Materials Chemistry and Physics xxx (2015) 1e11

Contents lists available at ScienceDirect

Materials Chemistry and Physics

journal homepage: www.elsevier.com/locate/matchemphys

Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic

supports: Inductive influence of various seeds and aluminosilicate gels
on the physicochemical properties and catalytic dehydrogenative
behavior
Marjan Razavian, Shohreh Fatemi*
School of Chemical Engineering, College of Engineering, University of Tehran, P.O. Box 11365-4563, Tehran, Iran

h i g h l i g h t s g r a p h i c a l a b s t r a c t

Synthesis of hierarchical ZSM-5 zeo-

lites with different Si/Al molar ratios.

TPAOH/TPABR-templated silicalite-1
seeding suspensions as structure
directing agents.

Presence of lamellar keatite assem-
blies at Si/Al  60 especially at
TPAOH-silicalite-1-directed ZSM-5.

Hybrid zeolites with supplementary
meso/macro pores as catalytic sup-
ports in PDH reaction.

Highest propene yield using PtSn/
ZSM-5(Si/Al ¼ 60) prepared from
TABR-silicalite-1 seeds.

a r t i c l e i n f o a b s t r a c t

Article history: TPABr-silicalite-1 and TPAOH-silicalite-1 seeding suspensions were employed as structure directing agents
Received 4 March 2015 through a seed-directed synthesis strategy to prepare a series of organotemplate-free hierarchical ZSM-5
Received in revised form zeolites with three different Si/Al molar ratios of 30, 60 and 120. No secondary meso-generating agent was
16 August 2015
implemented to construct wide supplementary pores. The physicochemical properties of the products
Accepted 22 August 2015
Available online xxx
were characterized by XRD, FESEM, FT-IR, EDX, AAS, TGA and N2 adsorptionedesorption techniques. All
zeolite samples with moderate hierarchy factors possessed a bi/tri-model porosity based on the governing
nucleation mechanism and growth of zeolite crystals upon the surface of the silicalite-1 seeds. The pre-
Keywords:
Nanostructures
pared hybrid zeolites with intrinsic micropores and auxiliary meso/macro pores were served as the cata-
Chemical synthesis lytic supports in propane dehydrogenation reaction to ascertain the influence of the seeding type, Si/Al
Energy dispersive analysis of X-rays (EDS or ratio and consequent structural and particulate properties on the catalytic performance. The best catalytic
EDAX) result was acquired for PteSn-based ZSM-5 catalyst, with Si/Al ¼ 60, synthesized with less expensive
Crystal structure TPABr-silicalite-1 seeds. Considerable activity (34.5% conversion) and high selectivity towards light olefins
(90% for both ethylene and propylene); propylene in particular (84.5%) was observed after 4 h time on
stream. This result is attributed to the high BET surface area and meso transport pores of the sample
collaborated with high acidity to progressively impact on the reactivity and selectivity.
© 2015 Published by Elsevier B.V.

* Corresponding author.
E-mail address: shfatemi@ut.ac.ir (S. Fatemi).

http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
0254-0584/© 2015 Published by Elsevier B.V.

Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
2 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
1. Introduction
Zeolites are important industrial materials with broad range of
applications as gas separator, ion-exchanger and more importantly
catalyzer [1e3]. Socony Mobil#5 (ZSM-5) with MFI framework type
has been regarded as the most popular member of zeolite family
with well-defined channels system of molecular dimensions [4]. Its
unique intricate porous structure and intrinsic surface acidity give
it size and shape sieving and catalytic properties which are boon for
lots of petrochemical reactions like cracking, aromatization, MTO
and so forth [5e7]. ZSM-5 can be generally synthesized via the
crystallization of an alkali precursor aluminosilicate hydrogel. The
precursor hydrogel can be prepared following different protocols
including organic template molecules as structure directing agents
(SDA) and space fillers or excluding the usage of organic templates
introducing extraneous seeds to induce the formation of ZSM-5 [8].
In fact in the latter method, seeds are provided to the alumi-
nosilicalte gel to direct the zeolite units and drive the crystallization
towards the formation of desired structure [9]. Cundy et al. [10]
proposed different behaviors for gel-wrapped seeds wherein
seeds can remain inert and solely act as pure cores on which zeolite
units can deposit and grow or partially dissolve and lead to the
secondary nucleation. Although seeds may act complicated and
play role in more than one way, their essential role in supplying
surface and exposing specific sites for the migration of TOH ter-
minal units to attach and pile up in order to promote the crystal-
lization process is evident. Short crystallization time, high
crystallization rate, elimination of self-nucleation time-consuming
step, low template consumption, high product yield and great
control over crystal size distribution are the advantages accompa-
nied by when using seed-directed environmentally benign syn-
thesis method [11,12].
The importance of similarity of the building units between the
seed and target zeolite, seed crystals surface area and recognition
capacity of seeds TOH units on the crystallization behavior of
template-free ZSM-5 have been studied by several groups [13,14].
Various seed-assisted synthesis systems were investigated by
introducing seeds with different frameworks (FAU, MEL, and MFI)
and pore structures. Framework similarity and existence of ter-
minal TOH units proved to be the key factors for the nucleation
process while large external surface area affected crystal size
distribution significantly. Effect of seeding step on the kinetic of
ZSM-5 crystallization process was examined sowing pre-fabricated
ZSM-5 and silicalite-1 seeds into the original self-induction system
[15]. Adoption of silicalite-1 nanoseeds had much more prominent
effect which led to the formation of relatively small aggregates
with promoted catalytic performance in benzene alkylation and
xylene isomerization. Despite great endeavors have been dedi-
cated to the investigation of the influence of synthesis parameters
and underlying crystallization behavior in seeded systems, reports
on the application of such materials are limited. A number of ZSM-
5 zeolites prepared through seed-induced synthesis methodology
using silicalite-1, ZSM-5 and NaY seeding suspensions were eval-
uated in the conversion of o-xylene [16], alkylation of naphthalene
[17], fluid catalytic cracking [18], and hydration of cyclohexane
[19] reactions and promising results were obtained. Well-
preserved activity and high selectivity towards light olefins
particularly propylene in FCC and o-xylene isomerization reactions
suggests that this is an efficient economic approach for the pro-
duction of ZSM-5 which will be a great benefit to industrial ap-
plications in selective production of light olefins in a controlled
trend.
In recent years, the growth in propylene demand exceeded the
growth in ethylene, opening up other technology opportunities
focused primarily on propylene [20,21]. It is predicted that the
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
demand for propylene derivatives will increase by 4% until 2020
and is expected to reach almost 100 million tons/yr in 2015
[22,23]. Generally, the worldwide propylene demand is mostly
supplied from steam crackers and fluid catalytic cracking (FCC)
units wherein propylene is produced as a byproduct [24]. How-
ever, the annual growing demand of propylene and the instability
of crude oil price, the FCC unit's feedstock, cannot meet the gap
between demand and productivity and consequent change of
market [21,25]. Propane dehydrogenation (PDH) involves cata-
lytic removal of hydrogen from propane to yield propylene which
has received significant attention as an on-purpose alternative
with high potential to cover the outgrowing propylene deficiency
[26,27]. However, it suffers from equilibrium limitation and
thermodynamical constraints due to strong endothermic char-
acter of the reaction disadvantageously. High reaction tempera-
ture is favorable, however it accelerates thermal cracking side
reaction towards lighter hydrocarbons and coke which leads to
low olefins selectivity and rapid deactivation. Therefore a great
deal of work has been conducted to formulate catalysts exhibiting
high catalytic properties and stability that can decelerate side
reactions under severe conditions. PteSn based ZSM-5 catalyst is
a hydrothermally stable catalyst with good reactivity and shape
selectivity because of its specific ten membered ring structure
which can hinder the formation of bulky molecules and coke
deposition [28,29].
It is well known that the application of a material is strictly
correlated with its properties such as chemical composition, crystal
size, morphology, network porosity, and acidity. In the light of this,
we were encouraged to prepare ZSM-5 zeolites employing facile
and low-cost seed-directed organotemplate-free synthesis strategy
to ascertain the influence of textural and physicochemical proper-
ties of obtained materials on their catalytic performance in PDH
reaction. Silicalite-1 high siliceous zeolite was selected as the
directing seeds based on the presence of common building units
between the parent seed and product zeolite and specific sites with
enhanced recognition capacity. These sites attract formed alumi-
nosilicate units in the gel to promote nucleation and subsequent
crystal growth of target zeolite. Thus, several hierarchical zeolites
with uniform crystal size distribution equipped with auxiliary
mesopores have been fabricated by varying the Si/Al molar ratio
and seeding suspension type namely, TPAOH and TPABr-templated
silicalite-1. It is well proved that the Si/Al ratio is an important
factor influencing physicochemical properties of zeolites, pre-
dominantly acidity. Several attempts have been made to clarify the
influence of Si/Al ratio of ZSM-5-derived catalysts in various re-
actions including alkane dehydrogenations [30,31], MTBE produc-
tion [32], craking of C4 alkanes [33], and alkylation [34]. Therefore,
the effect of Si/Al ratio on the catalytic features of prepared PteSn
based catalysts were assessed in PDH reaction aiming to produce
light olefins in a controlled selective trend; highlighting propylene
as the major product.
2. Experimental section
2.1. Materials
The chemical ingredients used in the syntheses were tetrapro-
pylammonium bromide (TPABr, 99%), tetrapropylammonium hy-
droxide (TPAOH, 40%), nitric acid (HNO3, wt. 65%), Platinum (IV)
chloride (PtCl4, 57.5% Pt), Tin (II) chloride (SnCl2), and sodium hy-
droxide (NaOH pellets, 98%) which were purchased from Merck
company. Colloidal silica (SiO2, wt. 30%), was obtained from Aldrich
and sodium aluminate (NaAlO2, 54% Al2O3 and 41% Na2O) was
bought from Riedel-de-Hae €n company.
 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
2.2. Synthesis of zeolites
The synthesis of seed-directed ZSM-5 zeolites was initially
started by the fabrication of silicalite-1 seeding suspension. Firstly,
aqueous silica sol was dissolved in the alkali solution containing
organic template and sodium hydroxide under vigorous stirring.
After complete hydrolysis, the obtained mixture with molar
composition of SiO2: 0.2 T: 0.03 Na2O: 40H2O (T ¼ TPABr or TPAOH)
was hydrothermally treated at 150  C for 24 h [35]. Secondly,
aluminosilicate hydrogels with different Si/Al ratios were prepared
by homogenous mixing of sodium hydroxide, sodium aluminate
and silica sol as aluminum and silicon sources and altering the
aluminum content of the synthesis precursor. The molar ratio of the
reaction mixtures were as follows 60 SiO2: x Al2O3: 3 Na2O: 1200
H2O where x ¼ 1, 0.5, and 0.25. Afterward, a certain amount of
silicalite-1 suspension with respect to 5% of total silica content was
added to the mixtures directly without any extra purification step.
No additional quantity of organic template was added to the
hydrogel, so the entire amount of template arised from seeding
suspension. Details on the composition and seed type of samples
are listed in (Table 1). Finally, crystallization of seeded systems was
performed at 170  C for 24 h to ensure high crystallinity. The ob-
tained products were recovered by centrifugation and repeatedly
re-suspended in deionized water until the sol exhibited a pH value
of about 9. The samples were then allowed to dry overnight at
100  C. Removal of the organic species was carried out at 500  C C
for 6 h in air with heating ramping rate of 1  C/min.
The solid yield of the obtained materials was defined as the
weight ratio percentage (g/g  100) of the solid obtained after
drying to the sum of the dry silica, sodium aluminate, and the dry
seeds in the starting Naaluminosilicate gel.
2.3. Preparation of the catalysts
The bimetallic PtSn/ZSM-5 catalysts were prepared by sequen-
tial wet impregnation method at 65  C with very dilute acidic so-
lution of SnCl2 and aqueous solution of PtCl4 with 0.03 and
0.0133 M concentration, respectively. Each impregnation step was
followed by drying and calcination at 500  C for 6 h. The nominal
loading of metals was considered as 0.5% of Pt and 1% of Sn.
2.4. Characterization
The X-ray powder diffraction (XRD) patterns of products were
recorded at ambient temperature using a Philips Xpert system in
the 2q range of 5e50 using Cu-Ka1 radiation (l ¼ 1.54056 A). The
Relative crystallinity (RC %) of the samples were calculated based on
