-11 emerge after 1 h, presumably owing to the effect of tumbling, that accelerates nucleation. The intensity of peaks grows quickly over the initial 4 h, and slows down during the remaining time of crystallization. From the SEM image (Fig. 7a and b), it is seen that the initial amorphous gel does not exhibit any characteristic morphological features. After 1 h (Fig. S2† and 7c and d), the initial gel becomes 5–6 μm sized particulates, together with smaller ones. At this stage, crystallization has started, as evidenced by the corresponding XRD patterns. Small nuclei belonging to SAPO-11 become noticeable in the SEM image after 2 h of crystallization (Fig. 7e and f), as they appear at the margins of large particulates. The size of the nuclei ranges from 50 to 100 nm, and they are embedded into the huge amorphous gels, showing that they are still not fully converted. After crystallization times of 4 and 6 h (Fig. 7g–j), the nucleus grows at the cost of the inner amorphous parts, gradually consuming the starting materials. The crystallization is completed after 96 h (Fig. 7k and l). The crystal growth takes place towards an inward direction, as can be seen from the progressively covered external surfaces by crystalline building blocks. Meanwhile, the inner amorphous gel parts are gradually consumed, which can be observed from both the bottom and the side of the barrel-shape particulates. One interesting phenomenon is that the nuclei do not attach into each and the void space in between has been retained during the crystallization. The hindered attachment is ascribed to the presence of the 1,2,3-hexanetriol additive, that to a large extent inhibited the fusion of primary building blocks. In contrast, crystallization of SAPO-11-C (in the absence of 1,2,3-hexanetriol) does not show such a feature (Fig. S3†), because the initially formed nucleus is attached to neighbouring ones, thus forming an orientated attached superstructure (or single crystal when grain boundary eliminates) without void spaces in between.

Fig. 6 Time-dependent XRD patterns for the synthesis of SAPO-11-H.

Fig. 7 The SEM images for varied time in the synthesis of SAPO-11-H. (a and b) 0 h, (c and d) 1 h, (e and f) 2 h, (g and h) 4 h, (i and j) 6 h, (k and l) 96 h.

The crystal growth study reveals that SAPO-11-H crystallization can be roughly divided into three consecutive stages: amorphous gel palletisation into round-shaped particulates, nucleation at the margins of the amorphous pellets (i.e., heterogeneous nucleation) and the inward growth process. Such a growth mechanism can be sketched in Scheme 1. Substantial temporal overlap of the first two stages is observed, as nucleation occurs quickly under tumbling conditions. In the first stage, the irregularly shaped amorphous gel that is liable to changes through processes such as agglomeration or agitation-induced fission has been pelletized into spheres. The dimensions of tumbling-made particulates are system-dependent, and is equilibrated with the co-existing mother-liquid.⁵⁷ In the subsequent second stage, it is reasonable to visualize that SAPO-11 crystals have indeed nucleated heterogeneously at the pellet–liquid interface. Such a situation is credible since nutrient concentration gradients will be maximized at the pellet–liquid interface, and the embedded tiny zeolite crystallites will grow at the expense of the surrounding amorphous gel agglomerates, until the latter were completely consumed.⁵⁸ The interface of the amorphous pellets will provide sites at which the free energy change necessary for nucleus formation is lower than would be the case of homogeneous nucleation.⁵⁷ Nucleation starting from margins of amorphous gels has been observed before, as proposed for the crystallization of zeolite A (LTA topology)⁵⁸ and analcime (ANA topology).⁵⁹ It is therefore no wonder that a from-shell-to-core growth occurred during the crystallization stage. Nonetheless, often a close packed nanocrystal or single crystal are produced afterwards, henceforth, neither auxiliary porosity nor nanocrystal size could be retained in the final structure of the product. In the third stage, further growth of SAPO-11 takes place from the shell where heterogeneous nucleation started, towards the inner amorphous parts of pellets. This growth orientation preference is tentatively ascribed to the influence of 1,2,3-hexanetriol inhibitor, which can be strongly adsorbed on nucleus–liquid interfaces and inhibits the outward growth, or the fusion of primary nuclei into bulky crystals through orientated attachment. The inhibiting role has been observed by Rimer et al. for zeolite L³⁹ and by some of us for SAPO-34 recently.⁴⁷ As the unidimensional channel consists of alternating Al–O–P or Al–O–Si bonds, the polar nature of such bonds will build up a polarity perpendicular to the channel. Consequently, the surface energy normal to the channel is relatively higher than other facets, leading to the diminishing of such facets during crystal growth (Gibbs–Wulff theory), as can be seen for SAPO-11-C. Therefore, rod-shaped unidimensional SAPO-11 is frequently generated. Conversely, for SAPO-11-H, the high surface energy leads to a stronger adsorption of 1,2,3-hexanetriol on the (100) planes. The adsorption of growth inhibitor will hinder the growth along the [100] axis and decrease the aspect ratio of the primary SAPO-11 nanocrystals. On the other hand, the nucleus–pellets interface is free of growth inhibitor and a close nucleus–pellets contact can be anticipated. Under this scenario, the inward growth becomes kinetically favourable, which can also be regarded as spontaneous epitaxial growth as no energetic barrier exists for its further growth. The really fascinating part in morphology is that the growth created a hierarchical superstructure that contains auxiliary larger pores which are open, while the primary building blocks are still kept within nanometer scale.

