Dual template-directed synthesis of SAPO-34 nanosheet assemblies with improved stability in the methanol to olefins reaction

Article abstract of DOI:10.1039/c4ta06124a

A spherical self-assembly of SAPO-34 nanosheets was hydrothermally synthesized by using a quaternary ammonium-type organosilane surfactant [3-(trimethoxysilyl)propyl]octadecyldimethylammonium chloride (TPOAC) as the mesoporogen and a part of the silica source, and diethylamine (DEA) as the microporous template. The prepared materials were well characterized by XRD, XRF, SEM, TEM, N₂ adsorption-desorption, NH₃-TPD, NH₃-adsorbed IR and pyridine-adsorbed IR measurements. It revealed that the morphologies, compositions and acidity of the SAPO-34 products changed significantly with the increased TPOAC/TEOS ratio, showing that the TPOAC/TEOS ratio had a significant impact on the crystallization process of SAPO-34. Solid state ¹³C, ²⁹Si NMR and TG-DTA analyses were further conducted to investigate the status of TPOAC in the final products. A possible crystallization process involving TPOAC was proposed. The catalytic performances of the synthesized SAPO-34s were evaluated using the MTO reaction. The optimal SAPO-34 nanosheet assemblies with a hierarchical porous structure displayed a remarkably enhanced catalytic lifetime and high yields of light olefins.

Full text of DOI:10.1039/c4ta06124a

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
Dual template-directed synthesis of SAPO-34
nanosheet assemblies with improved stability in the
Cite this: J. Mater. Chem. A, 2015, 3,
methanol to olefins reaction†
5608
Chan Wang,abc Miao Yang,ab Peng Tian,ab Shutao Xu,ab Yue Yang,ab Dehua Wang,abc
ab
Yangyang Yuanabc and Zhongmin Liu
*
A spherical self-assembly of SAPO-34 nanosheets was hydrothermally synthesized by using a quaternary
ammonium-type
organosilane
surfactant
[3-(trimethoxysilyl)propyl]octadecyldimethylammonium
chloride (TPOAC) as the mesoporogen and a part of the silica source, and diethylamine (DEA) as the
microporous template. The prepared materials were well characterized by XRD, XRF, SEM, TEM, N2
adsorption–desorption, NH3-TPD, NH3-adsorbed IR and pyridine-adsorbed IR measurements. It revealed
that the morphologies, compositions and acidity of the SAPO-34 products changed significantly with the
increased TPOAC/TEOS ratio, showing that the TPOAC/TEOS ratio had a significant impact on the
crystallization process of SAPO-34. Solid state 13C, 29Si NMR and TG-DTA analyses were further
conducted to investigate the status of TPOAC in the final products. A possible crystallization process
involving TPOAC was proposed. The catalytic performances of the synthesized SAPO-34s were
evaluated using the MTO reaction. The optimal SAPO-34 nanosheet assemblies with a hierarchical
porous structure displayed a remarkably enhanced catalytic lifetime and high yields of light olefins.
Received 12th November 2014
Accepted 28th January 2015
DOI: 10.1039/c4ta06124a
www.rsc.org/MaterialsA
microporous structures is considered as an attractive solution.
Wei et al. obtained hierarchical SAPO-34 composed of decussate
1 Introduction
zeolite slice units by using natural layered kaolin as a special Si
and Al source.14 The enhanced conversion and olen selectivity
in the dimethyl ether to olen reaction are attributed to the
unique hierarchical structure. Xie et al. prepared a hierarchical
SAPO-34 monolith through a dry gel conversion of the amor-
phous silicoaluminophosphate monolith.18 Schmidt et al.
synthesized hierarchical SAPO-34 by using carbon materials as
hard templates.19 Yu et al. prepared the hierarchically macro-
porous SAPO-34 by using an in situ growth-etching approach.20 It
is surprising that although the so-template directing method is
very popular in the synthesis of mesoporous aluminosilicate
zeolites,21–25 it is not well developed to synthesize mesoporous
aluminophosphate and silicoaluminophosphate molecular
sieves. Progress has been made in the synthesis of mesoporous
SAPO-34 by using certain polymers as mesopore-directing
agents,26,27 but obtaining a high-quality silicoaluminophosphate
catalyst with uniform mesopores through this method still poses
many challenges. Quaternary ammonium-type organosilane
surfactants are one class of the most promising mesopore-
generating agents for the synthesis of mesoporous zeolites. They
can simultaneously act as parts of the silica source and
mesoporogens during the synthesis, preventing the phase sepa-
ration of zeolites and mesoporous materials.21,28,29 By using
3-(trimethoxysilyl)propyl hexadecyldimethylammonium chloride
([(CH3O)3SiC3H6N(CH3)2C16H33]Cl, TPHAC), Ryoo et al. success-
fully synthesized mesoporous AlPO-5, AlPO-11 and SAPO-5;
Silicoaluminophosphate (SAPO) molecular sieves, invented by
Union Carbide Corporation in the 1980s, are important crys-
talline materials.1–4 Incorporation of silicon atoms into the
neutral AlPO4 lattice generates acidity, which gives SAPO
molecular sieves many signicant applications as catalysts,
catalyst supports and adsorbents.5–9 SAPO-34 with a CHA
topology is one of the most investigated silicoaluminophos-
phate molecular sieves due to its excellent performance in the
methanol-to-olen (MTO) reaction (100% methanol conversion
and over 90% selectivity to C2–C4 light olens).10–12 Unfortu-
˚
nately, the small 8-ring pore opening (3.8 ꢀ 3.8 A) and large
˚
CHA cavity (9.4 A in diameter) structure cause a severe diffusion
limitation resulting in rapid deactivation by coke deposition
during the MTO process. It is highly desirable to improve the
stability of the SAPO-34 catalyst without sacricing its inherent
selectivity and catalytic activity.