the summation of peak intensities of three main characteristic
peaks of MFI type zeolite at Bragg angles of 2q ¼ 22e25. Synthe-
sized zeolite with the highest crystallinity was considered as the
reference sample. Field-emission electron microscope (Zeiss) was
used to capture FE-SEM images to investigate the particulate
properties of the samples. The particle size distribution histograms
were illustrated by ImageJ software. The chemical composition of
the calcined samples was analyzed by energy dispersive X-ray
Table 1
Summary of the chemical composition and employed seed type for the synthesis of
different samples.
Samples Si/Al ratio Na2O content Seed type
Z-1 30 0.07 TPABr-Silicalite-1
Z-2 60 0.06 TPABr-Silicalite-1
Z-3 120 0.055 TPABr-Silicalite-1
Z-4 60 0.06 TPAOH-Silicalite-1
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
3
spectroscopy and atomic absorption spectrometry. FT-IR spectra
were measured using KBr-diluted pellet on an IR spectrophotom-
eter (PerkinElmer Version 10.03.06). Thermo-gravimetric analysis
(TGA) was performed on NETZSCH TG 209 F1 instrument under air
flow at a ramp of 20  C min1 from 25  C to 800  C. N2 adsorp-
tionedesorption isotherms were determined in a Micromeritics
Tristar-3000 porosimeter. Textural properties including total sur-
face area, total pore volume, micropore volume and average pore
width were measured using BET, t-plot and BJH methods. Before
the surface area measurements samples were degassed at 300  C
for 3 h.
2.5. Catalytic tests
Propane dehydrogenation was carried out in a fixed-bed tubular
quartz micro reactor with a propane and hydrogen mixture as feed.
The reactor was placed in an electrical furnace equipped with a
temperature controller using three thermocouples along the
reactor to maintain the temperature homogenously within ±1  C of
the desired set point. 0.5 g catalyst was placed into the center of
reactor and was heated up to 500  C and maintained at this tem-
perature under nitrogen flow for 2 h to get activated. The reactor
temperature was increased to the desired value of 600  C with a
temperature ramp of 5  C/min shortly afterward. Feed gas flow rate
was controlled by mass flow controllers. Reaction conditions
including pressure, weight hourly space (WHSV), and H2/C3H8
molar ratio were fixed at 1 atm, 4 hr1 and 0.8. Finally products
were analyzed with an online YL-6100 GC gas chromatograph
equipped with a HID detector. Propane conversion, product selec-
tivities, and propylene yield were defined as follows:
P ni
i3 Ciout  Cpropaneout
Propane Conversion ¼ P ni  100
i 3 Ciout
ni
Ciout
Selectivity ¼ P ni 3
 100
i3 Ciout  Cpropaneout
conversionð%Þ  selectivityð%Þ
Yield ð%Þ ¼
100
Where i include all the carbon containing components in the
effluent stream, ni is the number of carbon atoms of component i
and Ci is its molar composition.
A schematic flow chart of experimental setup is shown in
(Fig. S1).
3. Results and discussions
3.1. Zeolites characterization results
The XRD diffractograms of the prepared samples are repre-
sented in (Fig. 1) to investigate the structure and phase purity. All
major characteristic reflections at 2q angles of 7.8, 8.9, 23, 24 and
24.5 corresponding to the MFI topology are present in all products.
As shown by the relative crystallinity values, with increasing the Si/
Al ratio crystallization got more difficult and tended towards (011)
and (020) crystal planes and crystallinity decreased as the result
(see Table 2). But it is important to recall the complication faced
with aluminum incorporation into the framework of zeolites at low
Si/Al ratios. It has been reported that aluminum content in the
batch strongly affects the nucleation and crystallization of ZSM-5
zeolite [11,36].
Appeared peaks around 25.5e27 2q in samples with high Si/Al
ratios (Si/Al  60) especially that synthesized under the direction of
4 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
Fig. 1. XRD patterns of synthesized Z-1, Z-2, Z-3 and Z-4 samples; (7) indicates
keatite phase.
basic TPAOH-silicalite-1 shows that the keatite secondary phase
was formed next to ZSM-5 crystals. Keatite is a dense non-acidic
metamorphic tetragonal modified form of silicon (IV) dioxide. It
accompanies with ZSM-5 and co-crystallizes in the form of semi-
spherical aggregates composed of sheet-like particles that lowers
the efficiency of catalytic application of zeolite. Low acidity disfa-
vors the hydrogen transfer reaction (adsorption of hydrogen from
alkane) which is the rate-limiting step in alkane dehydrogenation
conducted over Pt metallic centers [30]. This result is consistent
with what Kim et al. [36,37] observed in their experiments. They
concluded that the purity and type of the final product using
template-free synthesis strategy was a strict function of Na2O
content. Naþ cations are believed to act as inorganic SDA and charge
balancing agent in organo-template free synthesis method. Sodium
structure-forming cations interact with the complex aluminosili-
cate species and favor the formation of secondary building units by
stabilization of Al-containing structural 5-1 subunits which can
form nuclei to deposit on the surface of seeds and grow. Moreover,
the aluminum content in the framework is directly proportional to
the Naþ content and Naþ cations balance negatively charged
aluminum in the framework. In other words, in the absence of high
amount of tetrahedarlly coordinated Al (high Si/Al), the deficiency
of Naþ cations direct the crystallization path towards the formation
of keatite rather than ZSM-5. However, employing high quantity of
sodium cations could increase the pH of synthesis media and result
in the dissolution of silica species and a decrease in Si/Al ratio.
Therefore, it seems a challenge to regulate the Na2O content pre-
cisely so as to inhibit either secondary phase formation and silica
dissolution problems.
FESEM micrographs along with the relevant particle size dis-
tribution histograms of synthesized zeolites with different Si/Al
ratios and seed types are provided in (Fig. 2). As shown in Fig. 2, Z-1
sample contained ellipsoidal aggregates with mean size of 1 mm
with rugged surfaces that were assembled by small thin stick-
shaped nanocrystals of approximately 400 nm length. Increasing
the Si/Al ratio has led to the formation of hexagonal prism-shaped
crystals of ca. 1.2 mm in both Z-2 and Z-3 samples. Thus, sizes of
particles tended to increase slightly and particle size distributions
Table 2
Physical and chemical features of the crystallized products.
Samples Precursor PH Si/Al in starting gel Si/Al in product RC (%)
EDS AAS
Z-1 12 30 20.2 21.7 100
Z-2 12 60 28.3 33.5 81.1
Z-3 12 120 38.5 45 78.5
Z-4 13.5 60 28.8 e 83
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
became broader with an increase in the Si/Al molar ratio which can
be assigned to the easier formation of crystals form thermody-
namical point of view due to the presence of less supplied Al spe-
cies. It suggests that at high Si/Al ratio (Si/Al  60) each crystal is
created from a single nucleus while at lower Si/Al ratios (Si/Al ¼ 30)
a multi-nuclei site is the origin of a cluster of crystals. This result
can also be found in other reports regarding the seeded zeolite
yielding systems [11,38,39]. Only Hu et al. [40] synthesized several
ZSM-5 samples with different SiO2/Al2O3 ratios under microwave
irradiation and received contrary results on the effect of Si/Al ratio
on the size of crystals. It is assumed that the differences in
morphology and sizes of particles in seed-induced systems are
directly attributed to the quantity of aluminum and sodium cations
of the synthesis media. Aggregation process derived by intercrystal
forces due to the high surface Gibbs free energy of nanocrystals is
responsible for the morphology of Z-1 sample. Moreover, higher
Na2O content of the synthesis reaction mixture can be contem-
plated as another reason for the formation of huge number of
tablet-like nanocrystals. This phenomenon further confirms the
role of Naþ cations as the SDA instead of organic template which led
to an enhancement in number of nuclei and formation of smaller
crystals. The distinct morphology of this sample may result in the
disparate characteristics and catalytic performance as a rule.
As evident from (Fig. 2(c)), unreacted silica species are present
in the Z-3 sample to some extent implying that the obtainable Si/Al
molar ratio in the formation of ZSM-5 molecular sieve in the
absence of organic template is restricted and minimum aluminum
should exist. Consequently it can be concluded that the Si/Al molar
ratio of 60 is the upper limit to consume the whole silica content in
precursor and achieve the complete transformation of primary
amorphous phase to fully crystalline structure.
For the sample Z-4 that was synthesized under the direction of
TPAOH-silicalite-1 seeds, besides the MFI typical morphology:
prism shaped, a great portion of clusters with a flaky morphology
consisting of very thin slices was also observed (Fig. 2(d)). These
clusters with an average size of 6 mm are ascribed to the all silica
compound, second crystalline keatite phase. EDS analysis of these
unusual woolly aggregates in template-free ZSM-5 samples in
Machado et al. [41] work gave an average Si/Al ratio of 43 which
was far form batch Si/Al ratio of 20 confirming the silica-rich
structure of these aggregates. Ren et al. [42] performed a careful
analysis on the chain events in the crystallization/transformation
process of seed-induced OSDA-free synthesis of sub-micrometer
ZSM-5 (see Fig. S2). Keatite secondary phase with lamellar struc-
ture was recognized as the intermediate in the crystallization
process during a time interval (30e115 h) which was in a pseudo-
equilibrium with MFI phase to transform into hollow a-quartz
particles. After z110 h, both as-formed ZSM-5 crystals and keatite
aggregates started to transform into a-quartz, however the rate of
transformation of keatite was evidently faster. It was shown that
keatite slices were formed directly from unexpended amorphous
species being aggregated/condensed in the form of CAs (condensed
aggregates). The slice-like features arranged as the spherical as-
semblies, having the size of about 5e10 composed of small
nanometer-sized flakes.
Solid yield (%) Product phase
90.7 ZSM-5
95.4 ZSM þ keatite (minor by-product)
90.4 ZSM þ keatite (minor by-product)
93 ZSM þ keatite (large population of lamellar aggregates)
 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11 5