Scheme 1 Proposed formation mechanism of SAPO-11-H.

Acidity measurements

The acidity is a crucial factor which determines the catalytic activity of zeotype materials, such as SAPOs. In order to detect the influence of our synthesis on acidity, pyridine absorption IR has been used to measure the acid type, amount and strength for the two samples. The Py-IR spectra for the two samples are displayed in Fig. 8. The bands located at 1545 and 1455 cm⁻¹ can be assigned to pyridine adsorbed on Brønsted and Lewis acid sites, respectively.^60,61 The peak at 1490 cm⁻¹ is a synergetic result from both Lewis and Brønsted acid sites.⁶² The areas of the peaks decrease gradually with the raising of outgassing temperatures from 423 to 623 K, as a consequence of weakly adsorbed pyridine desorption at elevated temperatures. Table 3 gives the quantitative results about the acid properties of different samples from Py-IR spectra. The total acid sites of SAPO-11-H (1.32 mmol Py g⁻¹) is slightly greater than that of SAPO-11-C (1.11 mmol Py g⁻¹). The number of strong Brønsted acid sites (measured at 623 K) are 0.32 and 0.34 mmol Py g⁻¹ for SAPO-11-C and SAPO-11-H, respectively. This result clearly indicates that SAPO-11-H possesses a comparable number of strong Brønsted acid sites as SAPO-11-C. These observations are consistent with ²⁹Si MAS NMR data, corroborating that the introducing of hierarchical pores onto SAPO-11 does not cause a loss of acidity.

Fig. 8 Pyridine infrared spectra for SAPO-11-C and SAPO-11-H.

Table 3 Comparison of acid amount and Brønsted/Lewis type determined by Py-IR SAPO-11-C and SAPO-11-H

Sample	Total acidity^a (mmol g^−1)	Acidity^a (mmol Py g^−1)
		Brønsted (1545 cm^−1)			Lewis (1455 cm^−1)
		423 K	473 K	623 K	423 K	473 K	623 K
a Calculated by Py-IR.
SAPO-11-C	1.11	0.89	0.67	0.32	0.22	0.18	0.08
SAPO-11-H	1.32	1.11	0.75	0.34	0.21	0.16	0.06