Considerable efforts have been made to optimize the SAPO-34
catalyst,13–17 and integration of meso- or/and macropores into the
aDalian National Laboratory for Clean Energy, Dalian Institute of Chemical Physics,
Chinese Academy of Sciences, Dalian, P. R. China. E-mail: liuzm@dicp.ac.cn
bNational Engineering Laboratory for Methanol to Olens, Dalian Institute of
Chemical Physics, Chinese Academy of Sciences, Dalian, P. R. China
cUniversity of Chinese Academy of Sciences, Beijing, P. R. China
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c4ta06124a
5608 | J. Mater. Chem. A, 2015, 3, 5608–5616
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry A
however, hierarchical SAPO-5 lacks strong acid sites, though a
The yields of products were calculated using the following
high silica content (molar composition of Al0.50P0.14Si0.36O2) is equation: yield ¼ (Msample ꢀ 75%)/M(Al O
ꢀ 100%,
+ P2O5 + SiO2)gel
3
2
claimed.30 In 2010, Wang et al. synthesized hierarchical SAPO-34 where Msample and M(Al O
stand for the weights of
+ P2O5 + SiO2)gel
3
2
with
[3-(trimethoxysilyl)propyl]octadecyldimethylammonium the products and the dry mass of inorganic oxides in the
chloride (TPOAC) as the mesoporous template and sole silica starting mixture respectively. 75% is the estimated average
source, together with tetraethylammonium hydroxide (TEAOH) value of framework compounds included in the samples.
as the microporous director.31 Still, the obtained material had
much weaker acidity than conventional SAPO-34 molecular
sieves. Although abundant mesopores are obtained, the appli-
2.2 Characterization
cation of SAPO-34 as a catalyst is limited. Very recently, Yu et al. The powder XRD pattern was recorded on a PANalytical X' Pert
˚
reported success in the preparation of hierarchical SAPO-34 with PRO X-ray diffractometer with Cu-Ka radiation (l ¼ 1.54059 A),
enhanced MTO catalytic performance also by using TPOAC as operating at 40 kV and 40 mA. The chemical composition of the
the mesoporous director and morpholine (MOR) as the micro- samples was determined with a Philips Magix-601 X-ray uo-
porous template.32 The disparate results from the same meso- rescence (XRF) spectrometer. The crystal morphology was
porogen TPOAC suggest that the microporous template may also observed by scanning electron microscopy (SEM) using a
play a crucial role in the synthesis of hierarchical SAPO-34. In TM3000 (Hitachi) and eld emission SEM (Hitachi SU8020).
addition, the role of TPOAC as a silica source during the crys- Transmission electron microscopy (TEM) images and selected
tallization process still deserves investigation.
area electron diffraction (SAED) patterns were recorded with a
In this work, exploration on the synthesis of hierarchical JEM-2100 electron microscope. Textural properties of the
SAPO-34 was continued by using TPOAC as the mesoporogen calcined samples were determined by N2 adsorption–desorp-
and diethylamine (DEA) as the microporous template. Sphere- tion at 77 K on a Micromeritics ASAP 2020 system. The total
like SAPO-34 molecular sieves consisting of interpenetrating surface area was calculated based on the BET equation. The
nanosheets were synthesized and well characterized. The micropore volume and micropore surface area were evaluated
inuence of various synthetic parameters, roles of the dual using the t-plot method. Mesopore volume and mesopore
templates as well as the catalytic properties of SAPO-34s were surface area were evaluated from the adsorption isotherm by
investigated in detail. The deactivation behavior of the hierar- the Barrett–Joyner–Halenda (BJH) method. The temperature-
chical SAPO-34 catalyst was studied and discussed.
programmed desorption of ammonia (NH3-TPD) was carried
out with an AutochemII2920 equipment (Micromeritics). 0.2
grams of the sample particles (40–60 mesh) were loaded into a
U-quartz tube and pretreated at 650 ꢁC for 60 min under He
ow. Aer cooling down to 100 ꢁC, a gas mixture of NH3 and He
ow was introduced to saturate the sample surface with NH3
2 Experimental
2.1 Synthesis
Orthophosphoric acid (H3PO4, 85 wt%), tetraethyl orthosilicate adsorption (60 min). Aer this, He ow was purged through the
(TEOS, 98 wt%), aluminium isopropoxide (Al(i-C3H7O)3, 99 sample for 30 min to remove the weakly adsorbed NH3 mole-
wt%) and diethylamine (DEA, 99 wt%) were purchased from cules. The measurement of the desorbed NH3 was performed
Tianjin Kemiou Chemical Reagent Company. Silica sol (30.1 from 100 ꢁC to 650 ꢁC (10 ꢁC minꢂ1) under He ow (20 ml
wt%) was purchased from Shenyang Chemical Company, and minꢂ1). 13C magic-angle-spinning (MAS) nuclear magnetic
pseudoboehmite (72.5 wt%) was purchased from Fushun resonance (NMR), 29Si cross-polarization (CP)/MAS NMR and 1H
Petroleum No. 3 Factory. [3-(Trimethoxysilyl)propyl]octade- MAS NMR analyses were carried out on a Bruker AvanceIII 600
cyldimethylammonium chloride ([(CH3O)3SiC3H6N(CH3)2- spectrometer at 150.9 MHz for 13C, 119.2 MHz for 29Si and
1
C
18H37]Cl, TPOAC, 65 wt%) was purchased from Aladdin 600.13 MHz for H. 29Si CP/MAS NMR spectra were measured
Chemical Incorporation. All chemicals were used as received with a contact time of 3 ms and a recycle delay of 2 s. 29Si CP/
without further purication. MAS NMR spectra were recorded with a 4 mm MAS probe with a
