A fast route for synthesizing nano-sized ZSM-5 aggregates

Article abstract of DOI:10.1039/c4ta04073b

Nano-sized ZSM-5 aggregates have been rapidly synthesized from leached metakaolin by solid-like state conversion. The influence of synthesis conditions such as TPA⁺/SiO₂, NaOH/SiO₂ and SiO₂/Al₂O₃ molar ratios on the final products was investigated. The properties of nano-sized ZSM-5 aggregates were characterized by XRD, SEM, HRTEM, ²⁹Si and ²⁷Al MAS NMR, NH₃-TPD, TG, N₂ adsorption-desorption and particle size analysis. The results clearly showed that nano-sized ZSM-5 aggregates could be obtained within 2 h via solid-like state conversion. SEM revealed that the obtained ZSM-5 aggregates were irregular spheres that consisted of nano-sized crystallites with 30-50 nm. The crystallization process indicated that the size of flaky raw materials gradually decreased and formed nano-sized particles as time prolonged. Therefore, the transformation mechanism followed the solution mediated mechanism, though only a little water was contained in the system. Compared with the conventional hydrothermal route, the solid-like state conversion not only significantly shortened the crystallization time, but also totally avoided the emission of waste liquids. In addition, for the methanol dehydration reaction, the nano-sized ZSM-5 aggregates obtained by this method showed much better catalytic performance, C2-C4 olefin selectivity and longer lifetime to endure coke deposition than the sample obtained by the hydrothermal route.

Full text of DOI:10.1039/c4ta04073b

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
A fast route for synthesizing nano-sized ZSM-5
aggregates
Cite this: DOI: 10.1039/c4ta04073b
a
Feng Pan, Xuchen Lu, ac Qingshan Zhu,a Zhimin Zhang,a Yan Yan,a Tizhuang Wanga
*
and Shiwei Chenab
Nano-sized ZSM-5 aggregates have been rapidly synthesized from leached metakaolin by solid-like state
conversion. The influence of synthesis conditions such as TPA+/SiO2, NaOH/SiO2 and SiO2/Al2O3 molar
ratios on the final products was investigated. The properties of nano-sized ZSM-5 aggregates were
characterized by XRD, SEM, HRTEM, 29Si and 27Al MAS NMR, NH3-TPD, TG, N2 adsorption–desorption
and particle size analysis. The results clearly showed that nano-sized ZSM-5 aggregates could be
obtained within 2 h via solid-like state conversion. SEM revealed that the obtained ZSM-5 aggregates
were irregular spheres that consisted of nano-sized crystallites with 30–50 nm. The crystallization
process indicated that the size of flaky raw materials gradually decreased and formed nano-sized
particles as time prolonged. Therefore, the transformation mechanism followed the solution mediated
mechanism, though only a little water was contained in the system. Compared with the conventional
hydrothermal route, the solid-like state conversion not only significantly shortened the crystallization
time, but also totally avoided the emission of waste liquids. In addition, for the methanol dehydration
reaction, the nano-sized ZSM-5 aggregates obtained by this method showed much better catalytic
performance, C2–C4 olefin selectivity and longer lifetime to endure coke deposition than the sample
obtained by the hydrothermal route.
Received 7th August 2014
Accepted 17th October 2014
DOI: 10.1039/c4ta04073b
www.rsc.org/MaterialsA
50%. Another strategy is to restrict the growth of the nano-
crystals, named as the conned space7,8 or silanization
method.9,10 Madsen and Jacobsen8 obtained nano-sized ZSM-5
1. Introduction
ZSM-5 is a medium pore zeolite having a 10-membered ring
structure with two types of channels, straight and zigzag. It has
been widely used as a catalyst and as a selective adsorbent in the
petrochemical industry owing to its high thermal stability,
intrinsic acidity, high surface area and well-dened porosity.1
Nano-sized zeolites have a dramatic effect on the rates of
diffusion and reaction, which can signicantly inuence the
catalytic performance when they are used as catalysts. There-
fore, nano-sized ZSM-5 zeolites have received much attention
due to their good performances such as very shape-selective
properties, high steam stability, high metal resistance and slow
coke accumulation.2,3 Various attempts to synthesize ZSM-5
with nano-sized crystals have been developed. In a typical
approach, monodispersed or nano-sized aggregates of ZSM-5
were synthesized by carefully controlling the gel composition
and crystallizing conditions in a hydrothermal environment.4–6
However, the crystallization time was as long as several days,
and the yield of nano-sized ZSM-5 aggregates was no more than
crystals by using the conned space method for the rst time.
The procedure consisted of incipient wetness impregnation of
mesoporous carbon black with a zeolite precursor, then, the
carbon black was consequently subjected to autoclave treat-
ment in saturated steam. Serrano et al.9 adopted organosilane
as the growth inhibitor to synthesize nano-sized ZSM-5. The
synthesis is based on reducing the size of zeolite crystals by
silanization of the zeolitic seeds or precursors to hinder their
further growth. The obtained sample consisted of 300–400 nm
particles which were formed by the aggregation of small units of
10 nm. The third measure is to provide enough crystal nuclei for
the formation of nanocrystals via adding crystal seeds, called
seed-induced crystallization. Majano et al.11 studied the inu-
ence of synthesis parameters on the crystallization of ZSM-5
nanoparticles, established the relationship between the factors
and the nanosized zeolite particles, and revealed the formation
mechanism of the nanocrystallites.
