Preparation of a preferentially oriented SAPO-34 membrane by secondary growth under microwave irradiation

Abstract

Piece-like SAPO-34 crystals were directly assembled into a highly oriented seed layer on the α-Al₂O₃ substrate by using spin-coating, and then a preferentially oriented SAPO-34 membrane with no large defects was prepared on an oriented SAPO-34 seed layer by secondary hydrothermal microwave growth. The synthesized membrane exhibited a high CO₂ permeance of 1.57 × 10⁻⁶ s⁻¹ Pa⁻¹ at 302 K and 138 kPa pressure drop, with a CO₂/CH₄ separation selectivity of 109.

---

Zeolites are crystalline materials with angstrom scale pores and channels in their crystal lattice. Zeolite powders are widely used in the chemical industry as shape-selective catalysts,¹ ion exchangers and adsorbers,² and zeolite membranes, which offer great potential for a variety of applications such as gas or liquid separation, membrane reactors, chemical sensors, corrosion-resistant coatings and micro-electronic devices.^3–5 Ways to control the membrane microstructures, including manipulation of grain size and shape,⁶ membrane thickness,⁷ and orientations,⁸ minimization of defects,⁹ are very important for improving membrane performances.

Oriented zeolite membranes exhibit superior performance compared to randomly oriented ones. For example, appropriate orientation of crystals can effectively enhance the mass transport and mechanical stress of the membranes which related to membrane separation and chemical sensors.¹⁰ Despite great effort over the past several years, challenges still exist to fabricate high-quality oriented zeolite membranes. Compared with other methods, including in situ hydrothermal synthesis, the secondary growth method is more reliable for achieving preferentially oriented membranes because the membrane microstructure (i.e., thickness, grain size, crystal orientation) can be easier controlled by decoupling the nucleation and crystal growth steps during membrane formation.

In the preparation of oriented zeolite membranes by secondary growth starting from randomly oriented seed layers, grains grow out rapidly along their fastest growth direction, while grains in other orientations grown slower will be gradually buried by faster grains. The final membrane orientation is attributed to the competitive growth of grains along different crystallographic directions. During secondary growth, the orientation of the seed layer is critical for the preparation of membranes with the same orientation and often enhances their separation performance by providing desired pore channels perpendicular to the membrane surface. Tsapatsis and co-workers prepared oriented ZSM-5 membranes by secondary growth of b-oriented monolayers. Compared with c-oriented membranes synthesized under the same conditions except for the use of randomly oriented seed layers, these b-oriented membranes are thinner, less defective and more robust to crack formation.¹¹ Huang et al. reported that an oriented AlPO₄ LTA zeolite membrane prepared using a highly oriented AlPO₄ LTA monolayer showed superior gas separation performances compared to a random AlPO₄ LTA membrane.¹² SAPO-34 has a CHA framework topology and a three-dimensional 8-ring channel system with a pore diameter of 0.38 nm. The pores in SAPO-34 are essentially the same size as the kinetic diameter of CH₄ (0.38 nm), so the small-pore SAPO-34 favors the diffusion of CO₂ (0.33 nm) over CH₄. Besides, SAPO-34 adsorbs CO₂ more strongly than CH₄, which could improve separation performance in CO₂/CH₄ binary mixtures when it is used as a membrane. SAPO-34 has straight channels with an estimated pore opening of 0.38 nm × 0.38 nm along its a-axis (Scheme 1), so the growth of a-oriented crystals in a membrane can effectively reduce the mass transport resistance and control the thermal stress when related to gas separation. Zhu et al.¹³ synthesized an a-oriented SAPO-34 membrane on a macroporous stainless-steel-net. This oriented membrane displayed CO₂/CH₄ separation selectivity of less than 10, although CO₂ permeance was as high as 2.5 × 10⁻⁶ mol m⁻² s⁻¹ Pa⁻¹.

Scheme 1 The topology of SAPO-34 structure along the [100] direction (According to the database of zeolite structures available from the International Zeolite Association: http://www.iza-structure.org/databases/).