A certain amount of TPOAC was rstly added into deionized spinning rate of 8 kHz. 13C MAS NMR spectra and 1H MAS NMR
water and stirred overnight. Then, the desired amounts of Al(i- spectra were recorded with a 4 mm MAS probe at a spinning rate
C3H7O)3, DEA, TEOS and H3PO4 were added in sequence. Aer of 12 kHz. Chemical shis were referenced to 4,4-dimethyl-4-
further stirring for 2–5 h, the obtained gels with a molar silapentane sulfonate sodium salt (DSS) for 29Si, and ada-
composition of (2.0–2.5) DEA : 1.0Al2O3 : 1.0P2O5 : (0.4–0.6) mantane for 13C and 1H. Thermogravimetric and differential
SiO2 : 150H2O were transferred into stainless steel autoclaves, thermal analysis (TG and DTA) were performed on a TA
and heated at 200 ꢁC for 72 h statically under autogenous SDTQ600 analyzer with a temperature-programmed rate of 10
pressure. The products were ltered, washed thoroughly and ꢁC minꢂ1 in air. The ammonia-adsorbed infrared (NH3-adsor-
dried in air. Then the samples were calcined at 600 ꢁC for 2 h bed IR) and pyridine-adsorbed IR spectra were carried out on a
to remove the template if necessary. For comparison, Bruker Tensor 27 instrument. The calcined samples were
conventional SAPO-34 with a similar silica content was dehydrated at 500 ꢁC for 1 h under a vacuum of 1.0 ꢀ 10ꢂ2 Pa,
prepared by using the same template DEA. The detailed and then pure ammonia or pyridine vapor was adsorbed at
synthesis and characterization information are provided in room temperature for 10 min. Aer reaching equilibrium, the
the ESI.†
ammonia-adsorbed system was evacuated at 200 ꢁC, 300 ꢁC and
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A, 2015, 3, 5608–5616 | 5609

View Article Online
Journal of Materials Chemistry A
Paper
400 ꢁꢁC, and the pyridine-adsorbed system was evacuated at crystallization process, which disappears aer a considerable
ꢁ
100 C and 300 C. The IR spectra were recorded at the corre- reaction time (Fig. S3 and S4†). In addition, the addition of
sponding temperatures, respectively.
TPOAC slows down the crystallization rate signicantly, while it
takes at least 72 hours in order to obtain a SAPO-34 phase with
satisfactory crystallinity. Comparatively, well-crystallized SAPO-
34 with a minor DNL-6 impurity could be obtained in 24 h
under the same conditions but without TPOAC (sample 5,
Fig. S5†). Keeping the SiO2/Al2O3 ratio of the gel at 0.5, the
molar ratio of TPOAC/TEOS was adjusted from 1/9 to N. All
synthesized samples exhibit the typical XRD peaks of SAPO-34
with high intensity except sample 9. Sample 9 is synthesized by
using TPOAC as the only silica source, which leads to SAPO-5
with low crystallinity. More interestingly, samples 6–8 have an
obviously increased silica content due to the higher TPOAC/
TEOS molar ratio. It is deduced that the interaction between
TPOAC and TEOS may enhance the activity of the silica source
which promotes the introduction of silica atoms into the
framework. When TEOS is absent, the activity of TPOAC is low,
and only a small amount of the Si source is involved in the
reaction which leads to the SAPO-5 phase.
2.3 Catalytic test and retained coke analysis
The methanol to olens (MTO) reaction was performed in a
quartz tubular xed-bed reactor at atmospheric pressure. 0.3 g
calcined catalyst (40–60 mesh) was loaded in the quaꢂrt1z reactor
and activated at 550 ꢁC in a He ow of 30 ml min for 1 h
before starting each reaction run, and then the temperature was
adjusted to a reaction temperature of 450 ꢁC. Methanol was fed
by passing the carrier gas (42 ml minꢂ1) through a saturator
containing methanol at 30 ꢁC, which gave a weight hourly space
velocity (WHSV) of 3 hꢂ1. The reaction products were analyzed
using an online gas chromatograph (Agilent GC 7890N),
equipped with a ame ionization detector (FID) and Plot-Q
column.
The retained organics in the deactivated SAPO-34 catalysts
aer the MTO reaction were analyzed according to the proce-
dure introduced by Guisnet.33 The completely deactivated
catalysts were rst dissolved with a HF solution (20 wt%), and
then the organic compounds present in them were extracted by
using CH2Cl2 and analyzed by a gas chromatograph equipped
with a FID detector and a mass sensitive detector (Agilent
7890A/5975C) using a capillary column of HP-5.
The morphologies of the SAPO-34 products change much
with the increased molar ratio of TPOAC/TEOS. As seen in Fig. 2,
sample 6 presents as spherical agglomerates composed of many
micrometer-sized cubic crystals, while sample 7 evolves into a
self-assembly of spherical nanosheets as the TPOAC/TEOS ratio
3 Results and discussion
3.1 Synthesis and characterizations
The synthesis conditions of hierarchical SAPO-34 and the cor-
responding results are listed in Table 1. The selected XRD
patterns are shown in Fig. 1. A pure SAPO-34 phase can be
prepared when the molar ratio of SiO2/Al2O3 is higher than 0.4,
and the DEA/Al2O3 ratio is above 2.0. The yields are around 42%,
which are slightly lower than that of SAPO-34 synthesized
without TPOAC (48.2%, sample 5) due to the introduction of the
mesoporogen. It is noted that a lower SiO2/Al2O3 ratio and
insufficient DEA result in impurities such as SAPO-11 (sample 1)
and SAPO-5 (sample 2), respectively. Even under optimized
Fig. 1 XRD patterns of as-synthesized samples 6 (a), 7 (b), 8 (c) and
conditions, DNL-6 appears as a metastable phase during the 9 (d).
