A top-down approach to prepare silicoaluminophosphate molecular sieve nanocrystals with improved catalytic activity

Article abstract of DOI:10.1039/c3cc48264b

Silicoaluminophosphate SAPO-34 molecular sieve nanocrystals have been prepared by a post-synthesis milling and recrystallization method, which is further proven to be universally applicable to other SAPO molecular sieves. The obtained SAPO-34 with reduced Si enrichment on the external surface shows considerably improved catalytic performance in the MTO reaction.

Full text of DOI:10.1039/c3cc48264b

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
A top-down approach to prepare
silicoaluminophosphate molecular sieve
nanocrystals with improved catalytic activity†
Cite this: Chem. Commun., 2014,
50, 1845
Received 29th October 2013,
Accepted 16th December 2013
Miao Yang,a Peng Tian,a Chan Wang,ab Yangyang Yuan,ab Yue Yang,a Shutao Xu,a
Yanli Hea and Zhongmin Liu*a
DOI: 10.1039/c3cc48264b
www.rsc.org/chemcomm
Silicoaluminophosphate SAPO-34 molecular sieve nanocrystals
The traditional top-down method has been widely used to create
have been prepared by a post-synthesis milling and recrystallization mesopores in aluminosilicate zeolites through post treatment of
method, which is further proven to be universally applicable to zeolites in an acidic (dealumination) or basic (desilication) solution.12
other SAPO molecular sieves. The obtained SAPO-34 with reduced However, it is difficult to employ this method to modify SAPO
Si enrichment on the external surface shows considerably improved molecular sieves because the frameworks of SAPOs are not as stable
catalytic performance in the MTO reaction.
as those of aluminosilicates during acidic/basic treatment. Recently,
a novel top-down approach has been proposed to prepare nanosized
Molecular sieves have been widely used in industries as hetero- zeolite A and ZSM-5.13 The zeolite crystals were first crushed by bead
geneous catalysts.1 Their regular channel structures, which are in milling. After that, the crystallinity was recovered by recrystallizaion of
the size range of many molecules, make them excellent shape- the milled precursor in a dilute aluminosilicate solution.
selective catalysts.2 However, the molecular dimension of the
In this work, we successfully developed the post-synthesis milling
micropores may suffer diffusion limitation and cause fast coke and recrystallization method to prepare small SAPO-34 crystals. Both
deposition.3 Preparation of nanoscale molecular sieves or hierar- the mother liquid collected from SAPO-34 synthesis and the self-
chical molecular sieves is considered to be an effective solution, made solution could be employed as the recrystallization solution.
which could improve the catalytic efficiency because of the The obtained SAPO-34 crystallites showed prolonged catalytic life-
shortened diffusion paths and increased external surface area.4
SAPO-34, one of the most important members in the SAPO family,
time and higher selectivity to light olefins in the MTO reaction.
Micrometer-sized SAPO-34 was first synthesized using DEA as a
has exhibited excellent catalytic performance in the methanol-to- template according to our previous work.14 The SAPO-34 precursor
olefins (MTO) reaction due to the contribution of the CHA structure, was calcined followed by milling using a planetary ball mill with water
medium/strong acidity, and high hydrothermal stability.5 It is as a dispersing phase. After that, the milled SAPO-34 was hydro-
acknowledged that decreasing the crystal size of SAPO-34 could thermally recrystallized in a mother liquid or a self-made alumino-
improve its MTO catalytic lifetime.6 Therefore, the synthesis of phosphate solution at 180 1C for 2–24 hours (Table 1). It should be
nano-SAPO-34 is very attractive from both academic and practical noted that the small crystal size remained almost unchanged even
perspectives. Many synthetic strategies, such as the colloidal solution after recrystallization at 200 1C for 24 hours (Fig. S1, ESI†). The Si
method,7 dry gel conversion,8 microwave,9 sonochemical10 and two-
step hydrothermal crystallized methods,11 have been tried to obtain
small SAPO-34 crystals. All these processes are bottom-up methods,
Table 1 Preparation of SAPO-34 by the recrystallization methoda
Recrystallization
and involve the use of expensive tetraethylammonium hydroxide
(TEAOH) as the template. Other organic templates such as triethyl-
amine (TEA), diethylamine (DEA), and morpholine (MOR) generally
lead to micrometer-sized SAPO-34 crystals.
