Effect of SiO2/Al2O3 ratio on the performance of nanocrystal ZSM-5 zeolite catalysts in methanol to gasoline conversion

Article abstract of DOI:10.1016/j.apcata.2016.05.032

In this study, the effect of SiO₂/Al₂O₃ ratio on the catalytic performance of nanocrystal ZSM-5 zeolite catalyst in the methanol to gasoline conversion (MTG) was investigated. A series of zeolite samples with different SiO₂/Al₂O₃ ratios of 23, 47, 107, 217 and 411 were synthesised. Through systematically controlling the material synthesis conditions, these nanocrystal ZSM-5 zeolite samples were produced to have very similar crystal sizes and structural properties, thus providing an ideal opportunity to study the intrinsic effect of SiO₂/Al₂O₃ ratio on the performance of the ZSM-5 samples in MTG. The MTG experimentation was carried out in a fixed-bed reactor under a set of constant conditions of temperature 375?°C, pressure 1?MPa and WHSV 2?h^?1. A steady methanol conversion was sustained with increasing the SiO₂/Al₂O₃ ratio, and a progressive decrease in methanol conversion was found over catalysts with low SiO₂/Al₂O₃ ratios (≤107) after 5?h on stream. Decreasing the SiO₂/Al₂O₃ ratio promoted C₁-C₄ selectivity and thus decreased gasoline yield. It was also found that decreasing the SiO₂/Al₂O₃ ratio promoted aromatisation reactions and hence higher durene selectivity and more coke formation, resulting in rapid catalyst deactivation. The sample with SiO₂/Al₂O₃ ratio of 217 showed the highest methanol conversion, gasoline yield, and very low coke formation.

Full text of DOI:10.1016/j.apcata.2016.05.032

Accepted Manuscript
Title: Effect of SiO2/Al2O3 Ratio on the Performance of
Nanocrystal ZSM-5 Zeolite Catalysts in Methanol to Gasoline
Conversion
Author: Zhijian Wan Wei Wu Gang Kevin Li Chuanfu Wang
Hong Yang Dongke Zhang
PII:
S0926-860X(16)30307-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.05.032
APCATA 15919
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
6-3-2016
24-5-2016
30-5-2016
Please cite this article as: Zhijian Wan, Wei Wu, Gang Kevin Li, Chuanfu Wang, Hong
Yang, Dongke Zhang, Effect of SiO2/Al2O3 Ratio on the Performance of Nanocrystal
ZSM-5 Zeolite Catalysts in Methanol to Gasoline Conversion, Applied Catalysis A,
General http://dx.doi.org/10.1016/j.apcata.2016.05.032
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Effect of SiO2/Al2O3 Ratio on the Performance of Nanocrystal ZSM-5
Zeolite Catalysts in Methanol to Gasoline Conversion
Zhijian Wan, Wei Wu, Gang (Kevin) Li*, Chuanfu Wang, Hong Yang and Dongke Zhang
Centre for Energy (M473), The University of Western Australia, 35 Stirling Highway,
Crawley, WA 6009, Australia
(A manuscript offered to “Applied Catalysis A: General”)
* Corresponding author:
Dr Gang (Kevin) Li
Email: Kevin.li@uwa.edu.au
Phone: +61 8 6488 4636
Fax:
+61 8 6488 1024
1

Graphical abstract
Highlights
 Nanocrystal ZSM-5 of different SiO2/Al2O3 ratios but same sizes were obtained
 The effect of SiO2/Al2O3 ratio on methanol to gasoline was systematically studied
 Increasing SiO2/Al2O3 ratio reduced durene yield and improved catalyst durability
 Low SiO2/Al2O3 ratios promoted C1-C4 and durene selectivity
 Catalysts with low SiO2/Al2O3 ratios deactivated more rapidly due to coking
Abstract
In this study, the effect of SiO2/Al2O3 ratio on the catalytic performance of nanocrystal ZSM-
5 zeolite catalyst in the methanol to gasoline conversion (MTG) was investigated. A series of
zeolite samples with different SiO2/Al2O3 ratios of 23, 47, 107, 217 and 411 were synthesised.
Through systematically controlling the material synthesis conditions, these nanocrystal ZSM-
5 zeolite samples were produced to have very similar crystal sizes and structural properties,
thus providing an ideal opportunity to study the intrinsic effect of SiO2/Al2O3 ratio on the
performance of the ZSM-5 samples in MTG. The MTG experimentation was carried out in a
2