In fact, the above mentioned methods belong to the tradi-
tional hydrothermal route. Few shortcomings are still inevi-
table, such as time-consumption, low yield, generating a large
amount of waste solvents, high cost, and higher requirements
for equipment due to the high pressure. If the synthesis is
performed in a solvent-free system or by directly adopting the
aState Key Laboratory of Multiphase Complex Systems, Institute of Process
Engineering, Chinese Academy of Sciences, No.
1 Bei-er-tiao, Zhong-guan-cun,
Haidian District, Beijing 100190, P. R. China. E-mail: xclu@home.ipe.ac.cn; Tel:
+86-10-8254-4889
bUniversity of Chinese Academy of Sciences, Beijing 100039, P. R. China
cUnited Research Center for Resource and Materials, Wuhai 016000, P. R. China
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
solid-state raw material, some obvious advantages could be the solid-like conversion and traditional hydrothermal route
brought:12,13 (1) high yields of products, (2) reducing the waste has been compared.
liquids signicantly, (3) shortening the crystallization time
greatly and enhancing the production efficiency, and (4) saving
energy and simplifying the synthesis procedures. Therefore,
solid-state transformation12 and steam-assisted crystallization
2. Experimental
2.1. Synthesis of nano-sized ZSM-5 aggregates
(SAC), also called vapor-phase-transport or dry gel conversion
The raw material powders of kaolin (Inner Mongolia, China)
(DGC),13–16 have been developed for the synthesis of zeolites.
Especially, the DGC method is also quite effective for the
synthesis of nano-sized ZSM-5. It is worth mentioning that the
DGC or SAC method has gained much attention recently. A
series of studies have been devoted to synthesize ZSM-5 with
nano-sized crystals or hierarchical structure. Zhou et al.15
were converted to metakaolin by calcining kaolinite at 1073 K
for 2 h. Then, the metakaolin powders were leached in 6 M HCl
solution (solid/liquid ratio 1 : 5, g mLꢀ1) with stirring. Aer
that, the leached metakaolin was ltered, washed with deion-
ized water, dried and collected as a precursor for the synthesis
of nano-sized ZSM-5 aggregates.
obtained highly crystalline ZSM-5 with a micro/mesoporous
In a typical synthesis of nano-HZSM-5 zeolite, dealuminated
structure by treatment for 12 h. More recently, Song17 and
metakaolin was uniformly mixed with NaOH and TPAOH in a
molar ratio of SiO2 : xAl2O3 : yTPAOH : zNaOH : mH2O (the water
Nandan18 synthesized hierarchical ZSM-5 consisted of nano-
crystals via steam-assisted crystallization. The dry-gel conver-
only came from TPAOH solution)
¼
1 : 25.68–80 : 0.026–
sion method could shorten the crystallization time obviously,
enhance the yield of the products, and avoid the emission of
waste liquids. Nevertheless, the procedures of dry gel prepara-
tion are complicated, in which the solvents are necessary for the
preparation of homogeneous gels, followed by vaporizing the
solvents to obtain the dry gels.13 In addition, the organic
template could get into the solvent at the bottom of the Teon.
Even so, these abovementioned methods have provided good
ideas for the synthesis of nano-sized ZSM-5 or other zeolites.
Actually, ZSM-5 zeolites have been synthesized from
chemical silica and aluminium sources in a solvent-free
system. However, the crystals of the samples were micro-
sized particles and the crystallization time was as long as 24
0.065 : 0.012–0.258 : 0.89–2.67, and transferred into a 50 mL
Teon-lined autoclave. Then, the Teon-lined autoclave was le
to stand for 0–3 h at 190 ꢁC. The obtained sample was calcined at
550 ꢁC for 6 h to remove the organic templates. In order to
investigate the catalytic performance, the as-synthesized Na-ZSM-
5 sample needs to transform into the H-form ZSM-5, the as-
synthesized sample (SiO2/Al2O3 ¼ 37.74) was exchanged with 1 M
NH4NO3 solution at 90 ꢁC for 2 h, calcined at 550 ꢁC for 6 h, and
named as HSZ-5. A contrast sample was synthesized by the
traditional hydrothermal method27 and named as HCZ-5.
2.2. Characterization
h.14 Compared with the traditional chemical silicon and Powder XRD patterns of the dried solid products were
aluminum sources, natural clay is considered to be an recorded by using a PANalytical X'pert diffractometer with
alternative material for the synthesis of zeolites because it is CuKa radiation. The morphology and chemical composition
readily available and very economical. Therefore, the of the samples were characterized by eld-emission scanning
conversion of natural clay to crystalline zeolite has been electron microscopy (SEM) (JEM-7001F, JEOL, Japan).
gaining much attention in academic and industrial elds. Transmission electron microscopy (TEM) images were
Kaolin [Al2Si2O5(OH)4] is one pf the most important clays, obtained on a JEM 2100F electron microscope operating at an
which is connected by one siliconoxygen (SiO4) tetrahedral accelerating voltage of 200 kV. FT-IR spectra were recorded
layer and one alumina [Al(O, OH)6] layer. It has been proved on an ALPHA FT-IR spectrometer (Bruker, Germany) using
to be a good combinational silicon and aluminum source to the dried KBr disk technique, in the range of 400–4000 cmꢀ1
.