Das and co-workers^14–16 prepared preferentially oriented SAPO-34 membranes on clay-Al₂O₃ supports by selective deposition of oriented seed crystals, followed by epitaxial secondary growth. In this procedure, cubic SAPO-34 particles (thicker than 1 mm) were used as seeds, and the membrane thickness was found to be about 25 μm. Considering the shortest dimension perpendicular to the large basal plane in piece-like particles would be beneficial in the formation of an oriented layer, Choi et al.¹⁷ selectively deposited piece-like Si-CHA particles onto the α-Al₂O₃ disk by a sonication-assisted method with the successful formation of uniform h0h-out-of-plane-oriented Si-CHA layers. Furthermore, compared with the cubic crystals, the nanosized piece-like crystals would be beneficial in decreasing the thickness of the membrane. Unlike other strategies, including in situ growth, seed growth, convective assembly and electric-field-driven assembly, spin-coating is a simple, quick, effective and highly reproducible strategy to prepared zeolite layer. In the present work, an oriented SAPO-34 seed layer was prepared by spin-coating the α-Al₂O₃ substrate with piece-like SAPO-34 crystals, and then a preferentially SAPO-34 membrane was prepared on an oriented SAPO-34 seed layer under microwave irradiation, and was used for the separation of equimolar CO₂/CH₄ gas mixtures.

The SAPO-34 crystals were synthesized by using a microwave irradiation method and the synthesis procedure was similar to that described previously.¹⁸ After hydrothermal treatment at 453 K for 2 h, pure SAPO-34 crystals with high crystallinity are formed (Fig. 1). The SAPO-34 crystals that are used as seeds for membrane preparation are piece-like with the approximate dimension of 400 × 400 × 40 nm (Fig. 2a). An a-oriented SAPO-34 seed layer was prepared as follows. The seed suspension with a concentration of 0.5 wt% in ethanol was irradiated with ultrasound for 1 h. The suspension was then spin-coated onto the treated α-Al₂O₃ substrate with a spinning rate of 1280 rpm for 40 s. After dried at 333 K overnight, uniform layer was obtained. Fig. 2b shows the SEM top view of the SAPO-34 seed layer deposited by spin-coating. The porous α-Al₂O₃ support surface is completely covered by uniform SAPO-34 crystals, and the seed layer is preferentially oriented. XRD pattern further verifies the formation of a preferentially a-oriented SAPO-34 seed layer on the α-Al₂O₃ support surface (Fig. 1). Compared with the randomly oriented XRD pattern of SAPO-34 zeolite powder, the oriented SAPO-34 seed layer shows only (100) peak besides the α-Al₂O₃ signals from the substrate.

Fig. 1 XRD patterns of (a) SAPO-34 crystals, (b) α-Al₂O₃ substrate, (c) SAPO-34 seed layer and (d) SAPO-34 membrane. The asterisk (*) indicates the XRD peaks of a bare α-Al₂O₃ support.

Fig. 2 SEM images of (a) SAPO-34 seeds, (b) top-view of SAPO-34 seed layer, (c) top-view and (d) cross-section of SAPO-34 membrane.

Zhu and co-workers prepared a-oriented SAPO-34 membrane by using conventional heating.¹³ In contrast to this traditional method, the use of microwave heating allows faster crystal nucleation and crystal growth in the form of powders or membranes. Fig. 2c shows the SEM top view of the preferentially oriented membrane prepared by microwave-assisted secondary growth. After microwave-assisted secondary growth for 4 h, zeolite seeds on the surface have grown bigger and merged with each other, and no visible cracks and grain boundary defects are observed. Meanwhile, the SAPO-34 membrane still keeps preferentially oriented. From the cross section shown in Fig. 2d, the membrane is dense through the whole membrane thickness, and the thickness is about 4 μm. The XRD pattern of SAPO-34 membrane is shown in Fig. 1. As clearly seen in the figure, the peak at 9.4° corresponding to the (100) peak is still observed with an enhancement in the peak intensity. It suggests that the SAPO-34 membrane retains the preferred orientation, which is in good agreement with the result of the SEM images.

