Synthesis of SAPO-34 Crystals in the Presence of Crystal Growth Inhibitors

Surendar R. Venna and Moises A. Carreon*Department of Chemical Engineering, UniVersity of LouisVille, LouisVille, Kentucky 40292

ReceiVed: October 21, 2008; ReVised Manuscript ReceiVed: NoVember 10, 2008

Microporous SAPO-34 molecular sieves were synthesized employing polyethylene glycol, polyoxyethylenelauryl ether, and methylene blue as crystal growth inhibitors. The synthesized SAPO-34 crystals displayedBET surface areas up to 700 m2/g, high CO2/CH4 adsorption ratios, and small crystal size in the ∼0.6-0.9µm range with narrow particle size distribution. The enhanced CO2/CH4 adsorption capacities were related tothe high N/H ratios observed in the phases prepared in the presence of crystal growth inhibitors. The synthesizedSAPO-34 crystals may find potential applications to prepare membranes for CO2 purification.

Introduction

The demand for novel functional microporous materials withcontrolled desired properties is steadily increasing. SAPO-34,(SixAlyPz)O2, where x ) 0.01-0.98, y ) 0.01-0.60, and z )0.01-0.52, a type of silicoaluminophosphate microporouszeolite, is of particular interest in separation, catalytic, andadsorption technologies because of its chemical and thermalstability, unique shape selectivity, molecular sieving properties,and atomically ordered ∼0.38 nm pore structure. SAPO-34 hasbeen successfully employed to separate carbon dioxide1 andhydrogen2 from different gases. Properties such as fairly strongBrønsted acidity,3 adsorption of desired components,4 andexcellent shape selectivity4,5 make SAPO-34 an ideal active andselective catalyst in methanol-to-olefin reaction6 and hydrocar-bon transformation.4,7 SAPO-34 has been used for the trappingof hydrocarbons, in particular, for cold start emission controlin the automobile industry.8 It has also been used for thethermochemical storage of heat because of its unique wateradsorption properties.9 In addition, due to its unique cage sizeand shape, SAPO-34 has been found to be suitable for selectiveformation of lower olefins from methanol.4,5a

SAPO-34 has been prepared by several synthetic routes; forexample, Lok and co-workers first reported the hydrothermalsynthesis of SAPO-34 using tetraethylammonium hydroxide(TEAOH) as a structure-directing agent.10 Different structure-directing agents such as morpholine,11 piperidine,12 diethy-lamine,13 triethylamine,14 isopropylamine,15 TEAOH-dipropyl-amine,1e and TEAOH-cyclohexylamine1e have been employedto prepare SAPO-34. Other synthetic approaches for SAPO-34involve different heat treatment methods such as two-stagetemperature varying hydrothermal synthesis,11 microwave syn-thesis,16 the vapor phase transport method,17 ultrasound irradia-tion synthesis,18 the use of various sources of silica and aluminaas precursors,6a,19 and the incorporation of different metals.20

However, the control of the crystal size of SAPO-34 with highsurface area, high crystallinity, homogeneity, and preferentialadsorption capacities over light gases is still challenging. SAPO-34 with small crystal size and narrow particle size distribution

potentially leads to larger-accessible surface area, reduceddiffusion resistance, and increased adsorption capacity phases,which may impact positively its performance in functionalapplications such as gas separation1,2 and heterogeneouscatalysis.6,7 Herein, we describe the successful synthesis ofSAPO-34 crystals by using polyethylene glycol (PEG), poly-oxyethylene lauryl ether (Brij-35), and methylene blue (MB)as crystal growth inhibitors. The use of these additives resultedin higher surface area and smaller SAPO-34 crystals with narrowcrystal size distribution. The influence of crystal growthinhibitors in the adsorption capacities of SAPO-34 crystals onCO2 and CH4 was studied.