synthesize zeolites or functional materials, such as A,19 X,20 N2 adsorption–desorption analysis was carried out on an
mesoporous g-Al2O3,21 SAPOs,22,23 ZSM-5,24ꢀ26 etc. To obtain AUTOSORB-1 analyzer (Quantachrome). Prior to measure-
nano-sized zeolites, the key point is to control the concen- ment, the samples were degassed at 573 K for 10 h under
tration and supersaturation of the crystal nucleus. Kaolin vacuum. The total acidity and acid strength distribution of
gradually could dissolve into the system and form the crystal the catalyst were determined by calorimetric measurement of
ꢁ
nucleus heterogeneously during the crystallization differential adsorption of NH3 at 60 C and subsequent TPD
process.21,22 Therefore, it is difficult to synthesize ZSM-5 with of the adsorbed NH3 with a ramp of 10 ꢁC minꢀ1 up to 700 ꢁC
the nano-sized crystals via the traditional hydrothermal (Quantachrome CHEMBET 3000, USA). The coke amount was
method.
determined using a TG/ꢁDTA6300 (Japan) thermogravimetric
Based on the ideas of the DGC or SAC method, we tried to analyzer from 30 to 800 C under air ow with a heating rate
further reduce the crystallization time and the emission of the of 10 ꢁC minꢀ1. Particle size analysis was performed on a
waste solvents. Thus, this paper describes that the nano-sized laser particle size analyzer (Mastersizer 2000E, Malvern). 29Si
ZSM-5 aggregates were directly synthesized by solid-like state and 27Al magic angle spinning nuclear magnetic resonance
transformation using tetrapropylammonium hydroxide (MAS-NMR) spectroscopy was performed at room tempera-
(TPAOH) as a template and kaolin as a silica–alumina source. In ture using a Bruker Advance III 400 spectrometer operating at
addition, the catalytic performance of the obtained ZSM-5 by 104.0 MHz for 27Al and 79.3 MHz for 29Si. The recycle delay
was 1.0 s for 27Al and 2.0 s for 29Si. The pulse duration (or
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
pulse width) was 1.0 ms for 27Al and the contact time was 5.0
ms for 29Si.
2.3. Catalytic performance
ꢁ
Catalytic reaction tests were conducted at 425 C under atmo-
spheric pressure in a conventional continuous down ow, xed-
bed stainless steel reactor (inner diameter 6 mm). 0.3 g of the
HZSM-5 catalyst (40–60 mesh) was placed in the center zone of
the reactor. Prior to the catalytic measurements, the fresh
catalyst was activated at 500 ꢁC for 2 h with nitrogen (15 mL
minꢀ1). Then, the reactant (50 vol% methanol solution) was
pumped into the pre-heater reactor through a piston pump
(WHSV of ethanol was 7.0 hꢀ1) along with nitrogen (15 mL
minꢀ1).
Fig. 1 XRD patterns of as-synthesized samples with different TPA+/
The feed and products were analyzed on-line by gas chro-
matography (SHIMADZU GC-2014) equipped with a ame
ionization detector (FID) and a thermal conductivity detector
(TCD). The FID with a TG-BOND capillary column (50 m ꢂ 0.32
mm ꢂ 15 mm, Thermo, USA) analyzed the un-reacted methanol,
dimethyl ether and hydrocarbons (C1–C5). The TCD with a
GDX-101 packed column detected CO, H2 and CO2. All products
were identied according to the standard gases (Beijing Gaisi
Chemical Gases Center), and the concentration of the compo-
nents in the reactor effluent stream, expressed as molar
percentages, was determined from the on-line chromatographic
results. The response factors in gas chromatography analysis
were evaluated from the effective carbon number. In addition,
the conversion of methanol to coke was neglected because the
instantaneous formation of coke was dependent on the reaction
time and difficult to estimate. The conversion of methanol was
calculated by applying the molar balance between the inlet and
outlet of the reactor. The selectivity for the products of interest
was expressed as the mass percentage of each product and
calculated according to the carbon balance between the inlet
and the outlet of the reactor.
SiO2 molar ratios.
depended on the content of TPAOH solution. Fig. 1 shows the
XRD patterns of as-synthesized samples with different TPA+/
SiO2 molar ratios. According to the results, the characteristic
diffraction peaks of ZSM-5 appeared when the molar ratio of
TPA+/SiO2 was as low as 0.026, indicating that the leached
metakaolin transformed into highly crystalline ZSM-5 within 2
h. This amount of TPAOH was much lower than that reported by
Ren.12 However, nano-sized ZSM-5 aggregates could not be
obtained under this condition, as could be seen in Fig. 2(a). The
intensity of the diffraction peaks decreased, and the half-peak
width increased signicantly as the amount of TPAOH
increased, suggesting the existence of nano-sized crystals.9,28
The nal particle size decreased from 1 mm to 40 nm when the x
value of TPAOH increased from 0.026 to 0.052. The results
demonstrated that there was a remarkable dependence of the
nal particle size on the content of TPAOH. Further increasing
the TPAOH content (x > 0.052) did not lead to smaller crystals,
which was consistent with the results reported in the
3. Results and discussion
3.1. Factors affecting the synthesis of nano-sized ZSM-5
aggregates
The presence of TPAOH can promote the condensation reaction
between –Si–O–H groups to form –Si–O–Si– groups. In the
synthesis of nano-sized zeolites, the high supersaturation
conditions are a key factor for the formation of nano-crystalline
zeolites. This condition is usually achieved by the utilization of
proper amounts of organic structure directing agents (SDAs).