Table 1 Synthesis conditions,a yields and composition of products analyzed by XRF
Sample
x
y
TPOAC/TEOS
Product
Product yield
Product composition
1
2
3
4
5b
6
7
8
9
0.4
0.5
0.5
0.6
0.5
0.5
0.5
0.5
0.5
2.3
2.0
2.5
2.3
2.3
2.3
2.3
2.3
2.3
1/3
1/3
1/3
1/3
0
1/9
1/3
1/1
N
SAPO-34 + SAPO-11
SAPO-34 + SAPO-5
SAPO-34
SAPO-34
SAPO-34 + DNL-6
SAPO-34
SAPO-34
SAPO-34
SAPO-5 + amorphous
42.3%
43.3%
43.6%
45.5%
48.2%
42.1%
41.5%
39.4%
—
—
—
Al0.486P0.361Si0.153O2
Al0.467P0.367Si0.166O2
—
Al0.483P0.358Si0.141O2
Al0.491P0.360Si0.149O2
Al0.493P0.366Si0.159O2
—
a
The initial gel molar composition: Al2O3/P2O5/SiO2/DEA/H2O ¼ 1.0/1.0/x/y/150, crystallization conditions: 200 ꢁC, 72 h. b The crystallization time is
24 h, and the estimated value of framework compounds is 85% used for the yield calculation.
5610 | J. Mater. Chem. A, 2015, 3, 5608–5616
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry A
Fig. 4 Nitrogen adsorption–desorption isotherms (a) and BJH pore
size distributions (b) of samples 6, 7 and 8.
Fig. 2 SEM images of calcined samples 6 (left), 7 (middle) and 8 (right).
to the CHA type zeolite structure in the [001] direction, which
demonstrates that the nano-square crystals have a consistent
growth orientation and arrangement. It is deduced that the
crystals of sample 7 grow preferentially along the plane corre-
sponding to the {001} crystal plane group.
The textural properties of samples 6–8 were characterized by
N2 adsorption–desorption experiments. As illustrated in Fig. 4a,
all the three adsorption–desorption isotherms have steep
uptakes near P/P0 ¼ 0 due to the microporous structure.
Remarkable differences among the three isotherms lie in the
region of P/P0 > 0.4. A very small hysteresis loop exists in the
region 0.45 < P/P0 < 0.7 for sample 6, while there are manifest
hysteresis loops at P/P0 ¼ 0.45–0.9 for sample 7, and at P/P0 ¼
0.7–0.9 for sample 8. The increased adsorption quantity at the
medium-pressure region is interpreted as capillary condensa-
tion in the mesopores. The abrupt inclination at a relatively
high pressure of P/P0 > 0.9 is associated with multilayer
adsorption between the intercrystal pores of nanoparticles. The
mesopore size distribution for samples 6 and 7 is 3–15 nm and
3–30 nm, respectively (Fig. 4b). The latter consists of richer
mesopores which is reected in the higher distribution peak.
Sample 8 has a wider pore distribution with the maximum at 25
nm. Based on the above analyses, it is believed that sample 7
has the most abundant mesopores with a relatively narrow
distribution among the three samples. Their textural properties,
summarized in Table 2, further conrm our conclusions. All the
Fig. 3 TEM image and SAED pattern (inset) of the calcined sample 7.
increases from 1/9 to 1/3. The thickness of each nanosheet is
around 50 nm. When the TPOAC/TEOS ratio reaches 1/1
(sample 8), the nanosheets disappear. Instead, SAPO-34 with an
elliptical morphology appears which consists of aggregates of
ca. 50 nm particles. TEM was used to investigate the details of
the structural features of nanosheets belonging to sample 7.
Fig. 3 reveals that the nanosheets are spliced together by
countless nano-square pieces. Mesopores can be observed
within the nanosheets. Selected area electron diffraction (SAED)
further conrms the crystallinity of the nanosheets (Fig. 3,
inset). The distinct diffraction pattern can be indexed according
Table 2 Textural properties and the acid amount of the samples
Textural properties
Acid amountf (mmol gꢂ1
)
Medium and
Weak strong
Sample
SBETa (m2 gꢂ1
)
Smicb (m2 gꢂ1
)
SExtc (m2 gꢂ1
)
Vmicd (cm3 gꢂ1
)
Vmeso (cm3 gꢂ1
)
Total
e
6
7
8
631
591
565
584
484
464
586
47
107
101
2
0.27
0.23
0.21
0.28
0.05
0.18
0.17
0
1.48
1.37
0.50
1.68
1.42
1.38
0.54
1.93
2.90
2.75
1.04
3.61
Conventional SAPO-34 588
a
BET surface area. b t-Plot micropore surface area. c t-Plot external surface area. d t-Plot micropore volume. e BJH adsorption volume. f Calculated
from the peak area of NH3-TPD proles.
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A, 2015, 3, 5608–5616 | 5611

View Article Online
Journal of Materials Chemistry A
Paper
three samples have a high BET surface area above 560 m2 gꢂ1 stronger Brønsted and Lewis acid sites on the external surface
and a large micropore volume of 0.20–0.27 cm3 gꢂ1, conrming due to their larger external surface.
the high crystallinity of these samples. In addition, sample 7
has the largest external surface area of 107 m2 gꢂ1 and meso-
pore volume of 0.18 cm3 gꢂ1
.
3.2 Effect of TPOAC
The acidic properties of samples 6, 7 and 8 were investigated
by NH3-TPD measurements by comparing with those of
conventional SAPO-34. As seen in Fig. 5, there are two desorp-
tion peaks at about 180–200 ꢁC and 400–470 ꢁC for all samples,
which correspond to weak and strong acid sites, respectively.
The amounts of acid sites were roughly determined through the
NH3-TPD proles, and are listed in Table 2. The large amount of
acid sites is consistent with the high Si contents of the products
synthesized with the DEA template. It is clear that samples 6, 7
and 8 prepared with TPOAC show lower desorption temperature
and smaller acid amount than the conventional SAPO-34.