Experimental Section

Synthesis of SAPO-34 Crystals. Aluminum isopropoxide(Al(i-C3H7O)3, >99.99% metal basis, Aldrich), phosphoric acid(85 wt % aqueous solution, Sigma-Aldrich), and Ludox (40 wt% in suspension, Sigma-Aldrich) were used as the inorganicprecursors. Tetraethylammonium hydroxide (35 wt %, Sigma-Aldrich) and dipropylamine (99 wt %, Aldrich) were used asthe primary and secondary structure-directing agents (SDAs),respectively. The crystal growth inhibitors (CGIs) used in thesynthesis were polyethylene glycol-600 (Alfa Aesar), polyoxy-ethylene lauryl ether (Acros Organics), and methylene blue(Sigma-Aldrich). In a typical synthesis, aluminum isopropoxide,phosphoric acid, and H2O were mixed and stirred for about 2 hto form a homogeneous solution. Then, Ludox was added andthe resulting solution was stirred for another 2 h. The primaryand secondary templates were added to the precursor solution,and stirring continued for another 30 min before crystal growthinhibitors were added in the desired molar ratio. The composi-tion of the gel was 1Al2O3:1H3PO4:0.3SiO2:1TEAOH:1.6DPA:77H2O:xCGI, where 0.037 < x < 0.2. The final resulting gelwas aged for 3 days while stirring at 50 °C, since aging improvescontrol over zeolite nucleation by depolymerizing Ludox andforming the germ nuclei or even complete nuclei.21 Thehomogeneous gel was transferred to a Teflon lined stainless steelautoclave (Parr Instrument Company) and heated under autog-enous pressure under static conditions in a conventional ovenat 220 °C for 24 h. Alternatively, hydrothermal treatment wasalso done using a controlled temperature ramp as described

* Corresponding author. E-mail: macarr15@louisville.edu. Phone: 502-852-4103. Fax: 502-852-6355.

16261

10.1021/jp809316s CCC: $40.75 2008 American Chemical Society

Published on Web 11/21/2008

2008, 112, 16261–16265

elsewhere.22 After the gel was cooled to room temperature, itwas centrifuged at 4000 rpm for 20 min to separate the seedsfrom the mother liquor. The resultant precipitate was washedwith water. The centrifugation-washing process was repeated3 times. The resultant precipitate was dried overnight at 60 °C.The as-synthesized samples were calcined at 550 °C for 5 hwith heating and cooling rates of 1° and 10 °C/min, respectively,to remove both the structure-directing agents and the retainedcrystal growth inhibitors.

Characterization. The crystal size of SAPO-34 was deter-mined with a FE-SEM (FEI Nova 600) with an accelerationvoltage of 6 kV. The crystalline structure was analyzed bycollecting XRD patterns on a Bruker D8 Discover diffracto-meter. The quantitative analysis of elemental carbon, hydrogen,and nitrogen were carried out using a model 440 CHN/O/Sanalyzer (Exeter analytical, MA). N2 adsorption BET surfaceareas were determined in a Micromeritics Tristar-3000 poro-simeter. Before the surface area measurements, the samples weredegassed at 300 °C for 3 h. Adsorption isotherms for CO2 andCH4 were measured at 25 °C using water as the coolant. TEMimages and diffraction patterns were taken on a Technai F20FEI TEM using a field emission gun, operating with anaccelerating voltage of 200 kV. BX FTIR (Perkin-Elmer) wasused to determine the lattice vibrations in the inorganicframework.

Results and Discussion

The BET surface area and average crystal size of SAPO-34crystals synthesized employing crystal growth inhibitors (CGIs)are shown in Table 1. The samples synthesized with CGIdisplayed higher surface areas, most likely due to their smallercrystal size and due to the incorporation of extra microporosityin the SAPO-34 framework. Specific surface areas in the540-700 m2/g range were obtained when CGIs were incorpo-rated in the synthesis gel, while surface areas of ∼500 m2/gwere obtained in the absence of CGI.

XRD patterns of the calcined SAPO-34 samples synthesizedusing PEG, Brij-35, MB as CGI, and without CGI, respectively,are shown in Figure 1. All XRD patterns correspond to chabazitestructure,23 which is the typical structure of SAPO-34. Theintensity and peak positions are in agreement with previouslyreported SAPO-34 XRD spectra.1b,e,f,16 Except for the samplesynthesized using Brij-35, all samples were composed of pureSAPO-34. The sample synthesized with Brij-35 showed a peakat 2θ ∼ 23° (Figure 1b) which tentatively is assigned to AlPO-18. The AlPO-18 framework has a three-dimensional poresystem with a pore diameter of ∼0.38 nm which is essentiallythe same as SAPO-34.1e The interplanar spacings calculated

using Bragg’s law from the reflections at different 2θ valuesand TEM diffractograms are in good agreement with reportedliterature data for SAPO-34.23 Absorption FTIR spectra22

revealed the lattice vibration regions and their respective IRfrequencies of all SAPO-34 samples. These spectra correspondto SAPO-34 with chabazite structure.24 XRD and FTIR spectrasuggest that the crystal structure of SAPO-34 is not affectedby the incorporation of the CGI.