Therefore, the inuence of TPAOH content on the structure and
morphology of the nal products was investigated.
The organic template, TPAOH has four roles in the synthesis
system. It serves as a structure-directing agent in the assembly
of the zeolite framework, participates in the balance of the
framework charge, and maintains the alkalinity of the synthesis
solution. More importantly, it could provide the solvent for the
system. The system with the molar composition of
xTPAOH : SiO2 : yH2O has also been studied, where x was
Fig. 2 SEM images of as-synthesized samples: (a) TPA+/SiO2 ¼ 0.026;
between 0.026 and 0.065 and y was between 0.89 and 2.67 which (b) TPA+/SiO2 ¼ 0.039; (c) TPA+/SiO2 ¼ 0.052; (d) TPA+/SiO2 ¼ 0.065.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
literature.29 But, the amount of TPAOH used in this system was
much lower than that in the reference (TPA+/SiO2 $ 0.4).
It could be hypothesized that each TPA+ cation was an
effective site for the formation of a pentasil unit. Therefore, a
higher amount of TPAOH could provide more sites for the
formation of initial nuclei of ZSM-5 zeolite, which caused the
particle size of ZSM-5 to dramatically reduce. Thus, the content
of the structure directing agent played a very important role in
the formation of nano-sized ZSM-5. However, the effect of the
very little water that was present in the TPAOH solution could
not be ignored. Therefore, the very small amount of water
provided an environment for the dissolution of silicon and
aluminium species.
Nano-sized ZSM-5 aggregates could be obtained in such a
short time due to the little content of water. The nuclei formed
by silicon and aluminum species easily reached supersaturation
in the solid-like system, which were a key factor for the forma-
tion of nanocrystalline zeolites quickly. Compared with the
literature,30 where 55 nm silicalite-1 particles were synthesized
with a molar composition of 3TPAOH : 25SiO2 : 390H2O at low
temperature (<308 K) with a very long synthesis time (40
months), the crystallization time reduced signicantly in this
system.
The leached metakaolin still contained some metallic
elements (such as Fe, Ti, Mg, etc.), which might show different
crystallization characteristics from the chemical ones. There-
fore, it is necessary to investigate the effect of SiO2/Al2O3 molar
ratios on the nal products. Fig. 3 shows the XRD patterns of
the samples with different SiO2/Al2O3 molar ratios. It was
evident that all the samples showed strong diffraction peaks
except for the SiO2/Al2O3 sample which showed 25.68. The
results indicated that the precursor with lower SiO2/Al2O3 was
not benecial for the formation of nano-sized ZSM-5 aggre-
gates. As illustrated in Fig. 4(a), some aky material was still
visible. This situation might be ascribed to the high content of
aluminate ions which hindered the increase of the concentra-
tion of silicate ions that were involved in the formation of
crystal nuclei.31
Fig. 4 SEM images of the samples obtained from leached metakalin
with different SiO2/Al2O3 molar ratios: (a) SiO2/Al2O3 ¼ 25.68; (b) SiO2/
Al2O3 ¼ 37.74; (c) SiO2/Al2O3 ¼ 68.79; (d) SiO2/Al2O3 ¼ 80.
As the SiO2/Al2O3 increased to 37.74, the intense diffraction
peaks of ZSM-5 (JCPDS Card no. 00-044-0002) indicated that
highly crystalline nano-sized ZSM-5 could be obtained. From
the SEM images (Fig. 4(b)), the sample presented irregular
spherical agglomerates with a size of 200–600 nm. The aggre-
gates were composed of nano-sized crystals with ca. 40 nm.
Highly crystalline ZSM-5 was obtained as the silica alumina
ratio was increased, and all the three samples had a similar
morphology. The results indicated that there was no signicant
effect on the morphology as the SiO2/Al2O3 molar ratio changed.
Supersaturation of the nuclei can be achieved by the disso-
lution of excess silicon and aluminum species contained in the
precursors. So, the alkalinity of the system should satisfy the
requirement of nucleation and crystal growth. As a mineralizer,
the hydroxy group of NaOH facilitates the conversion of oxide
forms of silicon and aluminum to useful kinds of precursors in
Fig. 5 (a) XRD patterns of the samples obtained at different NaOH
Fig. 3 XRD patterns of the samples obtained from leached metakaolin contents; SEM images of the sample obtained as (b) NaOH/SiO2
¼
with different SiO2/Al2O3 molar ratios.
0.026 and (c) NaOH/SiO2 ¼ 0.258.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
tetrahedral coordination and condensation. The crystallization
system reaches the state of supersaturation, which is a neces-
sary condition to facilitate the process of nucleation and crystal
growth.
According to Fig. 5(a), the nal sample was an amorphous
phase without adding NaOH, indicating that the precursor did
not participate in the formation of crystal nuclei and the growth
of crystals. The obtained sample was still amorphous even aer
increasing TPAOH/SiO2 to 0.065, and prolonging the crystalli-
zation time to 24 h at 190 ꢁC. The results demonstrated that
NaOH not only provided OHꢀ to facilitate the polycondensation
reaction between silicon and aluminum species, but also
supplied Na+ to balance the charge of the framework. The
characteristic peaks of ZSM-5 could be detected at 2q ¼ 22–25ꢁ
as only little NaOH was added into the reaction system.