Fig. 2. FESEM images and particle size histograms of Z-1 (a) Z-2 (b) Z-3 (c) and Z-4 (d) samples.

Some other factors such as high crystallization temperature
(T ¼ 200  C), lack of room-temperature aging time, low alkalinity of
the synthesis media (Na2O/SiO2  0.05), and prolonged crystalli-
zation time (t  48 h) were named as the reasons for the simul-
taneous formation of keatite aggregates along with ZSM-5 zeolite
nanocrystals [43]. Also, keatite appeared in the product of alumi-
nosilicate seed-assisted system with fragments of NaY zeolite
resulted from poor depolymerization of silica and high viscosity
and inhomogeneity of the system due to the low water content
[44].
Close inspection of FESEM images of Z-2 and Z-4 samples with
similar Si/Al molar ratios which were fabricated from different
seeding suspensions in (Fig. 3) reveals high purity of Z-2 sample in
comparison with Z-4 sample since a minor amount of silice-like
features were present. On the contrary, a large population of ag-
gregates with a flaky habit co-crystallized with ZSM-5 crystals in Z-
4 sample. This has occurred because of high accumulation of OH
ions produced from solution and formation of silicate, aluminate,
and aluminosilicate species and surplus wandering hydroxide ions
in TPAOH-silicalite-1 seeding suspension. As the result of high
alkalinity (PH ¼ 13), dissolution of the silica species has taken over
the aluminum hydroxide ones and the liquid phase was enriched
with silicate anions. It has been reported that the silicate anions
dissolved in the liquid phase cannot act as active growth precursors
during the assembly of MFI-type crystals and only hydrogel solid
phase contribute to the crystallization procedure by the deposition
of the gel onto the surface of silicalite-1 crystals, reconstruction,
and ordering [45e47]. These low-molecular silicate anions serve as
nutrients for the nucleation and growth of more stable keatite
phase through solution-mediated process by polycondensation
mechanism. Besides, gradual dissolution of the formed ZSM-5
crystals which was catalyzed by hydroxide ions by the attack of
OH ions on the SieOeSi bridges further support the growth of
highly siliceous keatite particles. In case of Z-2 sample, Br ions
controlled the PH of reaction mixture by forming hydrobromic acid
and minimized the transformation of amorphous phase to the
undesired keatite by-product.
Owing to the high alkalinity of the Z-4 sample synthesis media,
number of primary nuclei decreased and larger crystals were pro-
duced compared to the Z-2 sample which is in agreement with
Persson et al. results [48]. Chao et al. studied the kinetic of the
formation of zeolite ZSM-5 from Naþ/TPAþ aluminosilicate gels and
disclosed that an increase in OH concentration as the mineralizing
agent could shorten the induction time, but high alkalinities
beyond a certain value would prolong the nucleation period [49].
Accordingly, nucleation rate was affected by alkalinity of the
hydrogel through two mechanisms namely, I) the dissolution of the
gel materials and II) the polymerization of silicate and aluminate
anions to form aluminosilicate ions which act as the source of
nuclei. Second step is restricted at very high alkalinities thereby

Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
6 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11

Fig. 3. Comparison of the amount of competing secondary phase (keatite) in Z-2 and Z-4 samples with identical Si/Al ratio synthesized from different silicalite-1 seeds; the MFI
crystals and keatite domains are bordered by yellow and green arrows in enlarged detail (For interpretation of the references to color in this figure legend, the reader is referred to
the web version of this article.).

decreasing the nucleating rate and lengthening the induction
period. Therefore, larger crystals in Z-4 were produced not only
because of the less nuclei formed as the result of very high alka-
linity of the system arising from sodium hydroxide and TPAþ ions
but also higher conversion rate of Z-4 system due to strong influ-
ence of TPAþ ions.
Some physical and chemical properties of the products are
summarized in Table 2. Among all the zeolitic products, Z-1 sample
with the highest diffraction intensities at the 2q ¼ 22e25 range was
assigned as the reference sample with 100% relative crystallinity and
the relative crystallinity of all other samples were determined
accordingly. The low relative crystallinity of Z-3 sample might be
ascribed to the presence of unreacted amorphous silica species. High
solid recovery yields (>90%) are the greatest values reported ever
and a privilege of employed seeded zeolite yielding approach over
organic-templated crystallization method [13,16,17,46e50]. This
parameter is of great significance for industrial scale productions.
Chemical and elemental analyses were conducted to ascertain
the incorporation of Si and Al into the framework of product ma-
terials. As can be seen from Table 2, The Si/Al molar ratios of the
products are lower than those in starting gels which is indicative of
low availability of silica entities than alumina ones in harsh alkaline
precursor media during synthesis. The dissolution rate of silica is
much higher than the alumina at high alkalinity, therefore the
collaboration of silicate anions in the condensationepolymeriza-
tion chain reactions and formation of growth precursor species in
the gel matrix is restricted. This is in conformity with the
“aluminum enrichment of the solid phase” statement which has been
reported as the reason for the decrease of Si/Al ratio of the solid
phase of the reaction mixture according to the literature [46,47]. In
other word, the low utilization of silica source in the presence of
solely sodium as the SDA can be named as the main reason for the
difficult insertion of Si elements into the framework and low Si/Al
ratio in this method. Similar trend was observed by Wu et al.
concluding that the consumption of aluminum during the nucle-
ation step led to the formation of ZSM-5 zeolites with low Si/Al
ratios [17]. At this juncture, it is interesting to mention that the Si/Al
ratio of the final product in the seed-induced synthesis method-
ology is a function of seed additional amount from the previous
literature report [51]. 32% of silicalite-1 seed could afford the
suitable condition for the formation of submicron sized ZSM-5
particles with Si/Al ratio of 30 equal to the nominal ratio while by
introducing 5% of seeds to the system Si/Al ratio of 17 could be
achieved. (Fig. 4) shows that a linear equation on the basis of Si/Al
ratio of precursor gel can well predict the Si/Al ratio of the final
product. Observed difference suggests that the Si/Al ratio of the
ZSM-5 materials can be easily tuned by adjusting the Si/Al ratio of
the starting gel in seed-induced synthesis strategy.
FT-IR test was carried out in the frequency range of
400e4000 cm1 to further identify the structural characteristics of
the resultant materials and the obtained graphical data are
exhibited in (Fig. 5). The position, strength and possible assignment
of various IR bands are gathered in (Table 3), in light of the relevant
reports [45e48]. These adsorption bands are typical characteristic
bands of highly siliceous MFI structure and are in good agreement
with those reported in literature [52e55]. Detected bands in the
region of 400e1300 cm1 are due to the framework vibrations and
give clues on the skeletal features while bands in 3400e3800 cm1
region signify the existence of bridging hydroxyl acidic sites. All
patterns are dominated by a broad stretching vibration frequency at
~1095 cm1 with a shoulder at ~1225 cm1 which shifted towards
higher wavenumbers with increasing the Si/Al ratio from 30 to 120.
Maximum of the SieOeT asymmetric stretching band in silica-
based materials is reported to be sensitive to the Al content in
the framework and shifts to lower frequencies when the Al con-
centration increases [55]. Additionally, the external asymmetric

Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11 7

Fig. 4. Si/Al ratio relationship between starting gels and products (Z-1, Z-2, and Z-3).
Fig. 6. TGA curves of the calcined and coked samples.