Figure 2a-c shows the SEM images of SAPO-34 crystalsexhibiting a pseudocubic plate-like morphology, which is typicalof SAPO-34. The average crystal size of these samples was inthe ∼0.6-0.9 µm range. Figure 2a shows crystals of ∼0.8 µmsize and thickness of ∼0.25 µm, synthesized employing PEGas a CGI. The crystal size decreased to ∼0.7 µm with ∼0.2 µmthickness, using Brij-35 as a CGI (Figure 2b). When MB wasemployed as a CGI, the crystal size and thickness were ∼0.9and ∼0.2 µm, respectively, as shown in Figure 2c. Interestingly,regular sized ∼0.15 µm nanocubes were also observed whenMB was used as a CGI. The sample prepared without CGIshows cubic morphology, displaying well-developed crystalsin the size range 1.5-2 µm22 with homogeneous crystal sizedistribution. Figure 2d shows the TEM image of SAPO-34prepared with Brij-35 as a CGI, confirming the plate-likemorphology and ∼0.7 µm in size crystals.

CO2 and CH4 were chosen as the probe molecules to studythe adsorption capacity of SAPO-34 due to their energy andenvironmental importance and relevance in gas separations.1,2

To study the adsorption capacity of SAPO-34, CO2 and CH4

isotherms were collected at room temperature using water asthe coolant. The adsorption isotherms of SAPO-34 crystalssynthesized with different CGIs are shown in Figure 3. TheSAPO-34 crystals adsorbed CO2 preferentially over CH4. Toevaluate the effect of CGI on the CO2 and CH4 adsorptionproperties, a reference material was synthesized without usingany CGI. At 1 atm pressure, the SAPO-34 crystals synthesizedusing CGI adsorbed ∼8-9 times more CO2 than CH4. For thereference material (without CGI), the CO2/CH4 adsorption ratiowas only ∼6.5. The high CO2/CH4 adsorption capacitiesobserved for samples prepared with CGI make these crystalsattractive for CO2/CH4 mixture separation, since it is known

TABLE 1: BET Specific Surface Area and Average CrystalSize of SAPO-34 Crystals Synthesized Using Different CGIs

sampleIDa

crystal growth inhibitor(CGI)

specific surfacearea (m2/g)

average crystalsize (µm)

1 polyethylene glycol (PEG) 645 0.7 ( 0.12 polyethylene glycol (PEG) 622 0.7 ( 0.23 polyoxyethylene lauryl ether

(Brij-35)698 0.6 ( 0.1

4 polyoxyethylene lauryl ether(Brij-35)

633 0.7 ( 0.1

5 methylene blue (MB) 540 0.9 ( 0.16 methylene blue (MB) 563 0.9 ( 0.17 methylene blue (MB) 700 0.6 ( 0.28 without CGI 496 1.4 ( 0.2

a The detailed gel composition for the synthesis of SAPO-34crystals is described elsewhere.22

Figure 1. XRD patterns of SAPO-34 crystals synthesized using (a)PEG, (b) Brij-35, (c) MB as CGI, and (d) without CGI. The inset showsthe SAPO-34 diffraction pattern. / indicates the peak of AlPO-18crystals. hkl Miller indexes are assigned on the basis of chabazitestructure.

16262 J. Phys. Chem. B, Vol. 112, No. 51, 2008 Letters

that the higher the CO2/CH4 adsorption ratio, the higher the CO2/CH4 separation selectivity.1d,e Since both CO2 (0.33 nm) andCH4 (0.38 nm) molecules are able to enter the porous structureof SAPO-34, the CO2/CH4 separation selectivity will dependon the preferential adsorption of CO2 even though it diffusesfaster than CH4.