However, a large amount of precursor did not transform into
ZSM-5 because the alkalinity was too weak to make silica and
alumina dissolve into the system. The alkalinity increased with
the increase of NaOH content, which was benecial to improve
the nucleation rate and shorten the crystallization time.
Therefore, as the NaOH/SiO2 increased to 0.026, the relative
crystallinity dramatically increased in such a short time. As
could be seem in Fig. 5(b) and (c), all samples were aggregates
that were composed of nanocrystals with 40 nm, indicating
there was no signicant inuence on the morphology of the
nal products.
3.2. Crystallization process of nano-sized ZSM-5 aggregates
In order to understand the growth process of the nano-sized
ZSM-5 aggregates, the products from different crystallization
times were studied.
Fig. 6 (a) XRD patterns, (b) IR spectra, (c) 27Al NMR and (d) 29Si NMR of
the samples obtained at different crystallization times.
From the XRD patterns (Fig. 6(a)), the peaks at about 2q ¼ 7–
9ꢁ and 22–25ꢁ began to appear aer heating the sample for 0.5
h; the relative crystallinity was only 12%. According to the N2
adsorption–desorption results (Table 1), the specic surface
area was only 200.5 m2 gꢀ1. The relative crystallinity rapidly
increased to ca. 90% aer treating for 1 h. As the time was
extended to 3 h, no signicant change was observed, indicating
that almost pure ZSM-5 crystals were formed. The SBET of the
sample increased to 440.8 cm2 gꢀ1, which was similar to the
value reported in the literature.32 In the IR spectra, the peaks at
550 cmꢀ1 and 1221 cmꢀ1 which belong to the ve-membered
ring and the asymmetric stretching vibration of T–O bonds33
gradually got strong as the time extended.34
coordinated Al. The coordination of Si species changed
dramatically during the crystallization process. The peak at
ꢀ100 ppm disappeared, while the peak at ꢀ110 ppm was
gradually enhanced, the results indicated that the coordination
of Si species changed from Si (1Al) (Q3) to Si (0Al) (Q4).
Furthermore, the peaks assigned to Q3 and Q4 shied from
ꢀ100 ppm to ꢀ102 ppm and from ꢀ110 ppm to ꢀ115 ppm,
respectively. This phenomenon might be ascribed to the fact
that the bond angle of Si–O–Si in ZSM-5 was bigger than the one
Table 1 The porous properties of synthesized ZSM-5 under different
Fig. 6(c) and (d) show the change of coordination of Al and Si
species with the change of the crystallization time. For the
initial sample, two resonance peaks at 50 ppm and 0 ppm could
be observed in 27Al NMR spectra, which can be assigned to the
tetrahedrally coordinated aluminium in the framework and the
octahedral non-framework aluminium, respectively.35 A major
peak at ꢀ100 ppm with a shoulder centered at ꢀ110 ppm could
be seen in 29Si NMR spectra, corresponding to Si atoms in Si
(1Al) and Si (0Al) congurations.36,37 The resonance peak of 27Al
NMR at around 0 ppm gradually disappeared with the extension
of thermal treatment, which demonstrated that the octahedral
non-framework aluminium transformed into the tetrahedrally-
times
Property
0.5 h
1.0 h
2.0 h
3.0 h
Total surface areaa/(m2 gꢀ1
)
200.5
36.8
163.7
0.27
0.06
0
434.7
307.3
127.4
0.28
0.12
0.55
446.4
305.8
140.6
0.30
0.12
0.55
440.8
304.7
136.1
0.27
0.13
0.55
Micropore areab/(m2 gꢀ1
)
External surface areab/(m2 gꢀ1
Pore volumec/(cm3 gꢀ1
)
)
Micropore volumeb/(cm3 gꢀ1
)
Micropore pore diameterd/nm
a
Determined by the BET method from P/P0 ¼ 0.02–0.07. b Determined
c
by the t-plot method from P/P0 ¼ 0.4–0.6. Calculated at P/P0 ¼ 0.99.
d
Determined by the HK method.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
in amorphous silica.38 The remarkable decrease in the Q3/Q4
ratio indicated that the hydrophilic surface transformed into a
hydrophobic one, or in other words, the amorphous alumino-
silicate wall was converted into a crystallized zeolite framework.
Table 2 The porous properties of as-synthesized HKZ-5 and HCZ-5
Leached kaolin
Property
treated with TPAOH Leached kaolin
Total surface areaa/(m2 gꢀ1
)
250.4
77.6
172.8
0.25
481.8
233.4
248.4
0.38
Micropore areab/(m2 gꢀ1
)
3.3. Scheme for the formation of nano-sized ZSM-5
aggregates
External surface areab/(m2 gꢀ1
)
Pore volumec/(cm3 gꢀ1
)
Micropore volumeb/(cm3 gꢀ1
)
0.037
0.10
Typical clay minerals like kaolin have received special attention
because of their potential use in intercalation nanocomposites
in which their unique layered structure can play a key role.39 So,
it is necessary to discuss the interactions between leached
metakaolin and other materials.
a
Determined by the BET method from P/P0 ¼ 0.02–0.07. b Determined
by the t-plot method from P/P0 ¼ 0.4–0.6. c Calculated at P/P0 ¼ 0.99.
addition, some small holes emerged on the surface of the
precursor, as can be seen from the inset image in Fig. 8(A)-(b).