synthesized material. (Fig. 6) displays the thermogram for

Fig. 5. FT-IR spectra of synthesized Z-1, Z-2, Z-3, and Z-4 samples.
stretching vibration around 1225 cm1 implies that structure
contains four chains of five-membered rings arranged around a
twofold screw axis [56]. Since the weak intensity of the band at
~550 cm1 indicates the low ordering of the structure and amor-
phous nature of the material, it is rational to conclude the high
crystallinity of Z-1 sample among TPABR-silicalite-1 directed zeo-
lites which were confirmed by XRD analysis as well. There is a weak
band around 3650 cm1 in all four samples which is associated
with the OH groups connected with extra lattice Al atoms that
possess lower acid strength compared to structural OH groups. The
absorbance got weakened and nearly vanished with increasing Si/
Al ratio because of the low aluminum content and better distri-
bution and incorporation of aluminum species. OeH stretching of
the bridging Si(OH)Al groups located at the internal zeolite surfaces
in Z-2, Z-3, and Z-4 samples is evident. These hydroxyl groups are
active acidic centers that are responsible for acid-catalyzed
reactions.
TGA analysis was employed to quantitatively determine the
degree of template removal from the backbone of seed-directed
calcined Z-1 sample. A major weight loss can be observed below
200  C which is attributed to the desorption of physically adsorbed
water. There is no considerable weight loss between 200 and
800  C which corresponds to the decomposition of organic tem-
plate molecules. This data demonstrates that most of the template
molecules introduced by the seeding feed were removed from the
structure of the material and a perfect network with well-
developed porosity formed. Additionally, TGA analysis was per-
formed for PtSn/Z-2 spent catalyst to calculate the amount of coke
and specify its nature. Recorded data in Fig. 6 shows three weight
loss in T < 200  C, 200  C < T < 400  C and 400  C < T < 800  C
temperature ranges. First major weight loss corresponds to the
release of moisture and other liquid reactants while second and
third weight losses imply the cracking and oxidation of carbona-
ceous deposits. Around 0.9% weight loss was observed in second
weight loss step which may arise from the removal of alkyl species
from aromatics (soft coke) while 0.5% third step loss is believed to
result from the removal of bulky polyaromatics (hard coke).
Therefore it is concluded that high temperature favors the forma-
tion of hard coke. It seems that air regeneration at high tempera-
tures (T ¼ 500e600  C) can efficiently remove a great portion of the
produced coke from catalyst structure and recover the catalytic
reactivity.
Textural properties of zeolite samples were measured using N 2
sorption data by means of BET, t-plot and BJH methods. Results are
given in (Table 4). As can be observed, Z-2 sample possess the
highest total and micropore surface area whereas Z-1 sample has
the highest mesopore volume. Lower surface area of the Z-3
product originates from the presence of unreacted silica species in
the form of amorphous phase matrix in the product domain. Per-
ezeRamiz et al. [57] gathered vast amount of represented literature
data and categorized different hierarchical zeolitic materials on the
basis of hierarchy factor to provide a generic overview on the
porosity of large number of hierarchal zeolite architectures. The

Table 3
Position, strength and possible assignment of the common IR bands of the Z-1, Z-2, Z-3, and Z-4 samples.

Wave number (cm1) Assignment

451-455 (s) T-O bending vibrations of SiO4 and AlO 4

547-548 (s) Asymmetric stretching modes of the double five-membered rings of MFI structure
794-798 (ms) Asymmetric and symmetric vibrations of TOT (T ¼ Si, Al) groups
1094-1099 (s) Asymmetric stretching vibrations of SieOT (T ¼ Si, Al) linkages
1223-1228 (w) Asymmetric stretching modes of TO (T ¼ Si, Al) external linkages
1626-1634 (w) Stretching vibrations of adsorbed water
3444-3458 (ms) Stretching vibrations of terminal SieOHeAl hydroxyl bridges
3648-3666 (w) Stretching vibrations of terminal AleOH groups

Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
8 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
hierarchy factor (HF) is a quantitative tool to classify hierarchically
structured materials which is defined as the product of (Vmicro/
Vpore)  (Smeso/SBET) [57]. Porous systems on the top left of the
curves in (Fig. S3) are deeply mesoporous materials exhibiting low
degree of microporosity while bottom right represents highly
microporous samples with low relative mesoporosity. Generally
mesopores are formed scarifying microporosity, thus all the efforts
have been focused on the coupling of the native micropores with an
auxiliary mesopore network of intercrystalline or intracrystalline
nature. Seeding is located slightly top and to the left of the con-
ventional zeolites with HF > 0.1 displaying highly microporous
systems with relatively low degree of mesoporosity. Our results are
in great consistency with the represented data confirming the
formation of hierarchical porous network without using a sec-
ondary mesogenous template.
N2 adsorptionedesorption isotherms and the relevant BJH pore
size distribution profiles are demonstrated in (Fig. 7). Isotherms are
clearly found to be a composite of types I and IV, corresponding to
microporous and mesoporous materials in accordance with IUPAC
classification. The steep uptake in the very low relative pressure
region (p/p0 < 0.1) is the typical property of the microporous
compounds. Hysteresis loop appearing in the physisorption iso-
therms at the relative pressure in the range of P/P0 > 0.45 is usually
associated with the capillary condensation taking place in the
mesopores. Shape of hysteresis loop in all samples is close to H4
with two nearly horizontal branches over a wide range of P/P0
which has been identified with specific slit-like pores. In another
interpretation, type H4 merely verifies the existence of large mes-
opores implanted in a matrix of tiny pores. Z-1 sample showed the
highest nitrogen adsorption amount and either of Z-1 and Z-3
isotherms did not show any limiting adsorption at high relative
pressures suggesting that both samples were open surface mate-
rials with trimodal porosity that allowed the continuous uptake
and formation of multiple adsorbate layers as the p/p0 increased.
BJH pore size distribution of these two samples assert the trimodal
porosity considering micro-meso and macro pores as the conse-
quence of abundant aggregation of the nano-crystallites in Z-1
sample and production of pinholes and distorted defects through
ordering the structure of amorphous material in Z-3 sample. Close
stacking of single stick-shaped crystals to ellipsoidal agglomerates
observed in Fig. 2(a) and mild upward turn at high P/P0 float the
idea of inter-crystalline mesoporosity based on the increscent in-
terspaces located in the vicinity of crystallite surfaces. Inter-
crystalline mesopores are the voids constructed upon the aggre-
gation of individual crystals wherein the size and intensity of
packing determines the severity of mesoporosity.
Sample Z-2 seems to be endowed with a tuned bimodal porous
network with a relatively huge number of mesopores with average
width of 5 nm embedded amid the micropore framework. These
supplementary mesopores would couple with the intrinsic
Table 4
Textural properties of the obtained products.
Samples SBET (m2/g) Sext (m2/g)a Vt (m3/g)b Vmeso (m3/g)c D (nm)d HFe
Z-1 275.7 61.5 0.180 0.083 8.4 0.12
Z-2 287.4 68.8 0.161 0.063 5 0.15
Z-3 238.4 60.7 0.146 0.067 7.5 0.14
Z-4 281.1 70.2 0.143 0.047 5.5 0.17
a
Sext ¼ SBETeSmic; the micropore area and volume are determined using t-plot
method.
b
Total pore volume was estimated based on the volume adsorbed at p/
p0 z 0.995.
c
Vmeso ¼ VteVmic.
d
Mean pore width was determined using BJH method.
e
The hierarchy factor ¼ (Vmicro/Vpore)  (Smeso/SBET).
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
micropores and supply extra transport pores to improve the
accessibility and molecular diffusions propitiously. Diffusion limi-
tations not only lead to a negative influence on catalytic activity, but
also can cause a loss in selectivity and a reduction in the lifetime of
the catalyst. Designated hybrid micro/mesoporous zeolites are
believed to be more functional than their purely micropore coun-
terparts in chemical reactions dealing with bulky intermediates
and/or products. Faster migration of guest molecules in the host
zeolite framework affects activity, selectivity and stability as the
three key features of the catalysts.
3.2. Zeolites catalytic performance
All prepared zeolitic samples were evaluated in direct propane
dehydrogenation reaction as supports to investigate the efficiency
of seed-directed low-cost synthesis procedure for obtaining reac-
tive catalytic sieve structures. But a blank test was conducted prior
to catalytic tests to emphasize on the influence of catalyst in the
conversion of propane to desirable olefinic products. (Fig. S5). A
small amount of propane was only converted. Low propylene
selectivity and high methane selectivity in comparison with other
catalytic tests suggest that thermal cracking was the only possible
reaction taking place at high temperature.
Performances of PtSn-drived ZSM-5 catalysts are depicted in
Figs. 8e10 in terms of propane conversion, propylene selectivity
and yield of products. The first point that can be extracted
considering propylene, methane and ethylene sectivities is the
substantial role of catalyst in restricting thermal cracking as the
most important side-reaction competing against dehydrogenation
reaction. It is important to point out the notable differences
observed between blank test and these catalytic runs suggesting
the huge effect of active metallic phase. Pt metal sites actively ab-
stracts hydrogen from paraffin and later the acid sites provided by
the support selectively convert intermediates into light olefins. Sn
improves the dispersion of Pt ensembles (geometrical effect) and
strengthens the interfacial interaction of the metallic phase with
the support which affects the conversion and propylene selectivity
in great extent. In fact Sn helps in selective control of the stereo-
chemistry of the reaction by reducing the catalyst acidity and the
size of Pt ensembles, thereby inhibiting the side reactions to be
carried out and formation of highly dehydrogenated species which
can be served as the coke precursor.
A noticeable difference in catalytic reactivity of all samples is
observed in (Fig. 8) which is believed to be strictly related to the
acidic pattern (Si/Al ratio), degree of purity, porosity characteristics,
crystallinity and crystal size distribution of the resultant materials.
In accordance with the current opinion on the inverse relationship
between Si/Al ratio of the zeolite and dominant acidic regime, we
anticipated Ptsn/Z-1 catalyst to represent the highest activity
converting propane in PDH reaction. However, we obtained inter-
esting different results concluding that Z-2 support with nominal
Si/Al ratio of 60 was the best carrier with respect to high catalytic
conversion (X ¼ 42.2e30.9) and propylene selectivity (85.7e78.5)
as the major desired product. It seems that different underlying
nucleation mechanism and non-uniform aluminum distribution
throughout the crystals caused high amount of extra framework
aluminum species in the form of AleOH hydroxyl groups with low
acid strength which were the main reasons contributing to the
lower activity of the PtSn/Z-1 catalyst. Stability factor (Xf/X0) for
three bi-metallic Z-1, Z-2, and Z-3 catalysts were estimated 80%,
73%, and 59%, respectively. This trend is consistent with that re-
ported in Ref. 27 elucidating the quick deactivation at high Si/Al
ratio. One reason for higher stability of PtSn/Z-1 catalyst might be
attributed to the presence of wider mesopores hampering pore
blockage with coke deposits and heavy carbonaceous compounds
 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11 9