Comparing samples 6, 7 and 8, the acidity decreases with
increase in the TPOAC/TEOS ratio in the initial gel. Obviously,
sample 8 possesses much weaker acid strength and much
smaller acid amount than the other two samples. These results
suggest that the introduction of TPOAC can weaken the acidity
of SAPO-34. The strength and type of acid sites were further
characterized by IR spectroscopic studies of ammonia and
pyridine sorption at varied temperatures. The NH3-adsorbed IR
spectra of samples 6–8 and conventional SAPO-34 at various
temperatures are shown in Fig. S6.† For NH3-adsorbed IR
spectra, the bands at 1456 and 1619 cmꢂ1 can be assigned to the
dasym of NH3 bonded to Brønsted and Lewis acid sites, respec-
tively.34,35 The strength of acid sites can be distinguished from
the differences of IR spectra obtained at different temperatures.
As shown in Fig. S6,† the IR signals attributed to the adsorbed
NH3 disappeared at 300 ꢁC and 400 ꢁC for samples 8 and 7
respectively, while minor signals were still detected for sample 6
and conventional SAPO-34 at 400 ꢁC. It is believed that the
strength of acid sites decreases in the order of conventional
SAPO-34/sample 6 > sample 7 > sample 8, which is consistent
with the result obtained from the NH3-TPD measurements. The
acidity of the external surface was also detected through the
pyridine-adsorbed IR spectra (Fig. S7†) because the bigger
pyridine (kinetic diameter: 0.57 nm) cannot enter the channels
of SAPO-34 (0.38 nm). This shows that samples 7 and 8 have
The above characterizations clearly demonstrate that the
introduction of TPOAC has a signicant impact on the crystal-
lization of SAPO-34. In order to investigate the role of TPOAC,
sample 7 was selected to measure the 13C and 29Si NMR. The 13
C
MAS NMR spectrum is illustrated in Fig. 6a. Two peaks at 42 (A)
and 10.7 ppm (B) corresponding to DEA can be easily identied
on the basis of our previous report.36 The strongest signal at 30.8
ppm (g) is attributed to methylene moieties belonging to the
long-alkyl chain of TPOAC. All the other peaks can also be
ascribed to the corresponding groups of TPOAC which proves
its presence.37,38 The signal at 59.3 ppm ascribed to Si–OCH3
is negligible, which indicates that Si–OCH3 is hydrolyzed
mostly during the reaction. The 29Si CP NMR spectrum of the
as-synthesized sample 7 and MAS NMR spectrum for the
calcined sample 7 are shown in Fig. 6b. Both 29Si spectra
give a clear resonance signal at ꢂ89 ppm due to Si(OAl)4
species.39 A broad signal at ꢂ55 to ꢂ75 ppm is detected for the
as-synthesized sample, which is attributed to the Tn groups
(Tn ¼ RSi(OAl)n(OH)3ꢂn, n ¼ 2 and 3).40,41 It indicates that the
organo-siloxanes are connected to the silicoaluminophosphate
framework. Aer calcination, the resonance signal at ꢂ55 to
ꢂ75 ppm disappears, and two new shoulder peaks at about ꢂ80
and ꢂ85 ppm arise instead. It may be caused by the breaking of
Fig. 6 13C MAS NMR spectrum of as-synthesized sample 7 (a), 29Si CP
NMR spectrum of as-synthesized sample 7 and MAS NMR spectrum of
Fig. 5 NH3-TPD profiles of samples 6, 7, 8 and conventional SAPO-34. calcined sample 7 (b).
5612 | J. Mater. Chem. A, 2015, 3, 5608–5616
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry A
Si–C bonds due to the removal of organic groups, where zeolite morphology. Fan et al. synthesized different sized SAPO-
Si(OAl)3OH and Si(OAl)2(OH)2 species are formed as the defect 11 with cationic surfactants.43 They proved the adsorption
sites aerwards.41 The appearance of the Si–OH group is also behavior of the surfactant on the SAPO precursor through zeta
1
conrmed by H MAS NMR spectroscopy (Fig. S8†).
potential measurements at different crystallization times of
TG and DTA analyses were performed to examine the SAPO-11. In our work, the product morphology changes obvi-
organics of the samples. As seen in Fig. 7, conventional SAPO-34 ously with the increased amount of TPOAC (Fig. 2). Besides, the
has only one obvious exothermic peak at 435 ꢁC in the DTA time-consuming experiments present an intergrowing
curve corresponding to the DEA loss of ca. 14.6 wt% from 380 ꢁC morphology in a 12 hour reaction (Fig. S3a and S4a†). It
to 650 ꢁC in the TG curve. The SAPO-34s prepared with TPOAC suggests that the interplay between the surfactants and the
ꢁ
exhibit broader exothermic signals before 435 C in their DTA SAPO precursors happens from the beginning of the reaction.
curves, and the weight losses occur smoothly aer 200 C. The Referring to the investigation results by Zhang et al.,44 the
ꢁ
total weight losses are 17.5, 21.6 and 25.7 wt% for samples 6, 7 surfactant may act as a crystal face inhibitor preferentially
and 8, respectively. The additional exothermic peaks and weight absorbed onto certain specic planes. The anisotropic growth
losses are ascribed to the combustion of TPOAC. Although the of the crystal is therefore triggered. Moreover, the surfactant
precise amount of TPOAC cannot be determined, it is certain may be a very potent surface stabilizer, which could lower the
that the introduced amount of TPOAC increases with increase energy cost for creating new surfaces, thus affecting crystal
in the TPOAC/TEOS ratio.
splitting and prompting the formation of hierarchical archi-
Surfactants have been widely used to control the morphology tectures from individual crystals. A possible crystallization
of well-structured materials due to their efficient self-assembly process is therefore proposed and illustrated in Scheme 1. SAPO
properties.29,42–44 Schwieger et al. synthesized house-of-cards- precursors with adsorbed cationic surfactants are rstly
like mesoporous zeolite X by using the organosilane surfactant formed. If a small amount of TPOAC is involved, the crystal
3-(trimethoxysilyl)propyl hexadecyl dimethylammonium chlo- growth of SAPO-34 is not obstructed so much by the surfactants,
ride (TPHAC).29 They believed that the interplay between the and the rhombohedral crystals still form, but with a certain
surface activity of the TPHA+ molecules and the charge- degree of twinning. When the concentration of TPOAC is in a
balancing effects of the inorganic cations leads to the formation suitable range, the self-assembly of TPOAC directs the forma-
of micellar assemblies and directs the formation of the special tion of the intergrowth of SAPO-34 nanosheets. However, if the
Fig. 7 TG (a) and DTA (b) curves of samples 6, 7, 8 and conventional SAPO-34.