This phenomenon indicated that metakaolin dissolved and
formed silicate and aluminate ions that formed the elementary
units of ZSM-5.
Firstly, due to its special layered and porous structure, the
leached metakalin might combine with TPA+ via chemical
bonds (–OH) or an acting force (e.g. Van der Waals forces). In
order to clarify the interaction between the leached kaolin and
TPA+, the leached kaolin (K) and leached kaolin treated with
TPAOH at room temperature (TK) were detected on a TG and N2
adsorption–desorption apparatus (Fig. 7). For the TK sample,
there was an obvious mass peak loss in the second stage, which
can be attributed to the decomposition of TPAOH.13 The result
proved that the TPAOH was still kept in the leached kaolin aer
washing with water. Furthermore, the total surface area
decreased obviously from 481.8 m2 gꢀ1 to 250.4 m2 gꢀ1, and the
Thirdly, the silicate ions and aluminate ions were formed by
the dissolution of leached kaolin under alkaline conditions.
Generally, the nal crystal size distribution strongly depends on
the total number of nuclei formed during the crystallization and
on the rate of their formation.40,41 It is worth mentioning that
ZSM-5 cannot be obtained in the absence of NaOH by solid-like
state conversion. So, it could be deduced that the addition of
NaOH provided alkalinity to dissolve the solid materials and
Na+ to balance the charge of the skeleton. In addition, as
previously described, no ZSM-5 phase could be generated in the
nal products as the system denitely did not contain water.
Therefore, the little water contained in the TPAOH was critical
for the formation of zeolites in the solid-like state system.
According to Chang,42 the silicon species contacted with TPA+ by
van der Waals interaction, and formed embryonic structures
total pore volume decreased from 0.38 cm3 gꢀ1 to 0.25 cm3 gꢀ1
,
aer treating with TPAOH (Table 2). The results indicated that
TPA+ cations have been inserted into the leached kaolin.12 The
above results implied that the adsorption of TPAOH molecules
onto the outer surfaces of kaolin or between the layers of
leached kaolin was not a simple physical adsorption.
Secondly, the leached metakaolin dissolved gradually into
the system under alkaline conditions. To prove this phenom-
enon, the products obtained from different crystallization times
were characterized by SEM and TEM (Fig. 8). The size of aky
raw materials gradually decreased, and the edge of the materials
became smooth, as the crystallization time prolonged. In
Fig. 7 TG curves of leached kaolin and leached kaolin treated with
TPAOH. Note: the leached kaolin treated with TPAOH was obtained by
the following procedures: the leached kaolin was mixed with TPAOH
(25% solution) uniformly, and left to stand for 2 h at room temperature. Fig. 8 SEM (A) and TEM (B) images of the samples obtained at different
Then, the mixture was washed with water, filtered and dried.
times: (a) 0 h, (b) 0.5 h, and (c) 2 h.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
rapidly upon heating by formation of a chlathrate-like water
structure around TPA+, followed by a fast isomorphous substi-
tution of silicate with water. So, it was benecial for the nuclei
to get supersaturated in the solid-like state system.
Finally, the nuclei grew into nano-sized ZSM-5 aggregates.
Aer the nuclei formation, a fast crystal growth of ZSM-5
happened probably by the continuous assembly of the nuclei.
Thereaer, the solid surface was further dissolved, gelated, and
a continuous migration took place from the solid surface to the
colloid until it was depleted. The crystal nuclei gradually grew
into nano-sized ZSM-5, and aggregated together due to the high
activity of the nanocrystals, as can be seen in the TEM images
(Fig. 8(B)).
Therefore, the schematic of the crystallization mechanism
for the nano-sized ZSM-5 aggregates could be illustrated as
follows: (Fig. 9).
3.4. Catalytic performance
As illustrated in Fig. 10(a) and (b), the morphology of ZSM-5
synthesized by solid-like state conversion and the hydrothermal
route exhibited signicant difference. The d50 values of particles
of both samples were 427 nm and 6.4 mm, which was in good
agreement with the SEM results. The HSZ-5 catalysts showed
very high activity to convert methanol completely to hydrocar-
bons in the initial 20 hours (Fig. 10(d)). Aer that, the activity of
HSZ-5 declined slightly and the conversion of methanol was
sustainably higher than 95% aer 27 h. In sharp contrast,
conversion of methanol over HCZ-5 dropped to 78% aer 27 h.
Generally, the deactivation of zeolite catalysts occurs predomi-
nantly by the poisoning of acid sites and/or the pore blocking
due to the accumulation of carbonaceous deposits. The prod-
ucts easily diffused out of the pores in HSZ-5 because of the
existence of nano-sized crystals, which retarded the secondary
reaction to form the precursor of coke and hence delayed the
blocking of the channel by coke.43 Therefore, carbon deposition
on HSZ-5 was better than that on HCZ-5 zeolite; this conclusion
could be proved by the TG results (Fig. 10(f)).
As the reaction time prolonged, the selectivity of C2–C4
olens on the HCZ-5 catalyst gradually decreased to ca. 40%,
then increased to ca. 46%. In a sharp contrast, though the
conversion of methanol decreased to 95%, the HSZ-5 catalyst
was still favorable to the formation of C2–C4 olens. The
selectivity of olens over HSZ-5 continued to increase to ca.