Figure 4 shows the effect of N/H molar ratio on the CO2 andCH4 adsorption capacities. The CHN analysis revealed that morenitrogen was retained or incorporated in the SAPO-34 frame-work when CGIs were employed in the synthesis gel. The N/H

molar ratios of the SAPO-34 crystals synthesized using CGIwere in the range of 11 × 10-4 to 18 × 10-4. However, it wasonly ∼8 × 10-4 for the samples synthesized without CGI.Furthermore, the N/H molar ratio increased with the amount ofCGI. For instance, The N/H molar ratio increased from 11.5 ×10-4 to 16.5 × 10-4 for the SAPO-34 crystals synthesizedemploying 2 and 4 g of PEG as a CGI, respectively. The sametrend was observed for Brij-35 and MB. Interestingly, the higherthe N/H molar ratio, the higher the CO2/CH4 adsorption ratio.CO2, a Lewis acid site, will adsorb preferentially on a rich

Figure 2. SEM images of SAPO-34 crystals synthesized using (a) PEG, (b) Brij-35, and (c) MB as CGI. (d) TEM image of SAPO-34 crystalssynthesized with Brij-35 as CGI.

Figure 3. CO2 and CH4 adsorption isotherms of SAPO-34 crystals synthesized using (a) PEG, (b) Brij-35, (c) MB as CGI, and (d) without CGI.

Letters J. Phys. Chem. B, Vol. 112, No. 51, 2008 16263

nitrogen basic zeolite framework surface. Therefore, in principle,one may fine-tune the CO2 and CH4 adsorption capacities ofthe crystals by simply adjusting the N/H content. The presenceof nitrogen and hydrogen in the SAPO-34 crystals is not unusualfor crystalline zeolite phases. In fact, it has been shown forcalcined SAPO-3416,25 that nitrogen and hydrogen from theamino group of the structure-directing agent form very stablesurface compounds bonded to the zeolite framework.

The formation of zeolite nanocrystals requires conditions thatfavor nucleation over crystal growth in the initial stages of theprocess. CGIs potentially interact with reactive sites of theinorganic precursors in solution, shortening the nucleation periodand thereby resulting in a larger number of smaller nuclei.26 Atthis stage, CGI separates these nuclei by adsorbing onto itssurface, inhibiting the growth of the crystals. Then, the adsorbedCGI decomposed at high temperature in later stages of thehydrothermal treatment. Therefore, a larger population of smallnuclei after the induction period explains the production ofsmaller crystals. A similar mechanism has been suggested forthe synthesis of pure-silica-zeolite MFI nanocrystals employingMB as a CGI.27 The incorporation of CGI in the synthesis gelalso increased the pH of the resultant solution. The pH of thesynthesis gel prepared without any CGI was 9.7 and increasedto ∼10 when PEG, Brij-35, and MB were incorporated. It iswell-known that alkalinity promotes the secondary nucleationat initial stages of the hydrothermal treatment and quicktransformation to a stable phase in later stages, leading to smallercrystals.1e,26 It is clear that CGIs did not act as structure-directingagents (SDAs), since the kinetic diameters of these moleculesare large enough to fit in the pore size of SAPO-34 (∼0.38nm). For instance, the MB molecule has a rectangular paral-lelepiped shape28 with dimensions of 1.6 × 0.7 × 0.37 nm3,which is much larger than the pore size of SAPO-34. Similarly,the micellar size of Brij-35 and PEG in water is ∼4 nm29 and∼0.69-3.93 nm,30 respectively. Due to their larger molecularsize, CGIs can only adsorb on the surface of the nuclei and donot enter into the porous system of SAPO-34, supporting theproposed mechanism.

In summary, SAPO-34 crystals displaying BET surface areasup to 700 m2/g, enhanced CO2/CH4 adsorption ratios, and crystalsizes in the ∼0.6-0.9 µm range, with narrow particle sizedistribution, employing PEG, Brij-35, and MB as crystal growthinhibitors were synthesized. The relatively high N/H molar ratiosobserved in the SAPO-34 crystals prepared with CGI led tohigh CO2/CH4 adsorption capacities. Due to its small crystalsize, narrow particle size distribution, higher surface area, andpreferential CO2 adsorption capacities over CH4, these crystalsrepresent ideal phases to prepare thin supported membranes forCO2/CH4 separations.