Sample solution P/Al/Si/TEA/H2O Product composition T (1C)/t (h)
1b
2b
3b
4
5
6
1.5/1/0.08/–/73
1.5/1/0.08/–/73
1.5/1/0.08/–/73
1.0/1.0/0/1.9/75
1.0/1.0/0/1.9/75
1.0/1.0/0/1.75/27.5
Al0.498Si0.116P0.386O2
Al0.485Si0.136P0.379O2
Al0.482Si0.148P0.370O2
Al0.575Si0.117P0.307O2
Al0.537Si0.124P0.339O2
Al0.508Si0.088P0.404O2
180/2
180/2
180/2
180/3
180/24
200/24
a National Engineering Laboratory for Methanol to Olefins, Dalian National
Laboratory for Clean Energy, Dalian Institute of Chemical Physics, Chinese
Academy of Sciences, Dalian 116023, P. R. China. E-mail: liuzm@dicp.ac.cn
b University of Chinese Academy of Sciences, Beijing, P. R. China
a
SAPO-34 as the precursor (Al0.448Si0.200P0.351O2), solid/recrystallization
solution mass ratio was 1/15 for sample 1, 1/10 for samples 2, 4–6 and
b
† Electronic supplementary information (ESI) available: More details of experi-
mental and characterization processes, TG, XRD, SEM, NH3-TPD, 27Al and 31P MAS
NMR spectra and the MTO reaction catalytic results. See DOI: 10.1039/c3cc48264b
1/5 for sample 3. Mother liquid as the recrystallization solution,
which was obtained from SAPO-34 synthesis using TEA as a template.
For more details see ESI.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 1845--1847 | 1845

View Article Online
Communication
ChemComm
Fig. 1 XRD patterns of (a) SAPO-34 precursor, (b) milled SAPO-34, and
(c) recrystallized SAPO-34 (sample 2).
Fig. 2 SEM micrographs of (a) SAPO-34 precursor, (b) milled SAPO-34,
and (c and d) recrystallized SAPO-34 (sample 2).
content in the product, which had great effects on the acidity of the
sample (catalytic properties), could be tuned by simply modifying the
reaction conditions. A similar self-made solution using DEA or MOR
as the template could also accomplish the recrystallization, but gave
SAPO-34 products with larger crystal size (0.2–2.0 mm). In addition,
using NaOH or NH3ÁH2O as the basic resource only resulted in the
formation of amorphous material or ammonium aluminum phos-
phate, suggesting that the organic template was necessary for the
structure recovery. In fact, TG analysis (Fig. S2, ESI†) of sample 2 gave
a total weight loss of 17.0 wt% including around 4 wt% water and
13 wt% organic template. This is comparable to that of the conven-
tional SAPO-34, confirming the incorporation of an organic template
into the recrystallized product.
The XRD patterns of the SAPO-34 precursor, together with its
milled and recrystallized samples, are given in Fig. 1. The precursor
showed typical diffraction peaks of CHA structure. The peak inten-
sity of the milled sample decreased a lot indicating a destruction of
Fig. 3 29Si MAS NMR spectra of (a) SAPO-34 precursor, (b) milled SAPO-34,
and (c) recrystallized SAPO-34 (sample 2).
the crystal structure. Upon recrystallization, the product regained the in Fig. 3, the 29Si spectra of the precursor gave one strong peak at
high quality XRD pattern in spite of some broadening of the peaks, À91.6 ppm due to Si(OAl)4 species, and several overlapped peaks
suggesting the recovery of the crystal structure and the reduced between À95.4 and À110.3 ppm ascribed to Si(OAl)n(OSi)4Àn (n = 3–0)
particle size. Fig. 2 shows the SEM images of the three SAPO species. Only a broad signal was observed in the spectrum of the
samples. A typical rhombohedral morphology was observed for the milled sample, corresponding to its low crystallinity. The 29Si spec-
SAPO-34 precursor, and the particle size was around 4–8 mm with trum of the recrystallized SAPO-34 (sample 2) looked more like a
tiny smaller crystals. After milling, the regular morphology disap- superimposed one from those of the raw and milled samples. The
peared forming disordered agglomerates instead. Nanoscale parti- shoulder peaks at –83 and À86 ppm in the spectrum should arise
cles with cubic-like morphology (50–350 nm in size) could be from Si(OAl)3(OH) and Si(OAl)2(OH)2 species respectively.15 Moreover,
recognized after recrystallization, in agreement with the XRD results. the broad peaks between À90 and À120 ppm may also contain the
The N2 adsorption properties of the samples are presented in contribution of Si species connecting with OH groups. The 27Al and
Fig. S3 and Table S2 (ESI†). The precursor exhibits typical adsorption 31P spectra of the samples are given in Fig. S4 (ESI†). They showed a
properties of SAPO-34, while the milled sample has a very low N2 similar change to the 29Si spectra after milling and recrystallization
adsorption ability suggesting that the CHA structure has been treatment where penta- (10 ppm) and hexacoordinated (À10 ppm) Al
seriously damaged by milling. Notably, the isotherms of recrystallized species and not fully condensed P species (À12 to À20 ppm) can be
samples 2 and 6 regain the type-I curve illustrating the rebuilding of observed. These results implied that there existed more defects in the
microporous structure. The increased external surface area confirmed recrystallized framework. After calcination, sample 2 exhibited a
the small size of the recrystallized product, and the high mesopore narrow tetrahedral 27Al peak like the precursor, indicating that the
volume was associated with multilayer adsorption between the sample was free of extra-framework Al atoms.