65%. Due to the smaller crystal size of HSZ-5, the coke forma-
tion rate over HSZ-5 was slower than that over HCZ-5. Moreover,
Fig. 10 (a) SEM image of ZSM-5 obtained by solid-like state conver-
sion; (b) SEM image of ZSM-5 obtained by the hydrothermal route; (c)
NH3-TPD curves of HSZ-5 and HCZ-5; (d) methanol conversions and
C2–C4 olefin selectivity on the fresh zeolites as a function of time-on-
stream; (e) methanol conversions and C2–C4 olefin selectivity on the
regenerated zeolites as a function of time-on-stream; (f) TG curves of
HSZ-5 and HCZ-5 after the MTO reaction.
appropriate deposition of coke can reduce the size of the
channel, which was benecial to enhance the selectivity of C2–
C4 olens. In addition, the lower surface acidity of HSZ-5
contributed to the higher selectivity of C2–C4 olens, as
described by Inoue et al.,44 lower surface acidity played vital
roles in the formation of olens in the methanol dehydration
reaction, while aromatics were promoted at a higher acidity.
Though the MTO reaction was closely related to the strong acid
sites, the amount of strong acidity should be appropriate. As
shown in Table 3, the HSZ-5 catalyst had a lower concentration
Table 3 The porous properties of as-synthesized HSZ-5 and HCZ-5
Property
HSZ-5
HCZ-5
Total surface areaa/(m2 gꢀ1
)
446.4
305.8
140.6
0.30
0.13
0.33
354.3
311.9
42.4
0.23
0.13
0.62
0.45
Micropore areab/(m2 gꢀ1
)
External surface areab/(m2 gꢀ1
Pore volumec/(cm3 gꢀ1
Micropore volumeb/(cm3 gꢀ1
)
)
)
Weak acidityd/(NH3, mmol gꢀ1
)
Strong acidityd/(NH3, mmol gꢀ1
)
0.12
a
Determined by the BET method from P/P0 ¼ 0.02–0.07. b Determined
c
by the t-plot method from P/P0 ¼ 0.4–0.6. Calculated at P/P0 ¼ 0.99.
d
Weak acidity: from 373 to 643 K; strong acidity: from 643 to 823 K.
Fig. 9 Schematic of the growth mechanism for the nano-sized ZSM-5
both the weak and strong acidity were quantied by titration.
aggregates.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
of strong acid sites than that of HCZ-5. Therefore, it was bene-
cial for the formation of light olens.
To investigate the regeneration ability of the nano-sized
ZSM-5 aggregates, the reacted HSZ-5 was burned off under a
ow of dry air (15 mL minꢀ1) at 823 K for 2 h. Then, the
7 C. H. Jacobsen, Chem. Commun., 1999, 673–674.
8 I. Schmidt, C. Madsen and C. J. Jacobsen, Inorg. Chem., 2000,
39, 2279–2283.
´
9 D. P. Serrano, J. Aguado, J. M. Escola, J. M. Rodrıguez and
´
A. Peral, Chem. Mater., 2006, 18, 2462–2464.
regenerated HSZ-5 and HCZ-5 were used for the recycle reac- 10 Y.-P. Guo, H. J. Wang, Y. J. Guo, L. H. Guo, L. F. Chu and
tion. As shown in Fig. 10(e), aer 27 hours of continuous C. X. Guo, Chem. Eng. J., 2011, 166, 391–400.
running, methanol could still maintain a higher conversion rate 11 G. Majano, A. Darwiche, S. Mintova and V. Valtchev, Ind. Eng.
on regenerated HSZ-5. Compared with the fresh HSZ-5, the Chem. Res., 2009, 48, 7084–7091.
selectivity of C2–C4 olens was much higher in the initial stage. 12 L. Ren, Q. Wu, C. Yang, L. Zhu, C. Li, P. Zhang, H. Zhang,
It was interesting to note that the conversion of methanol on
HCZ-5 was much better than on the fresh one, and the C2–C4
X. Meng and F. S. Xiao, J. Am. Chem. Soc., 2012, 134,
15173–15176.
olens increased obviously in the initial stage. This phenom- 13 R. Cai, Y. Liu, S. Gu and Y. Yan, J. Am. Chem. Soc., 2010, 132,
enon might be ascribed to the dealumination of the catalyst that 12776–12777.
´
happened during the reaction due to the presence of water. In 14 P. H. PrasadaRao, Chem. Commun., 1996, 1441–1442.
addition, the selectivity of the olens on regenerated HSZ-5 had 15 J. Zhou, Z. Hua, J. Zhao, Z. Gao, S. Zeng and J. Shi, J. Mater.
a similar tendency with the fresh one, and showed much better
Chem., 2010, 20, 6764–6771.
catalytic performance than the regenerated HCZ-5. The results 16 K. Shen, N. Wang, W. Qian, Y. Cui and F. Wei, Catal. Sci.
indicated that the HSZ-5 had excellent regeneration ability.
Technol., 2014, 4, 3840–3844.
17 Y. Song, Z. Hua, Y. Zhu, J. Zhou, X. Zhou, Z. Liu and J. Shi, J.
Mater. Chem., 2012, 22, 3327–3329.
18 D. Nandan, S. K. Saxena and N. Viswanadham, J. Mater.
Chem. A, 2014, 2, 1054–1059.