Scheme 1 The proposed crystallization process of SAPO-34s with different TPOAC/TEOS ratios.
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A, 2015, 3, 5608–5616 | 5613

View Article Online
Journal of Materials Chemistry A
Paper
amount of TPOAC is very high, the inhibition of crystal growth the diffusion paths, which greatly enhance the mass transfer
is so serious that only nanoparticles are formed. Mesoporosity during methanol conversion. The satisfactory selectivity to light
is generated when TPOAC is removed by calcination. Notably, olens could be attributed to the retention of its excellent
the loss of mesoporosity for sample 8 might be more because microporosity. Notably, sample 8 has a declining lifetime (226
the nanoparticles accumulate randomly and gather more vs. 308 min) and selectivity to light olens (78.94 vs. 81.16%) in
closely aer the removal of TPOAC. So the external surface and comparison with sample 7 although they have comparable
mesopore volume are comparable for samples 7 and 8 even if mesoporosity. A possible reason is that sample 8 possesses
the content of TPOAC is higher for sample 8.
much weaker acid strength and lower acid amount as shown in
NH3-TPD proles (Fig. 5). Normally, SAPO-34 with lower acid
strength and acidic amount exhibits slower catalyst deactiva-
tion and better MTO catalytic performance. However, a catalyst
with the fewest acid sites does not necessarily have the longest
lifetime.12 The available acid sites may be gradually covered by
coke, resulting in faster deactivation. Therefore, in order to
achieve optimal catalytic performance for the SAPO-34 catalyst,
it is sensible to balance both mesoporosity and microporosity,
and control the acidity precisely. Catalyst regeneration experi-
ments were performed for sample 7 by calcination at 600 ꢁC for
2 h. And then, the MTO reaction was repeated under the same
conditions. The recycled catalyst showed a slightly longer life-
time (342 vs. 308 min) and lower selectivity to light olens (80.89
vs. 81.16%). The result conrms the excellent stability of the
hierarchical SAPO-34. The changes in lifetime and selectivity
may be the result of the change in acidity during calcination.47
3.3 Role of the microporous template
Compared to the reported hierarchical SAPO-34 directed by
TEAOH31 and MOR,32 SAPO-34 synthesized herein using DEA
has distinct morphology and acidic property. The result
demonstrates again that the microporous template plays an
important role in the synthesis of mesoporous SAPO-34.
Generally, TEAOH tends to direct SAPO-34 crystallites, while
DEA and MOR lead to micrometer-sized SAPO-34 crystals. The
latter two templates tend to introduce more silica atoms into
the framework which results in more acidic sites.45 Considering
that the introduction of TPOAC can weaken the acidity of SAPO
molecular sieves, it is supposed that the template leading to
silica-rich SAPO-34 is more suitable for the synthesis of meso-
porous SAPO-34 together with TPOAC. Otherwise, the desired
acidity of conventional SAPO-34 bestowed by a microporous
template, such as TEAOH, might be further weakened.
3.4 Catalytic test
The catalytic properties of the samples were evaluated using the
MTO reaction in a xed-bed reactor, and the results are given in
Fig. 8 and Table 3. Sample 6 exhibits the shortest catalytic
lifetime of 72 min and 77.97% selectivity for ethylene and
propylene, which are even worse than those of conventional
SAPO-34 with a similar silica content and the same structure
directing agent DEA. The reason could be the bigger bulk crystal
size of ca. 30 mm for sample 6 which leads to a longer diffusion
path and accelerates coke deposition.46 Thankfully, the catalytic
performance of sample 7 has been remarkably improved. The
methanol conversion over 99% could be retained for 308 min,
which is much longer than that of conventional SAPO-34 (157
min). The selectivity for ethylene and propylene is comparable
to that of conventional SAPO-34 (81.16 vs. 81.99%). The gener-
Fig.
8 Methanol conversion variation with time-on-stream over
samples 6, 7, 8, conventional SAPO-34 and regenerated sample 7.
Experimental conditions: WHSV ¼ 3 hꢂ1, T ¼ 450 ꢁC, catalyst weight ¼
ated mesopores and the small crystal size of sample 7 shorten 300 mg.
Table 3 Lifetimea and product distributionb of SAPO-34s in the MTO reaction (WHSV ¼ 3 hꢂ1, T ¼ 450 ꢁC)
C3
C2 + C3
C3H8
(%)
C4–C6
(%)
Coke conꢂte1ntc
Rcoke
¼
¼
¼
c
(mg minꢂ1
)
¼
Sample
Lifetime (min)
C2 (%)
(%)
(%)
(%, g gcat
)
6
72
308
342
226
157
42.32
49.71
48.41
48.19
47.56
35.65
31.44
32.48
30.74
34.43
77.97
81.16
80.89
78.94
81.99
4.66
0.73
2.72
0.88
3.58
12.73
13.99
12.99
16.43
12.31
—
19.65
—
—
14.91
—
0.214
—
—
0.335
7
7d
8
Conventional SAPO-34
a
b
Catalyst lifetime is dened as the reaction duration with >99% methanol conversion. Based on the highest selectivity of ethene and propene
c
under >99% methanol conversion. Determined using a TG and DTA analyzer up to 800 ꢁC measured aer the MTO reaction; Rcoke ¼ coke
amount/reaction time. d Aer deactivation, sample 7 was recycled aer regeneration.