Acknowledgment. This research was supported in part by aUniversity of Louisville Intramural Research Incentive Grant.

We thank Dr. Jacek B. Jasinski, IAMRE, University ofLouisville, for his help in TEM experiments.

Supporting Information Available: Detailed synthesiscompositions of SAPO-34 crystals synthesized employing CGI.FTIR of SAPO-34 crystals prepared using CGI. SEM imagesof SAPO-34 crystals synthesized without using any CGI.Proposed route for the formation of SAPO-34 crystals in thepresence of CGI. This material is available free of charge viathe Internet at http://pubs.acs.org.

References and Notes

(1) (a) Poshusta, J. C.; Tuan, V. A.; Pape, E. A.; Noble, R. D.; Falconer,J. L. AIChE J. 2000, 46, 779. (b) Li, S.; Falconer, J. L.; Noble, R. D. J.Membr. Sci. 2004, 241, 121. (c) Li, S.; Falconer, J. L.; Noble, R. D. AdV.Mater. 2006, 18, 2601. (d) Carreon, M. A.; Li, S.; Falconer, J. L.; Noble,R. D. J. Am. Chem. Soc. 2008, 130, 5412. (e) Carreon, M. A.; Li, S.;Falconer, J. L.; Noble, R. D. AdV. Mater. 2008, 20, 729. (f) Li, S.; Falconer,J. L.; Noble, R. D. Microporous Mesoporous Mater. 2008, 110, 310.

(2) (a) Yu, M.; Li, S.; Falconer, J. L.; Noble, R. D. MicroporousMesoporous Mater. 2008, 110, 579. (b) Hong, M.; Li, S.; Falconer, J. L.;Noble, R. D. J. Membr. Sci. 2008, 307, 277.

(3) (a) Regli, L.; Bordiga, S.; Zecchina, A.; Bjorgen, M.; Lillerud, K. P.Stud. Surf. Sci. Catal. 2005, 155, 471. (b) Martins, G. A. V.; Berlier, G.;Coluccia, S.; Pastore, H. O.; Superti, G. B.; Gatti, G.; Marchese, L. J. Phys.Chem. C 2007, 111, 330.

(4) Zhou, H.; Wang, Y.; Wei, F.; Wang, D.; Wang, Z. Appl. Catal., A2008, 348, 135.

(5) (a) Park, J. W.; Lee, J. Y.; Kim, K. S.; Hong, S. B.; Seo, G. Appl.Catal., A 2008, 339, 36. (b) Wilson, S.; Barger, P. Microporous MesoporousMater. 1999, 29, 117.

(6) (a) Popova, M.; Minchev, Ch.; Kanazirev, V. Appl. Catal., A 1998,169, 227. (b) Inoue, M.; Dhupatemiya, P.; Phatanasri, S.; Inui, T.Microporous Mesoporous Mater. 1999, 28, 19. (c) Chen, D.; Moljord, K.;Fuglerud, T.; Holmen, A. Microporous Mesoporous Mater. 1999, 29, 191.(d) Lee, Y. J.; Baek, S. C.; Jun, K. W. Appl. Catal., A 2007, 329, 130.

(7) Oikawa, H.; Shibata, Y.; Inazu, K.; Iwase, Y.; Murai, K.; Hyodo,S.; Kobayashi, G.; Baba, T. Appl. Catal., A 2006, 312, 181.

(8) Elangovan, S. P.; Ogura, M.; Zhang, Y.; Chino, N.; Okubo, T. Appl.Catal., B 2005, 57, 31.

(9) Jaenchen, J.; Ackermann, D.; Weiler, E.; Stach, H.; Broesicke, W.Thermochim. Acta 2005, 434, 37.

(10) Lok, B. M.; Messina, C. A.; Patton, R. L.; Gajek, R. T.; Cannan,T. R.; Flanigen, E. M. J. Am. Chem. Soc. 1984, 106, 6092.

(11) (a) Vistad, Ø. B.; Akporiaye, D. E.; Taulelle, F.; Lillerud, K. P.Chem. Mater. 2003, 15, 1639. (b) Vistad, Ø. B.; Akporiaye, D. E.; Taulelle,F.; Lillerud, K. P. Chem. Mater. 2003, 15, 1650.