4. Conclusions
In conclusion, a facile and fast approach has been developed for
¨
the synthesis of nanocrystal aggregates via solid-like state 19 M. Alkan, Ç. Hopa, Z. Yilmaz and H. Guler, Microporous
conversion. The obtained ZSM-5 aggregates were irregular Mesoporous Mater., 2005, 86, 176–184.
spheres with a crystal size of 200–500 nm, which consisted of 20 I. Caballero, F. G. Colina and J. Costa, Ind. Eng. Chem. Res.,
nano-sized crystallites of 30–50 nm. The obtained samples 2007, 46, 1029–1038.
showed good catalytic performance and regeneration ability in 21 F. Pan, X. Lu, T. Wang, Y. Wang, Z. Zhang, Y. Yan and
the MTO reaction. This method provided an easy and effective S. Yang, Mater. Lett., 2013, 91, 136–138.
path for the synthesis of nano-zeolites. In addition, natural 22 T. Wang, X. Lu and Y. Yan, Microporous Mesoporous Mater.,
layered clays like kaolin, sepiolite and montmorillonite can be 2013, 168, 155–163.
used as raw materials for the synthesis of many other zeolites, 23 J. Zhu, Y. Cui, Y. Wang and F. Wei, Chem. Commun., 2009,
such as ZSM-5, zeolite A, zeolite X, and zeolite Y. Therefore, this 3282–3284.
synthesis approach may be a general way to produce different 24 S. M. Holmes, S. H. Khoo and A. S. Kovo, Green Chem., 2011,
monodispersed or nano-sized aggregate zeolites.
13, 1152–1154.
25 K. Shen, W. Qian, N. Wang, J. Zhang and F. Wei, J. Mater.
Chem. A, 2013, 1, 3272–3275.
26 F. Pan, X. Lu, Y. Wang, S. Chen, T. Wang and Y. Yan,
Microporous Mesoporous Mater., 2014, 184, 134–140.
Acknowledgements
The authors would like to thank Wuhai Tian-yu Chemical High-
tech Co. Ltd (China) for the nancial and raw material 27 C. D. Madhusoodana, R. N. Das, Y. Kameshima and
assistance.
K. Okada, J. Porous Mater., 2005, 12, 273–280.
28 M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki and R. Ryoo,
Nature, 2009, 461, 246–249.
29 Q. Li, S. Yang and A. Navrotsky, Microporous Mesoporous
Mater., 2003, 65, 137–143.
Notes and references
1 C. S. Cundy and P. A. Cox, Chem. Rev., 2003, 103, 663–702.
2 Y. Adewuyi, D. Klocke and J. Buchanan, Appl. Catal., A, 1995, 30 R. W. Corkery and B. W. Ninham, Zeolites, 1997, 18, 379–386.
131, 121–133.
31 E. G. Derouane, S. Determmerie, Z. Gabelica and N. Blom,
Appl. Catal., 1981, 1, 201–224.
32 D. Serrano, J. Aguado, J. Rodriguez and A. Peral, J. Mater.
Chem., 2008, 18, 4210–4218.
33 G. Lei, B. Adelman, J. Sarkany and W. Sachtler, Appl. Catal.,
B, 1995, 5, 245–256.
34 J. Jansen, F. Van der Gaag and H. Van Bekkum, Zeolites,
1984, 4, 369–372.
3 L. Tosheva and V. P. Valtchev, Chem. Mater., 2005, 17, 2494–
2513.
4 W. Song, R. Justice, C. Jones, V. Grassian and S. Larsen,
Langmuir, 2004, 20, 8301–8306.
5 R. Van Grieken, J. Sotelo, J. Menendez and J. Melero,
Microporous Mesoporous Mater., 2000, 39, 135–147.
6 T. M. Davis, T. O. Drews, H. Ramanan, C. He, J. Dong,
H. Schnablegger, M. A. Katsoulakis, E. Kokkoli, 35 W. Guo, C. Xiong, L. Huang and Q. Li, J. Mater. Chem., 2001,
A. V. McCormick and R. L. Penn, Nat. Mater., 2006, 5, 400–
11, 1886–1890.
408.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2014

View Article Online
Paper
Journal of Materials Chemistry A
´
´
´
´
36 R. Ryoo and J. M. Kim, J. Chem. Soc., Chem. Commun., 1995, 40 S. Bosnar, T. Antonic-Jelic, J. Bronic, I. Krznaric and
´
B. Subotic, J. Cryst. Growth, 2004, 267, 270–282.
711–712.
37 A. Steel, S. W. Carr and M. W. Anderson, Chem. Mater., 1995, 41 C. S. Cundy and P. A. Cox, Microporous Mesoporous Mater.,
7, 1829–1832. 2005, 82, 1–78.
38 S. L. Burkett and M. E. Davis, Chem. Mater., 1995, 7, 920–928. 42 C. D. Chang and A. T. Bell, Catal. Lett., 1991, 8, 305–316.
39 P. Podsiadlo, A. K. Kaushik, E. M. Arruda, A. M. Waas, 43 X. J. Cheng, X. S. Wang and H. Y. Long, Microporous
B. S. Shim, J. Xu, H. Nandivada, B. G. Pumplin, J. Lahann
Mesoporous Mater., 2009, 119, 171–175.
and A. Ramamoorthy, Science, 2007, 318, 80–83.
44 K. Inoue, K. Okabe, M. Inaba, I. Takahara and K. Murata,
React. Kinet., Mech. Catal., 2010, 101, 477–489.
This journal is © The Royal Society of Chemistry 2014
J. Mater. Chem. A