The FT-IR spectra of the oriented SAPO-34 membrane before and after calcination are shown in Fig. 3. The bands at around 2800–3200 and 1300–1500 cm⁻¹ are attributed to the template TEAOH, which disappeared after calcinations at 390 °C in flowing air.¹⁹ The low defect concentration of the SAPO-34 membrane could be confirmed by measuring the single gas permeance for i-C₄H₁₀.²⁰ Fig. 4 shows the single gas permeances through the oriented membrane as a function of molecular kinetic diameter at 302 K with a 0.4 MPa pressure difference. After template removal from SAPO-34 membrane, the single gas permeances of molecules larger than CO₂ decrease with the increase in kinetic diameter, which indicates that the SAPO-34 membrane has a relatively narrow pore size distribution and its gas permeation behavior can be controlled by zeolitic pores. Therefore, CH₄ molecules can permeate through SAPO-34 membrane to only a small extent, and n-C₄H₁₀ and i-C₄H₁₀ (0.43 and 0.50 nm, respectively) hardly at all, which indicates that the SAPO-34 membrane prepared in this study had no large defects. As shown in Fig. 4, CO₂ permeance is higher than those of He and H₂, although the kinetic diameter of CO₂ is larger than that of He (0.26 nm) and H₂ (0.29 nm). Because CO₂, He and H₂ are smaller than SAPO-34 pores, the size difference among CO₂, He and H₂ has a small effect on determining the order of their permeances. Thus, differences in adsorption are more important in determining permeance. According to the adsorption isotherms on SAPO-34 power reported by Noble and co-workers,^21–23 SAPO-34 crystals adsorb CO₂ much more strongly than He and H₂. So the single-gas permeation of CO₂ is more than that of He and H₂ though CO₂ is the larger molecule.

Fig. 3 FT-IR spectra of (a) as-synthesized and (b) calcined SAPO-34 membranes.

Fig. 4 Single-gas permeances through SAPO-34 membrane at 302 K as a function of kinetic diameter.

The separation performance of the SAPO-34 membrane with a thickness of 4 μm was measured by separating the equimolar CO₂/CH₄ binary gas mixture after organic templates were removed. As shown in Table 1, the CO₂/CH₄ selectivity and CO₂ permeance are 109 and 1.57 × 10⁻⁶ mol m⁻² s⁻¹ Pa⁻¹ respectively at 138 kPa pressure drop and 302 K. Table 1 also lists the CO₂/CH₄ separation of the SAPO-34 membranes in previous studies.^13,18,24,25 The synthesized thick membrane shows a relatively high CO₂ permeation as compared to previously reported SAPO-34 membranes with random orientations. For a preferentially oriented membrane, the pores are more aligned and the mass transport resistance through the aligned channel is less than that of the zig–zag path of the non-aligned randomly oriented pores. As a result, CO₂ permeance through the membrane is enhanced. Lower CO₂ permeance but higher CO₂/CH₄ selectivity compared to membrane reported in the literature¹³ is due to the use of the homogeneous and dense SAPO-34 membrane in this study.

Table 1 Comparison of CO₂/CH₄ separation performances of SAPO-34 membrane in this study with other membranes from literature

Membrane	Orientation	Thickness (μm)	Temperature (K)	Pressure drop (kPa)	CO2 permeance (10^−6 mol m^−2 s^−1 Pa^−1)	CO2/CH4 selectivity	References
SAPO-34	Random	2.5	295	138	0.40	115	24
SAPO-34	Random	3	303	100	1.20	95	25
SAPO-34	Random	2.5	302	138	0.91	130	18
SAPO-34	a	10–20	Room temp	120	2.52	9.3	13
SAPO-34	a	4	302	138	1.57	109	This work

---

Conclusions

In summary, highly a-oriented SAPO-34 seed layer was successfully prepared by spin-coating the α-Al₂O₃ substrate with piece-like SAPO-34 crystals. After secondary microwave hydrothermal growth, a well-intergrown oriented membrane with a preferential (100) orientation was grown on the α-Al₂O₃ substrate. The quality of the preferentially oriented SAPO-34 membrane was characterized by single gas permeances, which suggested that this membrane had no large defects and its gas permeation behavior could be controlled by zeolitic pores. Compared with the randomly oriented SAPO-34 membranes, the preferentially oriented SAPO-34 membrane displayed significantly higher CO₂ permeance. The membrane had CO₂ permeance as high as 1.57 × 10⁻⁶ s⁻¹ Pa⁻¹ with a CO₂/CH₄ separation selectivity of 109 at 302 K and a pressure drop of 138 kPa.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21136008 and 21476171).