---

Catalytic performance in hydroisomerization of n-heptane

Hydroisomerization of n-heptane has been used to investigate the catalytic performance of bifunctional catalysts 0.5 wt% Pt/SAPO-11-C and 0.5 wt% Pt/SAPO-11-H. Above this Pt loading threshold, the (de)hydrogenation becomes kinetically unimportant,^40,41 and thereby the catalytic performance is solely governed by the properties of SAPO-11. The Pt dispersion for 0.5 wt% Pt/SAPO-11-C and 0.5 wt% Pt/SAPO-11-H are measured to be 7.7% and 13.5%, respectively (Table 1). Catalytic hydroisomerization reaction tests have been carried out at varied temperatures from 553 to 693 K. As displayed in Fig. 9, conversion increases with increasing the reaction temperature. The n-heptane conversion for Pt/SAPO-11-H is higher than that for SAPO-11-C for the whole testing temperature range. The product yield versus conversion is depicted in Fig. 10. The yield of iso-C₇ for Pt/SAPO-11-H is higher than that of Pt/SAPO-11-C, and exhibits a maximum isomer yield of 56.6%, whereas Pt/SAPO-11-C only shows a maximum of 40.3%. The results indicate that SAPO-11-H promotes both catalytic activity and isomer selectivity, as has been observed for other hierarchical SAPO-11 catalysts.³

Fig. 9 Conversion of n-heptane versus temperature.

Fig. 10 Product yield versus conversion in n-heptane hydroisomerization. Total: yield of iso-C₇; MB: yield of mono-branched iso-C₇; DB: yield of di-branched iso-C₇.

On the basis of the isomerization mechanism of n-paraffins over the bifunctional catalyst,⁶³ for bifunctional catalysts Pt/SAPO-11, n-heptane is first dehydrogenated into n-heptene on Pt centers, and then diffuses onto the Brønsted acid sites of SAPO-11 where the primary products of mono-branched C₇ isomers are formed. The mono-branched C₇ alkenes can either be hydrogenated to mono-branched C₇ alkanes products or be converted further into the di-branched C₇ alkenes on Brønsted acid sites. Finally, the di-branched C₇ alkanes are produced through hydrogenation on Pt centers. Side cracking reaction mainly comes from cracking of mono-branched or di-branched isomers.⁴⁰ The reaction scheme for the formation of isomers and cracking products is shown in Fig. S4.† The hydroisomerization-cracking of n-heptane is a consecutive reaction, and the cracking occurs through β-scission. The cracking rate increases with degree of branching, i.e., a cracking rate order follows: di-branched C₇ isomers > mono branched C₇ isomers > n-heptane.⁶⁴ As the strong Brønsted acidity sites responsible for the isomerization catalytic activity are comparable for the two samples (Fig. 8 and Table 3), the change in catalytic performance can be ascribed to their different hierarchical structure. The smaller the crystallites are, the shorter the pore channels will be, leading to a reduced retention time for primary mono-branched C₇ isomer products (with respect to Thiele theorem). As illustrated in Fig. 10, the SAPO-11-C and SAPO-11-H samples produce predominant mono-branched C₇ isomers, and the mono-branched isomers yield of SAPO-11-H is higher than that of SAPO-11-C. The reason for preferential formation of mono-branched C₇ isomers over SAPO-11 is attributed to product shape selectivity.⁴⁰ In addition, the shortened retention time can also be expected by the presence of auxiliary macro-/meso-pores opening to the external facets, which has been found to be important for accelerating mass transfer.^48,49 Besides, the high surface covered with micropore apertures (as indicated by the low aspect ratio) may also promote the pore entrance process in mass transfer. Overall, the hierarchical structure of Pt/SAPO-11-H can promote the diffusion of C₇ isomers out of the SAPO-11 micropores and thus can suppress the consecutive cracking reactions. Consequently, the n-heptane hydroisomerization conversion and selectivity for Pt/SAPO-11-H are higher than those of Pt/SAPO-11-C. Our observations are also in accordance with recent reports for hierarchical SAPO-11 derived from other routes.^3,37