(12) Dumitriua, E.; Azzouza, A.; Huleaa, V.; Lutica, D.; Kesslerb, H.Microporous Mater. 1997, 10, 1.

(13) (a) Liu, G.; Tian, P.; Zhang, Y.; Li, J.; Xu, L.; Meng, S.; Liu, Z.Microporous Mesoporous Mater. 2008, 114, 416. (b) Liu, G.; Tian, P.;Zhang, Y.; Li, J.; Xu, L.; Meng, S.; Liu, Z. Microporous Mesoporous Mater.2008, 111, 143.

(14) (a) Wei, Y.; He, Y.; Zhang, D.; Xu, L.; Meng, S.; Liu, Z.; Su,B. L. Microporous Mesoporous Mater. 2006, 90, 188. (b) Zhou, H.; Wang,Y.; Wei, F.; Wang, D.; Wang, Z. Appl. Catal., A 2008, 341, 112.

(15) Nevenka, R.; Djordje, S.; Stanko, H.; Venceslav, K. Zeolites 1993,13, 384.

(16) Heyden, H. V.; Mintova, S.; Bein, T. Chem. Mater. 2008, 20, 2956.(17) (a) Yao, J.; Wang, H.; Ringer, S. P.; Chan, K. Y.; Zhang, L.; Xu,

N. Microporous Mesoporous Mater. 2005, 85, 267. (b) Yao, J.; Zeng, C.;Zhang, L.; Xu, N. Mater. Chem. Phys. 2008, 112, 637. (c) Zhang, L.; Yao,J.; Zeng, C.; Xu, N. Chem. Commun. 2003, 2232.

(18) Li-ming, K.; Xiao-qin, L.; Ding-hua, L. J. Nanjing UniV. Sci.Technol. 2007, 31, 528.

(19) Mertens, M.; Strohmaier, K. G. U.S. Patent 6696032 B2, 2004.(20) (a) Rivera-Ramos, M. E.; Ruiz-Mercado, G. J.; Hernandez-

Maldonado, A. J. Ind. Eng. Chem. Res. 2008, 47, 5602. (b) Uzunova, E. L.;Mikosch, H.; Hafner, J. J. Phys. Chem. C 2008, 112, 2632. (c) Wei, Y.;Zhang, D.; Xu, L.; Chang, F.; He, Y.; Meng, S.; Su, B. L.; Liu, Z. Catal.Today 2008, 131, 262. (d) Wei, Y.; He, Y.; Zhang, D.; Xu, L.; Liu, S. M. Z.;Su, B. L. Microporous Mesoporous Mater. 2006, 90, 188.

(21) Li, Q.; Mihailova, B.; Creaser, D.; Sterte, J. Microporous Meso-porous Mater. 2001, 43, 51.

(22) See the Supporting Information.(23) Szostak, R. Handbook of molecular sieVes; Van Nostrand Reinhold:

New York, 1992; p 416.(24) Akolekar, D. B.; Bhargava, S.; Bronswijk, W. V. Appl. Spectrosc.

1999, 53, 931–937.

Figure 4. CO2/CH4 adsorption ratio as a function of N/H molar ratio.

16264 J. Phys. Chem. B, Vol. 112, No. 51, 2008 Letters

(25) Marchese, L.; Frache, A.; Ginotti, E.; Martra, G.; Causa, M.;Coluccia, S. Microporous Mesoporous Mater. 1999, 30, 145.

(26) Hosokowa, H.; Oki, K. Chem. Lett. 2003, 32, 586.(27) Lew, C. M.; Li, Z.; Zones, S. I.; Sun, M.; Yan, Y. Microporous

Mesoporous Mater. 2007, 105, 10.(28) Hoppe, R.; Schulz-Ekloff, G.; Rathousky, J.; Starek, J.; Zukal, A.

Zeolites 1994, 14, 126.

(29) Garcia, J. M.; Wick, L. Y.; Harms, H. EnViron. Sci. Technol. 2001,35, 2033.

(30) Wang, S. C.; Wang, C. K.; Chang, F. M.; Tsao, H. K. Macromol-ecules 2002, 35, 9551.

JP809316S

Letters J. Phys. Chem. B, Vol. 112, No. 51, 2008 16265