intercrystal pores of nanoparticles, as reflected in the isotherm with
A remarkable feature of the present SAPO-34 crystallites is their
improved catalytic properties in the MTO reaction (Fig. 4 and
an abrupt inclination at relatively high pressure (P/P0 > 0.9).
Solid-state 29Si, 27Al, and 31P MAS NMR spectra were obtained to Table S1, ESI†). Compared with the performance of the SAPO-34
investigate the local atomic environments in the samples. As shown precursor, the catalytic lifetime of sample 2 was dramatically
1846 | Chem. Commun., 2014, 50, 1845--1847
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
also confirmed by comparing with the surface composition of SAPO-
34-TEA, which was synthesized by a conventional method and had
similar bulk composition to sample 6 (Table 2). Based on the above
results, it is thus concluded that the prolonged catalyst lifetime,
improved selectivity to light olefins together with suppressed propy-
lene hydrogenation to propane should be related to a combination of
enhanced utilization of the internal pore space and reduced acidity
caused by the reduced Si content especially the Si content on the
external surface of the nanosized catalysts.
Besides the success in preparation of SAPO-34 nanocrystals,
nanoscale SAPO-5 and SAPO-35 have been synthesized through
this top-down route; for more details see ESI† (Fig. S7 and S8).
These results demonstrate that the above method is universally
applicable to SAPO molecular sieves.
In summary, SAPO molecular sieves with small crystal size have
been prepared by an effective post-synthesis milling and recrystalliza-
tion method. The mother liquid from zeolite production could be
Fig. 4 Methanol conversion and selectivity of C2H4 plus C3H6 in the MTO
reaction on the SAPO-34 precursor, sample 2 and 6. Reaction conditions:
450 1C, WHSV = 4.0 hÀ1, 40 wt% methanol solution.
extended from 38 to 224 min, and the selectivity of ethylene plus directly used as a recrystallization solution. The Si content in the
propylene was increased from 71.6 to 82.6%. Meanwhile, the selec- product can be tuned by changing the recrystallization conditions. The
tivity to propane (the secondary product from propylene) was reduced considerably improved catalytic performance of SAPO-34 nanocrystals
from 6.3% to 0.6%. Similar catalytic results were observed for sample in the MTO reaction resulted from the shortened diffusion paths,
6 recovered from the self-made recrystallization solution, which decreased acid concentration, and reduced Si enrichment on the
showed a lifetime of 258 min and 83.2% selectivity of ethylene plus crystal surface. More useful nanosized molecular sieves are expected
propylene. The catalytic performance of the milled sample was to be prepared using this facile and promising top-down strategy.
also tested, which showed very low activity (methanol conversion o
We thank the financial support from National Natural
5 wt%, Fig. S5, ESI†). NH3-TPD experiments were carried out to probe Science Foundation of China (21101150 and 20901076).
the acid properties of the samples, and the curves are given in Fig. S6
(ESI†). The high-temperature desorption peak for the recrystallized
SAPO-34 shifted to lower temperature together with decreased inten-
sity as compared with that of the precursor, suggesting the reduced
acid strength and amount. This was consistent with the reduced Si
Notes and references
1 A. Corma, Chem. Rev., 1997, 97, 2373.
2 (a) Z. Wang, J. Yu and R. Xu, Chem. Soc. Rev., 2012, 41, 1729; (b) J. Yu
and R. Xu, Acc. Chem. Res., 2010, 43, 1195.