Fig. 7. N2 adsorptionedesorption isotherms and the relevant BJH pore size distribution profiles of synthesized samples: (a) Z-1, (b) Z-2, (c) Z-3, and (d) Z-4.

Fig. 8. Propane conversion versus time on stream over different PtSn-based zeolite Fig. 9. Propylene selectivity versus time on stream over different PtSn-based zeolite
samples. samples.

as coke precursors. Moreover, existence of unreacted silica matrix
in the Z-3 sample backbone was conducive to the intensification of
the deactivation phenomenon. Z-4 sample did not show a good
performance because of the large dense keatite aggregates and
large crystals with low degree of mesoporosity which downgraded
the efficiency of material. The catalytic test for this sample was
conducted twice but similar values were obtained, thus only 4 h
was considered to report the results.
Similar trend was observed for the selectivity of propylene in
Fig. 9. PtSn/Z-2 catalyst was recognized as the most selective
catalyst with more than 90% of selectivity towards light olefins
namely, propylene and ethylene. A credible theory has been pro-
posed by Koyama et al. [58] for the production of propylene on the
basis of zeolite pore volume introducing Di as the diameter of the
largest sphere that can be included in the zeolite framework. It was
assumed that propylene can be produced selectively over zeolite on
condition that its pore volume accommodates the volume of the
alkanes and their carbocations counterparts. Given that, 10-
membered rings ZSM-5 framework with Di ¼ 6.3 A sounds to be
a great candidate for the accommodation of propane (Di ¼ 5.5 A)
and Cþ 3 carbenium ions (Di ¼ 5 A) if we assume that PDH reaction
proceeds through Olah's monomolecular mechanism [59,60]. Ac-
cording to this mechanism, zeolite protonates alkane at high tem-
perature and creates a pentacoordinated carbonium ion which can
rearrange and yield a molecular H2 and carbenium ion. Protolytic
cracking side reaction (CeC bond dissociation) can further accel-
erate over strong acid sites by recombining an alkyl moiety with
initial formed Cþ3 carbenium ions lowering propylene selectivity.

Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
10 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
Fig. 10. Average yields of products over different PtSn-based zeolite samples (Reported
yields are time-averaged values).
Therefore a special arrangement of acid sites mostly consisting of
weak and mild acid sites is required to enhance control over the
stereo-chemistry of the reaction and boost up propylene yield
efficaciously. Change of the Si/Al ratio of the zeolitic support
directly altered the dominant acid character. Nominal Si/Al ratio of
60 provided the necessary conditions to convert propane properly
to olefinic products with controlled composition highlighting
propylene product content. Initial Selectivity of the PtSn/Z-3 was
slightly higher than that of PtSn/Z-2 sample since their real Si/Al
ratios differed a little from one another according to the AAS
analysis results. But then again higher selectivity dropping rate
underlines the negative influence of wandering unreacted silica
species on the propylene selectivity. PtSn/Z-4 sample was neither
reactive nor selective as the consequence of large population of
keatite slices as the secondary phase.
(Fig. 10) represents average product yields over different PtSn-
derived zeolite catalysts. Formation of ethylene over PtSn/Z-2
catalyst is impeded in comparison with other catalysts based on
the suitable distribution of acid sites (moderate Si/Al ratio) and
consequently homogenous metallic species as the active centers.
Available information on catalytic behavior of different bi-
metallic (PtSn-based) ZSM-5 catalysts in PDH reaction is gathered
in Table S1 from open literature [61e64]. Data are compared with
the performance of our optimum catalyst i.e. PtSn-Z2 to be able to
evaluate the efficiency of the synthesis procedure. High conversion
rate, prominent selectivity towards propylene and high stability
factor of designated ZSM-5 catalyst confirm the success of catalyst
formulation and highlight the influence of synthesis parameters in
enhancing the catalytic properties. It seems that piled up zeolite
units outfitted stable cores and improved catalyst stability param-
eter which was comparable to other represented values in Table S1.
Stability factor is directly assigned to the rate of deactivation in
which active sites and micropores are covered with heavy carbo-
naceous products. The regeneration of a similar zeolitic catalyst
consisting of nearly equal portions of ZSM-5 and SAPO-34 with hot
air at 480  C and its second time evaluation showed that catalyst
could attain its initial activity mainly. The catalyst could act like a
fresh catalyst with somewhat lower deactivation rate. Bringing in
mind soft nature of great portion of the deposited coke according to
TGA result we can anticipate similar good result about the ZSM-5
spent catalysts.
The metallic compositions of the three fresh and spent catalysts
were obtained from AAS analysis, and shown in Table 5. It is already
confirmed that the metallic contents are successfully loaded on the
support surface of the fresh catalyst. However after the reaction,
PtSn/Z-1 catalyst has lost a part of its active phase. It was previously
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
Table 5
Metallic content of fresh and spent catalysts.a
Samples Pt content (% w/w) Sn content (% w/w)
PtSn/Z-1 (fresh catalyst) 0.502 1.03
PtSn/Z-1 (spent catalyst) 0.32 1.15
PtSn/Z-3 (spent catalyst) 0.52 1.2
a
Results from AAS analysis.
demonstrated that a portion of aluminum in Z-1 support is not
incorporated in the structure and exists as extra lattice AleOH
groups with lower degree of acidity compared to the structural
SieOHeAl bridging groups. However, strong Bronsted acid sites are
required to adsorb metals and create active centers with well-built
interaction with support. The chemical composition and structural
characteristics of zeolite specially the environment and number of
Al framework atoms adjacent to silanol groups affect the strength
of acidic and consequently produced active sites. The Lewis acid
sites are usually associated with extra framework Al species (EFAL)
or metal ions exchanged for the protons of Bronsted acid sites
[65,66]. Activation of CeH bonds in alkanes on a zeolite matrix
mainly occur on strong Lewis acidic centers. Therefore, it seems
that the adsorption of metal species over the present acidic sites
have not occurred strictly and loose bonds were formed because of
the existence of defected aluminum groups in Z-1 sample and that
may be the reason for detachment and remove of a fragment of
platinic centers during the reaction. Investigation on the other
catalyst, PtSn/Z-3, reveals that no active metal is removed from the
surface, however steeper deactivation of this kind of catalyst may
be raised from sintering of the metals at the surface during high
temperature reaction [67e69].
4. Conclusion
In summary, four hierarchical ZSM-5 zeolites with versatile Si/Al
ratios (30e120) were synthesized under organo-template free
conditions with the aid of TPABr/TPAOH-templated -silicalite-1
seeds as the SDA without using a secondary mesogenerator agent.
This two-step synthesis technique exploits an enriched aluminum
template-free precursor gel containing zeolite nuclei that attach
and grow around high siliceous zeolite silicalite-1 cores. Therefore,
self-nucleation time-consuming step is mostly eliminated by
introducing extraneous seeds with similar framework and common
building units. Seeds supply the specific sites for the migration of
nuclei and growth of zeolite crystals. This is a facile, efficient, and
reproducible route minimizing the use of organic template to
fabricate ZSM-5 zeolite with uniform particle size distribution and
high solid yield. This method is a low-cost approach wherein the
credit goes to the feasibility of large-scale production.
Characterizing the resultant materials revealed that in the
products obtained at Si/Al  60, semi-spherical lamellar slice-like
particles coexisted with ZSM-5 as the secondary phase to some
extent depending on the type of employed seeding feed. The wooly
aggregates in these samples are believed to be the dense crystalline
tetragonal form of silica named keatite which is normally a com-
panion to MFI zeolites. Therefore, Si/Al < 60 was the threshold for
obtaining fully crystalline pure MFI structure in the seed surface
crystallization mechanism. However, the quantity of keatite phase
in the products obtained from TPABr-silicalite-1 seeds was negli-
gible. Therefore, ZSM-5 product from crystallization of TPABr-
silicalite-1 seed-induced system with more uniform small crystals
and composite porous structure containing supplementary meso-
pores was selected advantageously. The catalytic behavior of all
samples in bimetallic PteSn design was evaluated in PDH process
to investigate the influence of structural and textural features of the
 M. Razavian, S. Fatemi / Materials Chemistry and Physics xxx (2015) 1e11
prepared zeolites with different Si/Al molar ratios and seed types
on their performances as support. Si/Al ratio or in another word
acidity which is strictly associated with aluminum distribution
throughout the crystals, purity, and porosity aspects (mesopores
and 3-D channels) can be named as the main factors affecting
reactivity. These parameters play critical role in the production of
light olefins in a selective trend. PtSn/ZSM-5 catalyst with nominal
Si/Al ratio of 60 which was synthesized applying TPABr-silicalite-1