Conclusions

In conclusion, a new synthetic method to generate hierarchically organized SAPO-11 by a combined growth inhibitor and tumbling crystallization strategy has been proposed. Different from known methods that relies on the using of macro- or meso-porogens, the current method is based on the synergetic effects of growth modifier and crystallization process control. The small molecular inhibitor 1,2,3-hexanetriol possesses three C–OH groups that can effectively cover the surface of embryonic crystals and hinder their growth to full sizes. The tumbling crystallization facilitates formation of pellet-shaped particulates. The embryonic crystals attach and become embedded onto the particulates’ surfaces upon tumbling, triggering a from-shell-to-core growth, finally leading to the formation of a complicated organized architecture. Such a crystallization route provides an alternative pathway to generate hierarchical SAPOs. Three outstanding features can be seen from the structure determination: (1) small primary size of SAPO-11 crystals, (2) auxiliary macro-/meso-pores connecting the crystal external surfaces, and (3) strong acid strength. The former two are crucial for mass transfer and the latter is important to maintaining the intrinsic catalytic activity. The catalytic n-heptane hydroisomerisation confirms that the advantages of hierarchical SAPO-11 are high activity and isomer selectivity. The potential of using other growth inhibitors to manipulate hierarchical structure of SAPOs is currently underway in our lab, and will be reported in the future.

Acknowledgements

XZ is sponsored by the National Natural Science Foundation of China (U1162112). KZ is grateful for the financial support from National Natural Science Foundation of China (21576082), Fundamental Research Funds for the Central Universities (WB 1213004-1). WH is grateful for the financial support from the Science and Technology Commission of Shanghai Municipality (13DZ2275200). LZ is supported by China Postdoctoral Science Foundation Grant No. 2015M580299 and Fundamental Research Funds for the Central Universities (1514011).