Notes and references

1. M. Dusselier, P. V. Wouwe, A. Dewaele, P. A. Jacobs and B. F. Sels, Science, 2015, 349, 78–80.
2. M. E. Davis, Nature, 2002, 417, 813–821.
3. B. Q. Zhang, C. Wang, L. Lin, R. L. Cui and X. F. Liu, Adv. Funct. Mater., 2008, 18, 3434–3443.
4. C. M. Lew, R. Cai and Y. S. Yan, Acc. Chem. Res., 2010, 43, 210–219.
5. B. Q. Zhang, M. Zhou and X. F. Liu, Adv. Mater., 2008, 20, 2183–2189.
6. Y. Yoo, Z. P. Lai and H. K. Jeong, Microporous Mesoporous Mater., 2009, 123, 100–106.
7. L. C. Boudreau, J. A. Kuck and M. Tsapatsis, J. Membr. Sci., 1999, 152, 41–59.
8. J. Choi, S. Ghosh, Z. P. Lai and M. Tsapatsis, Angew. Chem., Int. Ed., 2006, 45, 1154–1158.
9. Y. S. Li, H. Bux, A. Feldhoff, G. L. Li, W. S. Yang and J. Caro, Adv. Mater., 2010, 22, 3322–3326.
10. J. H. Dong, Y. S. Lin, M. Z. C. Hu, R. A. Peascoe and E. A. Payzant, Microporous Mesoporous Mater., 2000, 34, 241–253.
11. Z. P. Lai, M. Tsapatsis and J. P. Nicolich, Adv. Funct. Mater., 2004, 14, 716–729.
12. A. S. Huang and J. Caro, Chem. Commun., 2011, 47, 4201–4203.
13. Y. Y. Tian, L. L. Fan, Z. Y. Wang, S. L. Qiu and G. S. Zhu, J. Mater. Chem., 2009, 19, 7698–7703.
14. J. K. Das, N. Das and S. Bandyopadhyay, Int. J. Hydrogen Energy, 2012, 37, 10354–10364.
15. J. K. Das, N. Das and S. Bandyopadhyay, J. Mater. Chem. A, 2013, 1, 4966–4973.
16. J. K. Das and N. Das, ACS Appl. Mater. Interfaces, 2014, 6, 20717–20728.
17. E. Kim, W. X. Cai, H. Baik and J. Choi, Angew. Chem., Int. Ed., 2013, 52, 5280–5284.
18. L. C. Bing, X. F. Liu and B. Q. Zhang, J. Mater. Sci., 2016, 51, 1476–1483.
19. M. Briend, R. Vomscheid, M. J. Peltre, P. P. Man and D. Barthomeuf, J. Phys. Chem., 1995, 99, 8270–8276.
20. H. H. Funke, M. Z. Chen, A. N. Prakash, J. L. Falconer and R. D. Noble, J. Membr. Sci., 2014, 456, 185–191.
21. J. C. Poshusta, V. A. Tuan, E. A. Pape, R. D. Noble and J. L. Falconer, AIChE J., 2000, 46, 779–789.
22. S. Li, J. L. Falconer and R. D. Noble, J. Membr. Sci., 2004, 241, 121–135.
23. S. Li, J. L. Falconer and R. D. Noble, Microporous Mesoporous Mater., 2008, 110, 310–317.
24. S. Li, J. L. Falconer and R. D. Noble, Adv. Mater., 2006, 18, 2601–2603.
25. X. F. Liu, S. Y. Du and B. Q. Zhang, Mater. Lett., 2013, 91, 195–197.

---

Footnote

† Electronic supplementary information (ESI) available: Experimental and characterization details. See DOI: 10.1039/c6ra09018d

---