seeds, demonstrated the highest activity and selectivity towards
light olefins mainly propylene. On the contrary, ZSM-5 sample
crystallized under the direction of TPAOH-silicalite-1 seeds showed
unsatisfactory results subsequent to large population of dense
keatite woolly slice-like assemblies and big MFI crystals with low
degree of mesoporosity.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.matchemphys.2015.08.051.
References
€cker, H.G. Karge, J. Weitkamp, Advanced Zeolite Science and
[1] J.C. Jansen, M. Sto
Applications, Elsevier Science B.V., Amsterdam, 1994.
[2] J. Cejka, A. Corma, S. Zones, Zeolites and Catalysis: Synthesis, Reactions and
Applications, Wiley-VCH, Weinheim, 2010.
[3] M.A. Martin-Luengo, M. Yates, J. Mater. Sci. 18 (1995) 4483e4491.
[4] E.G. Derouane, S. Determmerie, Z. Gabelica, N. Blom, Appl. Catal. 1 (1981)
201e224.
[5] J.K. Reddy, K. Motokura, T.R. Koyama, A. Miyaji, T. Baba, J. Catal. 289 (2012)
53e61.
[6] X.H. Vu, U. Bentrup, M. Hunger, R. Kraehnert, U. Armbruster, A. Martin,
J. Mater Sci. 49 (2014) 5676e5689.
[7] Y. Takamitsu, K. Yamamoto, S. Yoshida, H. Ogawa, T. Sano, J. Porous Mater. 21
(2014) 433e440.
[8] G. Reding, T. Maurer, B. Kraushaar-Czarnetzki, Microporous Mesoporous
Mater. 57 (2003) 83e92.
[9] B. Xie, H. Zhang, C. Yang, S. Liu, L. Ren, L. Zhang, X. Meng, B. Yilmaz, U. Muller,
F.S. Xiao, Chem. Commun. 47 (2011) 3945e3947.
[10] C.S. Cundy, P.A. Cox, Microporous Mesoporous Mater. 82 (2005) 1e78.
[11] T. Xue, Y.M. Wang, M.Y. He, Microporous Mesoporous Mater. 156 (2012)
29e35.
[12] Y. Tang, B. Li, N. Zhang, S. Wang, Y. Wen, P. Jin, X. Wang, CrystEngComm 14
(2012) 3854e3857.
[13] K. Itabashi, Y. Kamimura, K. Iyoki, A. Shimojima, T. Okubo, J. Am. Chem. Soc.
134 (2012) 11542e11549.
[14] Q. Yu, Q. Zhang, J. Liu, C. Li, Q. Cui, CrystEngComm 15 (2013) 7680e7687.
[15] Y. Lu, N.Q. Zhang, Q. Tong, J.K. Liu, D.J. Hong, Nano (2013), http://dx.doi.org/
10.1142/S1793292013500276.
[16] H. Zhang, K. Song, L. Wang, H. Zhang, Y. Zhang, Y. Tang, ChemCatChem 5
(2013) 2874e2878.
[17] G. Wu, W. Wu, X. Wang, W. Zan, W. Wang, C. Li, Microporous Mesoporous
Mater. 180 (2013) 187e195.
[18] H. Pan, Q. Pan, Y. Zhao, Y. Luo, X. Shu, M. He, Ind. Eng. Chem. Res. 49 (2010)
7294e7302.
[19] Y. Tang, B. Li, N. Zhang, S. Wang, Y. Wen, P. Jin, X. Wang, CrystEngComm 14
(2012) 3854e3857.
[20] January, Evolving Propylene Sources e Solution to Supply Shortages? Report
by ChemSystems Products, NEXANT, INC, 2012.
[21] January, Alternative Routes to Propylene. Report by ChemSystems Products,
NEXANT, INC, 2010.
[22] C.N. Eng, S.C. Kang, S. Choi, S.H. Oh, Y.K. Park, AIChE and EPC, Prepared for
Presentation at the 2007, Spring National Meeting, Houston, Texas, 2007.
[23] H.A. Wittcoff, B.G. Reuben, J.S. Plotkin, Industrial Organic Chemicals, John
Wiley and Sons, Inc.,, New Jersy, 2012.
[24] P.K. Ladwig, J.E. Asplin, G.F. Stuntz, W.A. Wachter, B.E. Henry (2000), US Patent
6,069,287.
[25] A. Corma, F.V. Melo, L. Sauvanaud, F. Ortegaa, Catal. Today 107e108 (2005)
699e706.
Please cite this article in press as: M. Razavian, S. Fatemi, Synthesis and evaluation of seed-directed hierarchical ZSM-5 catalytic supports:
Inductive influence of various seeds and aluminosilicate gels on the physicochemical properties and catalytic dehydrogenative behavior,
Materials Chemistry and Physics (2015), http://dx.doi.org/10.1016/j.matchemphys.2015.08.051
11
[26] M.M. Bhasin, J.H. McCain, B.V. Vora, T. Imai, P.R. Pujado , Appl. Catal. A General
221 (2001) 397e419.
[27] C. Yu, Q. Ge, H. Xu, W. Li, Catal. Lett. 112 (2006) 197e201.
[28] Y. Zhang, Y. Zhou, A. Qiu, Y. Wang, Y. Xu, P. Wu, Catal. Commun. 7 (2006)
860e866.
[29] H. Liu, Y. Zhou, Y. Zhang, L. Bai, M. Tang, Ind. Eng. Chem. Res. 48 (2009)
5598e5603.
[30] Z. Nawaz, T. Xiaoping, F. Wei, Korean J. Chem. Eng. 26 (2009) 1528e1532.
[31] Z. Nawaz, S. Qing, G. Jixian, X. Tang, F. Wei, J. Ind. Eng. Chem. 16 (2010) 57e62.
[32] M.A. Ali, B. Brisdon, W.J. Thomas, Appl. Catal. A General 252 (2003) 149e162.
[33] X. Zhu, S. Liu, Y. Song, L. Xu, Appl. Catal. A General 288 (2005) 134e142.
[34] S. Narayanan, S. Sultana, Q.T. Le, A. Auroux, Appl. Catal. A General 168 (1998)
373e384.
[35] M. Razavian, S. Fatemi, M. Masoudi-Nejad, Adsorpt. Sci. Technol. 32 (2014)
73e87.
[36] S.D. Kim, S.H. Noh, K.H. Seong, W.J. Kim, Microporous Mesoporous Mater. 72
(2004) 185e192.
[37] S.D. Kim, S.H. Noh, J.W. Park, W.J. Kim, Microporous Mesoporous Mater. 92
(2006) 181e188.
[38] S. Narayanan, S. Sultana, Q.T. Le, A. Auroux, Appl. Catal. A General 168 (1998)
373e384.
[39] M. Mediavilla, H. Morales, L. Melo, A.B. Sifontes, A. Albornoz, A. Llanos,
D. Moronta, R. Solano, J.L. Brito, Microporous Mesoporous Mater. 131 (2010)
342e349.
[40] Y. Hu, C. Liu, Y. Zhang, N. Ren, Y. Tang, Microporous Mesoporous Mater. 119
(2009) 306e314.
[41] F.J. Machado, C.M. Lopez, M.A. Centeno, C. Urbina, Appl. Catal. A General 181
(1999) 29e38.
[42] N. Ren, J. Broni c, S. Bosnar, M.D. Sikiri
c, T.A. Jeli
c, J.J. Maoa, B. Suboti c, Crys-
tEngComm 15 (2013) 5032e5044.
[43] Y. Cheng, R.H. Liao, J.S. Li, X.Y. Sun, L.J. Wang, J. Mater Process Technol. 206
(2008) 445e452.
[44] H. Pan, Q. Pan, Y. Zhao, Y. Luo, X. Shu, M. He, Ind. Eng. Chem. Res. 49 (2010)
7294e7302.
[45] E.G. Derouane, Z. Gabelica, J. Solid State Chem. 64 (1986) 296e304.
[46] N. Ren, J. Bronic, B. Subotic, X.C. Lv, Z.J. Yang, Y. Tang, Microporous Meso-
porous Mater. 139 (2011) 197e206.
[47] N. Ren, J. Bronic, B. Subotic, Y.M. Song, X.C. Lv, Y. Tang, Microporous Meso-
porous Mater. 147 (2012) 229e241.
[48] A.E. Persson, B.J. Schoeman, J. Sterte, J.E. Otterstedt, Zeolites 15 (1995)
611e619.
[49] Chao KJ, Tasi TC, Che MS, J. Chem. Soc. Faraday Trans. 1 77:547e555.
[50] G. Majano, A. Darwiche, S. Mintova, V. Valtchev, Ind. Eng. Chem. Res. 48
(2009) 7084e7091.
[51] N. Ren, Z.J. Yang, X.C. Lv, J. Shi, Y.H. Zhang, Y. Tang, Microporous Mesoporous
Mater. 13 (2010) 103e114.
[52] P. Jacobs, H. Beyer, J. Valyon, Zeolites 1 (1981) 161e168.
[53] T. Armaroli, L.J. Simon, M. Digne, T. Montanari, M. Bevilacqua, V. Valtche,
J. Patarin, G. Busca, Appl. Catal. A General 306 (2006) 78e84.
[54] L. Shirazi, E. Jamshidi, M.R. Ghasemi, Cryst. Res. Technol. 43 (2008)
1300e1306.
[55] W. Mozgawa, J. Mol. Struct. 596 (2001) 129e137.
[56] J.C. Jansen, F.J. Van der Gaag, H. Van Beckum, Zeolites 4 (1984) 369e372.
[57] J. Perez-Ramirez, D. Verboekend, A. Bonilla, S. Abello, Adv. Funct. Mater. 19
(2009) 3972e3979.
[58] T.R. Koyama, Y. Hayashi, H. Horie, S. Kawauchi, A. Matsumoto, Y. Iwase,
Y. Sakamoto, A. Miyaji, K. Motokura, T. Baba, Phys. Chem. Chem. Phys. 12
(2010) 2541e2554.
[59] G.A. Olah, R.H. Schlosberg, J. Am. Chem. Soc. 90 (1968) 2726e2727.
[60] G.A. Olah, G. Klopman, R.H. Schlosberg, J. Am. Chem. Soc. 91 (1969)
3261e3268.
[61] H. Li, Z. Shijian, Z. Yuming, Z. Yiwei1, Z. Shaobo, X. Jun, China Pet. Process Pe
14 (2012) 75e82.
[62] Z. Nawaz, X. Tang, Q. Zhang, S. Wang, W. Fei, Catal. Commun. 10 (2009)
1925e1930.
[63] S. Zhang, Y. Zhou, Y. Zhang, L. Huang, Catal. Lett. 135 (2010) 76e82.
[64] L. Bai, Y. Zhou, Y. Zhang, H. Liu, M. Tang, Catal. Lett. 129 (2009) 449e456.
[65] S.M.T. Almutairi, The Role of Lewis and Brønsted Acidity for Alkane Activation
over Zeolites, Eindhoven, Technische Universiteit Eindhoven, Eindhoven,
2013.
[66] N. Rahimi, R. Karimzadeh, Appl. Catal. A General 398 (2011) 1e17.
[67] S.E. Wank, P.C. Flynn, Catal. Rev. Sci. Eng. 12 (1) (1975) 93e135.
[68] L. Yaofang, Y. Jiujing, Y. Guanghua, Chem. Technol. Fuels Oils 30 (1995)
353e359.
[69] T. Kanazawa, Appl. Catal. B Environ. 65 (2006) 185e190.