Notes and references

1. G. Bellussi, A. Carati and R. Millini, in Zeolites and Catalysis, ed. S. Zones et al., Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2010, ch. 16, pp. 449–491.
2. C. Martinez and A. Corma, Coord. Chem. Rev., 2011, 255, 1558–1580.
3. M. Y. Kim, K. Lee and M. Choi, J. Catal., 2014, 319, 232–238.
4. R. Yadav, A. K. Singh and A. Sakthivel, Catal. Today, 2015, 245, 155–162.
5. L. Ying, X. S. Yuan, M. Ye, Y. W. Cheng, X. Li and Z. M. Liu, Chem. Eng. Res. Des., 2015, 100, 179–191.
6. B. M. Lok, C. A. Messina, R. L. Patton, R. T. Gajek, T. R. Cannan and E. M. Flanigen, J. Am. Chem. Soc., 1984, 106, 6092–6093.
7. G. Sastre, D. W. Lewis and C. R. A. Catlow, J. Mol. Catal. A: Chem., 1997, 119, 349–356.
8. P. Liu, J. Ren and Y. H. Sun, Microporous Mesoporous Mater., 2008, 114, 365–372.
9. R. Bertolo, A. Fernandes, F. Ribeiro, J. M. Silva, A. Martins and F. R. Ribeiro, React. Kinet., Mech. Catal., 2010, 99, 183–191.
10. Y. Fan, H. Xiao, G. Shi, H. Y. Liu and X. J. Bao, J. Catal., 2012, 285, 251–259.
11. Y. X. Liu, X. Cui, L. Han, Z. W. Yu and X. M. Liu, Microporous Mesoporous Mater., 2014, 198, 230–235.
12. L. Guo, Y. Fan, X. J. Bao, G. Shi and H. Y. Liu, J. Catal., 2013, 301, 162–173.
13. M. Liu, W. Wu, O. V. Kikhtyanin, L. F. Xiao, A. V. Toktarev, G. L. Wang, A. J. Zhao, M. Y. Smirnova and G. V. Echevsky, Microporous Mesoporous Mater., 2013, 181, 132–140.
14. C. M. Song, Y. Feng and L. L. Ma, Microporous Mesoporous Mater., 2012, 147, 205–211.
15. V. Valtchev and L. Tosheva, Chem. Rev., 2013, 113, 6734–6760.
16. Q. Liu, H. Zuo, T. Wang, L. Ma and Q. Zhang, Appl. Catal., A, 2013, 468, 68–74.
17. K. Moeller and T. Bein, Chem. Soc. Rev., 2013, 42, 3689–3707.
18. D. P. Serrano, J. M. Escola and P. Pizarro, Chem. Soc. Rev., 2013, 42, 4004–4035.
19. Y. Seo, S. Lee, C. Jo and R. Ryoo, J. Am. Chem. Soc., 2013, 135, 8806–8809.
20. X. Zhang, D. Liu, D. Xu, S. Asahina, K. A. Cychosz, K. V. Agrawal, Y. Al Wahedi, A. Bhan, S. Al Hashimi, O. Terasaki, M. Thommes and M. Tsapatsis, Science, 2012, 336, 1684–1687.
21. J. G. Jiang, L. L. Jia, B. T. Yang, H. Xu and P. Wu, Chem. Mater., 2013, 25, 4710–4718.
22. R. Bertolo, J. M. Silva, F. Ribeiro, F. J. Maldonado-Hodar, A. Fernandes and A. Martins, Appl. Catal., A, 2014, 485, 230–237.
23. D. Verboekend, M. Milina and J. Perez-Ramirez, Chem. Mater., 2014, 26, 4552–4562.
24. H. Tian and C. Li, J. Mol. Catal. A: Chem., 1999, 149, 205–213.
25. Y. Liu, W. Qu, W. W. Chang, S. X. Pan, Z. J. Tian, X. J. Meng, M. Rigutto, A. van der Made, L. Zhao, X. M. Zheng and F. S. Xiao, J. Colloid Interface Sci., 2014, 418, 193–199.
26. T. Blasco, A. Chica, A. Corma, W. J. Murphy, J. Agúndez-Rodríguez and J. Pérez-Pariente, J. Catal., 2006, 242, 153–161.
27. D. Verma, B. S. Rana, R. Kumar, M. G. Sibi and A. K. Sinha, Appl. Catal., A, 2015, 490, 108–116.
28. K. C. Park and S. K. Ihm, Appl. Catal., A, 2000, 203, 201–209.
29. Y. Liu, C. Liu, C. Liu, Z. Tian and L. Lin, Energy Fuels, 2004, 18, 1266–1271.
30. V. M. Benitez, J. C. Yori, J. M. Grau, C. L. Pieck and C. R. Vera, Energy Fuels, 2006, 20, 422–426.
31. P. Mériaudeau, V. A. Tuan, F. Lefebvre, V. T. Nghiem and C. Naccache, Microporous Mesoporous Mater., 1998, 22, 435–449.
32. A. K. Sinha and S. Sivasanker, Catal. Today, 1999, 49, 293–302.
33. P. Liu, J. Ren and Y. Sun, Catal. Commun., 2008, 9, 1804–1809.