5614 | J. Mater. Chem. A, 2015, 3, 5608–5616
This journal is © The Royal Society of Chemistry 2015

View Article Online
Paper
Journal of Materials Chemistry A
4 Conclusion
By using the quaternary ammonium-type organosilane surfac-
tant TPOAC and microporous template DEA, highly crystallized
hierarchical SAPO-34 molecular sieves have been hydrother-
mally synthesized in one step with improved catalytic stability
for the MTO process. 13C and 29Si NMR clearly prove the exis-
tence of TPOAC in the products which indicates that it is
involved in the formation of SAPO-34 as a mesoporogen and a
part of silica source. The morphology and acidity of SAPO-34 are
therefore optimized which contribute to enhance its catalytic
ability. This work demonstrates that the so-template method
is feasible for the synthesis of hierarchical SAPO-34. In the
meantime, the microporous template, responsible for retaining
the inherent microporosity and acidity of silicoaluminophos-
phate molecular sieves, should be paid sufficient attention.
More highly effective hierarchical SAPO catalysts are expected to
be prepared with the understanding of the crystallization
process and the roles of mesoporous and microporous
templates.
Fig.
9 GC-MS chromatograms of occluded organic species
retained in sample 7 and conventional SAPO-34 after the MTO
reaction. The structures annotated on the chromatograms are peak
identifications in comparison with the mass spectra of those in the
NIST database.
Acknowledgements
In order to understand well the deactivation process of the
hierarchical SAPO-34 catalyst, the amount of deposited coke
and species of deactivated sample 7 and conventional SAPO-34
catalysts were analyzed aer the MTO reaction. The weight
losses of sample 7 and conventional SAPO-34 are 19.65% and
14.91%, respectively (Table 3), which are due to the combus-
tion of the retained coke species. Considering that coke was
deposited for different reaction times for the two catalysts (481
min for sample 7 and 208 min for conventional SAPO-34), the
average coke formation rate was calculated using the
following equation: Rcoke ¼ coke amount/reaction time. Coke
formation rate of sample 7 (0.214) is much slower than that of
conventional SAPO-34 (0.335) which agrees with the lower
hydrogen transfer index (C3H8/C3H6, Fig. S9†) and longer
lifetime of sample 7. The GC-MS chromatograms corre-
sponding to various deposited coke species of the two cata-
lysts are shown in Fig. 9. It is obvious that naphthalene,
phenanthrene and pyrene are the main coke species of
conventional SAPO-34. However, the peak intensity of naph-
thalene and naphthalene derivatives are found to be weaker
for sample 7. Normally, it is believed that naphthalene and its
derivatives can get converted to heavier constituents such as
phenanthrene and pyrene when they have access to light
molecules together with catalytic centers.48,49 Thus, the
smaller amounts of residues of naphthalene and its deriva-
tives of sample 7 should be ascribed to the introduced mes-
opores which provide larger space for diffusion and make the
further reaction for naphthalene and its derivatives happen.
The peaks with a retention time of above 45 min should be
attributed to polycyclic aromatics with more than four fused
rings. These retained compounds are too big to be accom-
modated in the CHA cage of SAPO-34 and hence form on the
external surface of the catalysts.16 The richer external surface
of sample 7 leads to relatively more pyrene and bigger poly-
cyclic aromatics.
We are thankful for the nancial support from the National
Natural Science Foundation of China (21476228 and 21101150).
Notes and references
1 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.
2 B. M. Lok, C. A. Messina, R. L. Patton, R. T. Gajek,
T. R. Cannan and E. M. Flanigen, US Pat. 4440871, 1984.
3 R. L. Bedard, in Zeolites in Industrial Separation and Catalysis,
ed. S. Kulprathipanja, Wiley-VCH, Weinheim, Germany,
2010, pp. 61–84.
4 S. T. Wilson, Stud. Surf. Sci. Catal., 2007, 168, 105–135.
5 H. O. Pastore, S. Coluccia and L. Marchese, Annu. Rev. Mater.
Res., 2005, 35, 351–395.
6 Z. M. Liu and J. Liang, Curr. Opin. Solid State Mater. Sci.,
1999, 4, 80–84.
7 J. M. Campelo, F. Lafont and J. M. Marinas, J. Catal., 1995,
156, 11–18.
8 S. G. Li, J. L. Falconer and R. D. Noble, Adv. Mater., 2006, 18,
2601–2603.
9 X. Su, P. Tian, D. Fan, Q. H. Xia, Y. Yang, S. T. Xu, L. Zhang,
Y. Zhang, D. H. Wang and Z. M. Liu, ChemSusChem, 2013, 6,
911–918.
10 J. Liang, H. Y. Li, S. Zhao, W. G. Guo, R. H. Wang and
M. L. Ying, Appl. Catal., 1990, 64, 31–40.
11 A. J. Marchi and G. F. Froment, Appl. Catal., 1991, 71, 139–
152.
12 S. Wilson and P. Barger, Microporous Mesoporous Mater.,
1999, 29, 117–126.
13 L. Xu, A. P. Du, Y. X. Wei, S. H. Meng, Y. L. He, Y. L. Wang,
Z. X. Yu, X. Z. Zhang and Z. M. Liu, Chin. J. Catal., 2008, 29,
727–732.
This journal is © The Royal Society of Chemistry 2015
J. Mater. Chem. A, 2015, 3, 5608–5616 | 5615

View Article Online
Journal of Materials Chemistry A
Paper
14 J. Zhu, Y. Cui, Y. Wang and F. Wei, Chem. Commun., 2009, 30 M. Choi, R. Srivastava and R. Ryoo, Chem. Commun., 2006,
3282–3284. 4380–4382.
15 D. Fan, P. Tian, X. Su, Y. Y. Yuan, D. H. Wang, C. Wang, 31 L. Chen, R. W. Wang, S. Ding, B. B. Liu, H. Xia, Z. T. Zhang
M. Yang, L. Y. Wang, S. T. Xu and Z. M. Liu, J. Mater.
Chem. A, 2013, 1, 14206–14213.