content in the recrystallized samples. Moreover, the surface composi-
tions of the samples were also examined by XPS (Table 2), because of
the fact that the Si content and its coordination environment at the
crystal surface are very important for the MTO reaction perfor-
mance.16 Clearly, there existed an obvious Si enrichment on the
external surface of the precursor, reflecting a gradual increase of the
Si content in SAPO crystals from the core to the surface.17 After
milling, the Si enrichment phenomenon was weakened due to the
exposure of the internal surface to a relatively low Si content. The Si
content on the crystal surface was further decreased after recrystalli-
3 A. Corma, J. Catal., 2003, 216, 298.
4 (a) L. Tosheva and V. P. Valtchev, Chem. Mater., 2005, 17, 2494;
(b) K. Na, M. Choi and R. Ryoo, Microporous Mesoporous Mater., 2013,
166, 3; (c) V. Valtchev and L. Tosheva, Chem. Rev., 2013, 113, 6734.
5 (a) 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; (b) J. Liang,
H. Y. Li, S. Zhao, W. G. Guo, R. H. Wang and M. L. Ying, Appl. Catal.,
1990, 64, 31; (c) A. J. Marchi and G. F. Froment, Appl. Catal., 1991,
71, 139.
6 G. Yang, Y. Wei, S. Xu, J. Chen, J. Li, Z. Li, J. Yu and R. Xu, J. Phys.
Chem. C, 2013, 117, 8214.
7 H. van Heyden, S. Mintova and T. Bein, Chem. Mater., 2008, 20, 2956.
8 Y. Hirota, K. Murata, S. Tanaka, N. Nishiyama, Y. Egashira and
K. Ueyama, Mater. Chem. Phys., 2010, 123, 507.
zation, and the R value (the ratio of Si/(Al + P)surface over Si/(Al + P)bulk
)
9 S. Lin, J. Li, R. P. Sharma, J. Yu and R. Xu, Top. Catal., 2010, 53, 1304.
10 S. Askari and R. Halladj, Ultrason. Sonochem., 2012, 19, 554.
of the recrystallized samples was obviously lower than that of the
precursor, suggesting a weakened surface Si enrichment. This was 11 P. Wang, D. Yang, J. Hu, J. a. Xu and G. Lu, Catal. Today, 2013,
212, 62.e1.
12 V. Valtchev, G. Majano, S. Mintova and J. Perez-Ramirez, Chem. Soc.
Table 2 Elemental composition from XPS analysis
Surface elemental composition (mol%)
Rev., 2013, 42, 263.
13 (a) T. Wakihara, A. Ihara, S. Inagaki, J. Tatami, K. Sato, K. Komeya,
T. Meguro, Y. Kubota and A. Nakahira, Cryst. Growth Des., 2011,
11, 5153; (b) T. Wakihara, R. Ichikawa, J. Tatami, A. Endo, K. Yoshida,
Y. Sasaki, K. Komeya and T. Meguro, Cryst. Growth Des., 2011, 11, 955.
14 G. Liu, P. Tian, Y. Zhang, J. Li, L. Xu, S. Meng and Z. Liu, Microporous
Mesoporous Mater., 2008, 114, 416.
15 (a) Z. Yan, B. Chen and Y. Huang, Solid State Nucl. Magn. Reson.,
2009, 35, 49; (b) L. Zhang, J. Bates, D. Chen, H.-Y. Nie and Y. Huang,
J. Phys. Chem. C, 2011, 115, 22309.
Sample
Al
Si
P
Ra
SAPO-34 precursor 30.7
44.7
28.9
21.9
13.5
20.6
24.6
31.7
38.3
44.0
34.1
3.23
—
1.78
1.63
2.76
Milled sample
Sample 2
39.4
39.8
42.5
45.3
Sample 6
SAPO-34-TEAb
16 D. Mores, E. Stavitski, M. H. F. Kox, J. Kornatowski, U. Olsbye and
B. M. Weckhuysen, Chem.–Eur. J., 2008, 14, 11320.
a
b
R = Si/(Al + P)surface/Si/(Al + P)bulk
.
SAPO-34 synthesized with TEA as a
template for comparison. Its bulk composition (Al0.486Si0.080P0.434O2) is 17 P. Tian, B. Li, S. Xu, X. Su, D. Wang, L. Zhang, D. Fan, Y. Qi and
similar to that of sample 6.
Z. Liu, J. Phys. Chem. C, 2013, 117, 4048.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 1845--1847 | 1847