34. H. Yunfeng, W. Xiangsheng, G. Xinwen, L. Silue, H. Sheng, S. Haibo and B. Liang, Catal. Lett., 2005, 100, 59–65.
35. Y. Fan, H. Xiao, G. Shi, H. Liu and X. Bao, J. Catal., 2012, 285, 251–259.
36. R. Bértolo, J. M. Silva, F. Ribeiro, F. J. Maldonado-Hódar, A. Fernandes and A. Martins, Appl. Catal., A, 2014, 485, 230–237.
37. Y. Liu, W. Qu, W. Chang, S. Pan, Z. Tian, X. Meng, M. Rigutto, A. v. d. Made, L. Zhao, X. Zheng and F. S. Xiao, J. Colloid Interface Sci., 2014, 418, 193–199.
38. L. Guo, X. Bao, Y. Fan, G. Shi, H. Liu and D. Bai, J. Catal., 2012, 294, 161–170.
39. A. I. Lupulescu, M. Kumar and J. D. Rimer, J. Am. Chem. Soc., 2013, 135, 6608–6617.
40. T. L. M. Maesen, M. Schenk, T. J. H. Vlugt, J. P. d. Jonge and B. Smit, J. Catal., 1999, 188, 403–412.
41. M. Höchtl, A. Jentys and H. Vinek, J. Catal., 2000, 190, 419–432.
42. T. Onfroy, G. Clet and M. Houalla, Microporous Mesoporous Mater., 2005, 82, 99–104.
43. S. Zhang, S.-L. Chen, P. Dong, G. Yuan and K. Xu, Appl. Catal., A, 2007, 332, 46–55.
44. Y. Fan, D. Lei, G. Shi and X. Bao, Catal. Today, 2006, 114, 388–396.
45. Q. Wu, I. Nartey Oduro, Y. Huang and Y. Fang, Microporous Mesoporous Mater., 2015, 218, 24–32.
46. S. Zhang, S.-L. Chen and P. Dong, Catal. Lett., 2010, 136, 126–133.
47. J. Zheng, W. Zhang, Z. Liu, Q. Huo, K. Zhu, X. Zhou and W. Yuan, Microporous Mesoporous Mater., 2016, 225, 74–87.
48. P. Kortunov, S. Vasenkov, J. Kärger, R. Valiullin, P. Gottschalk, M. Fé Elía, M. Perez, M. Stöcker, B. Drescher, G. McElhiney, C. Berger, R. Gläser and J. Weitkamp, J. Am. Chem. Soc., 2005, 127, 13055–13059.
49. M. Milina, S. Mitchell, P. Crivelli, D. Cooke and J. Pérez-Ramírez, Nat. Commun., 2014, 5 DOI:10.1038/ncomms4922.
50. A. Buchholz, W. Wang, M. Xu, A. Arnold and M. Hunger, Microporous Mesoporous Mater., 2002, 56, 267–278.
51. Y. Watanabe, A. Koiwai, H. Takeuchi, S. A. Hyodo and S. Noda, J. Catal., 1993, 143, 430–436.
52. H.-L. Zubowa, E. Alsdorf, R. Fricke, F. Neissendorfer, J. Richter-Mendau, E. Schreier, D. Zeigan and B. Zibrowius, J. Chem. Soc., Faraday Trans., 1990, 86, 2307–2312.
53. D. Barthomeuf, Zeolites, 1994, 14, 394–401.
54. M. J. Franco, A. Mifsud and J. Perez-Pariente, Zeolites, 1995, 15, 117–123.
55. J. A. Martens, P. J. Grobet and P. A. Jacobs, J. Catal., 1990, 126, 299–305.
56. J. A. Martens, C. Janssens, P. J. Grobet, H. K. Beyer and P. A. Jacobs, Stud. Surf. Sci. Catal., 1989, 49, 215–225.
57. C. S. Cundy and P. A. Cox, Microporous Mesoporous Mater., 2005, 82, 1–78.
58. S. Mintova, N. H. Olson, V. Valtchev and T. Bein, Science, 1999, 283, 958–960.
59. X. Chen, M. Qiao, S. Xie, K. Fan, W. Zhou and H. He, J. Am. Chem. Soc., 2007, 129, 13305–13312.
60. J. A. Z. Pieterse, S. Veefkind-Reyes, K. Seshan, L. Domokos and J. A. Lercher, J. Catal., 1999, 187, 518–520.
61. Y. Fan, H. Xiao, G. Shi, H. Liu, Y. Qian, T. Wang, G. Gong and X. Bao, J. Catal., 2011, 279, 27–35.
62. E. P. Parry, J. Catal., 1963, 2, 371–379.
63. Y. Ono, Catal. Today, 2003, 81, 3–16.
64. J. Weitkamp, P. A. Jacobs and J. A. Martens, Appl. Catal., 1983, 8, 123–141.

---

Footnote

† Electronic supplementary information (ESI) available: Time-dependent XRD patterns in the synthesis of SAPO-11-C, panoramic SEM image for crystallization of SAPO-11-H after crystallization of 1 h, and SEM images for samples crystallized for varied time duration in the synthesis of SAPO-11-C. See DOI: 10.1039/c6ra03039d

---