16 G. J. Yang, Y. X. Wei, S. T. Xu, J. R. Chen, J. Z. Li, Z. M. Li,
and S. L. Qiu, Chem. J. Chin. Univ., 2010, 9, 1693–1696.
32 Q. M. Sun, N. Wang, D. Y. Xi, M. Yang and J. H. Yu, Chem.
Commun., 2014, 50, 6502–6505.
J. H. Yu and R. R. Xu, J. Phys. Chem. C, 2013, 117, 8214–8222. 33 M. Guisnet, J. Mol. Catal. A: Chem., 2002, 182, 367–382.
17 M. Yang, P. Tian, C. Wang, Y. Y. Yuan, Y. Yang, S. T. Xu, 34 Z. M. Wang, Z. J. Tian, F. Teng, G. D. Wen, Y. P. Xu, Z. S. Xu
Y. L. He and Z. M. Liu, Chem. Commun., 2014, 50, 1845–1847.
and L. W. Lin, Catal. Lett., 2005, 103, 109–116.
18 H. Q. Yang, Z. C. Liu, H. X. Gao and Z. K. Xie, J. Mater. Chem., 35 J. Datka, B. Gil and A. Kubacka, Zeolites, 1995, 15, 501–506.
2010, 20, 3227–3231.
19 F. Schmidt, S. Paasch, E. Brunner and S. Kaskel, Microporous
Mesoporous Mater., 2012, 164, 214–221.
36 D. Fan, P. Tian, S. T. Xu, Q. H. Xia, X. Su, L. Zhang, Y. Zhang,
Y. L. He and Z. M. Liu, J. Mater. Chem., 2012, 22, 6568–6574.
37 K. Fujii, S. Hayashi and H. Kodama, Chem. Mater., 2003, 15,
1189–1197.
20 D. Y. Xi, Q. M. Sun, J. Xu, M. Cho, H. S. Cho, S. Asahina, Y. Li,
F. Deng, O. Terasaki and J. H. Yu, J. Mater. Chem. A, 2014, 2, 38 Z. L. Peng and X. J. Wu, Fine Chemicals, 2006, 23, 873–877.
17994–18004.
39 W. L. Shen, X. Li, Y. X. Wei, P. Tian, F. Deng, X. W. Han and
X. H. Bao, Microporous Mesoporous Mater., 2012, 158, 19–25.
40 T. M. Suzuki, T. Nakamura, E. Sudo, Y. Akimoto and K. Yano,
J. Catal., 2008, 258, 265–272.
21 M. Choi, H. S. Cho, R. Srivastava, C. Venkatesan, D. H. Choi
and R. Ryoo, Nat. Mater., 2006, 5, 718–723.
22 F. S. Xiao, L. F. Wang, C. Y. Yin, K. F. Lin, Y. Di, J. X. Li,
R. R. Xu, D. S. Su, R. Schlogl, T. Yokoi and T. Tatsumi, 41 Z. T. Xue, J. H. Ma, W. M. Hao, X. Bai, Y. H. Kang, J. H. Liu
Angew. Chem., Int. Ed., 2006, 45, 3090–3093. and R. F. Li, J. Mater. Chem., 2012, 22, 2532–2538.
23 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki and R. Ryoo, 42 H. Xue, Z. H. Li, H. Dong, L. Wu, X. X. Wang and X. Z. Fu,
Nature, 2009, 461, 246–249. Cryst. Growth Des., 2008, 8, 4469–4475.
24 K. Na, C. Jo, J. Kim, K. Cho, J. Jung, Y. Seo, R. J. Messinger, 43 L. Guo, X. J. Bao, Y. Fan, G. Shi, H. Y. Liu and D. J. Bai, J.
B. F. Chmelka and R. Ryoo, Science, 2011, 333, 328–332. Catal., 2012, 294, 161–170.
25 J. Zhu, Y. H. Zhu, L. K. Zhu, M. Rigutto, A. van der Made, 44 K. Liu, H. P. You, G. Jia, Y. H. Zheng, Y. J. Huang, Y. H. Song,
C. G. Yang, S. X. Pan, L. Wang, L. F. Zhu, Y. Y. Jin, Q. Sun,
Q. M. Wu, X. J. Meng, D. L. Zhang, Y. Han, J. X. Li,
M. Yang, L. H. Zhang and H. J. Zhang, Cryst. Growth Des.,
2010, 10, 790–797.
Y. Y. Chu, A. M. Zheng, S. L. Qiu, X. M. Zheng and 45 G. Y. Liu, P. Tian and Z. M. Liu, Chin. J. Catal., 2012, 33, 174–
F. S. Xiao, J. Am. Chem. Soc., 2014, 136, 2503–2510. 182.
26 Y. Cui, Q. Zhang, J. He, Y. Wang and F. Wei, Particuology, 46 D. Chen, K. Moljord, T. Fuglerud and A. Holmen,
2013, 11, 468–474. Microporous Mesoporous Mater., 1999, 29, 191–203.
27 F. Wang, L. Sun, C. L. Chen, Z. Chen, Z. W. Zhang, G. H. Wei 47 Z. B. Li, J. Martinez-Triguero, P. Concepcion, J. H. Yu and
and X. M. Jiang, RSC Adv., 2014, 4, 46093–46096. A. Corma, Phys. Chem. Chem. Phys., 2013, 15, 14670–14680.
28 K. Cho, H. S. Cho, L. C. de Menorval and R. Ryoo, Chem. 48 H. Fu, W. G. Song and J. F. Haw, Catal. Lett., 2001, 76, 89–94.
Mater., 2009, 21, 5664–5673.
29 A. Inayat, I. Knoke, E. Spiecker and W. Schwieger, Angew.
Chem., Int. Ed., 2012, 51, 1962–1965.
49 J. F. Haw, W. G. Song, D. M. Marcus and J. B. Nicholas, Acc.
Chem. Res., 2003, 36, 317–326.
5616 | J. Mater. Chem. A, 2015, 3, 5608–5616
This journal is © The Royal Society of Chemistry 2015