fixed-bed reactor under a set of constant conditions of temperature 375 ºC, pressure 1 MPa
and WHSV 2 h-1. A steady methanol conversion was sustained with increasing the SiO2/Al2O3
ratio, and a progressive decrease in methanol conversion was found over catalysts with low
SiO2/Al2O3 ratios (≤107) after 5 h on stream. Decreasing the SiO2/Al2O3 ratio promoted C1-C4
selectivity and thus decreased gasoline yield. It was also found that decreasing the SiO2/Al2O3
ratio promoted aromatisation reactions and hence higher durene selectivity and more coke
formation, resulting in rapid catalyst deactivation. The sample with SiO2/Al2O3 ratio of 217
showed the highest methanol conversion, gasoline yield, and very low coke formation.
Keywords: catalyst deactivation; coke formation; methanol to gasoline; nanocrystal ZSM-
5; SiO2/Al2O3 ratio
1. Introduction
Methanol to gasoline (MTG) conversion is one of several promising processes to produce
synthetic transport fuel from sources that are independent of the conventional petroleum
resource. MTG is a well-known process first discovered in the 1970s by Exxon Mobil [1, 2],
which uses methanol as feedstock and ZSM-5 zeolite as catalyst to produce gasoline range
hydrocarbons ‒ a sulphur free synthetic liquid fuel with a research octane number (RON) of
90-96, comparable to that of conventional gasoline [3]. Furthermore, the methanol is easy to
transport and store, and more importantly, can be made at various scales from various sources
such as natural gas, coal and/or biomass [4-7], or even the utilisation of greenhouse gas CO2
[8], which makes the MTG process attractive.
The most commonly used catalyst for the MTG process is ZSM-5 zeolite. This well-known
catalyst has a porous aluminosilicate framework composed by tetrahedral T atoms (T = Si, Al,
etc.) connected by oxygen atoms. The framework is negatively charged due to partial
3

substitution of framework Si by Al, and extra-framework cations (usually Na+) are used to
balance the charge of the unit cells. These cations are ion exchangeable, thus protonated
zeolites can be derived by thermally decomposing the ammonium exchanged ones. The
proton ions provide strong acid sites which gives ZSM-5 zeolite the high catalytic activity.
As can be inferred from above, the activity of ZSM-5 zeolite is determined by the strength
and density of its acid sites, more precisely, the number of accessible Brønsted acid sites per
unit cell, which are proportional to the tetrahedrally coordinated Al content in the zeolite
framework [9, 10]. Therefore, the SiO2/Al2O3 ratio of ZSM-5 zeolite plays an important role
in determining its catalytic activity and life span. A ZSM-5 zeolite with high Al content (i.e.
low SiO2/Al2O3 ratio) generally has high amount of acid sites, and thus high activity.
However, the high Al content in a zeolite will also lead to high coke formation rate that
results in fast catalyst deactivation [11-13]. This trade-off between the acidity and life span of
the ZSM-5 zeolite implies a suitable SiO2/Al2O3 ratio is pivotal to ensure the zeolite to have
high activity as well as a long life span. A suitable range of SiO2/Al2O3 ratio was reported to
be 30-70 for micronsized zeolitic catalysts in MTG conversion [14, 15].
Apart from the intrinsic effect of SiO2/Al2O3 ratio, crystal size also impacts greatly on the
performance of ZSM-5 catalyst. The ZSM-5 zeolite has three-dimensional microporous
channels with 10-membered ring openings with an accessible diameter of 0.55 nm [16-18].
These micropore channels cause diffusion restrictions that make the catalyst prone to coking
[19, 20]. The formation of coke subsequently deactivates the catalyst by blocking pore
entrances and covering active sites inside the channels [21-23]. This has always been a major
concern in industrial applications.
A potentially simple way to overcome diffusion limitations within ZSM-5 zeolite is to reduce
its crystal size to the nanoscale, namely, via synthesis of nanocrystal ZSM-5. A reduction in
4

the crystal size provides ZSM-5 zeolite with increased external surface area, providing more
active sites accessible to reactants. More importantly, it greatly shortens the diffusion paths in
ZSM-5 nanocrystals, leading to significantly enhanced mass transfer for both reactant and
product species to enter and exit the micropores. Nanocrystal ZSM-5 zeolites have been
reported to outperform its micronsized counterparts in various catalytic reactions [24-26]. For
example, Rownaghi and co-workers [25] reported a nanocrystal ZSM-5 sample showing long-
term catalytic stability, with sustained complete methanol conversion within 10 h on stream,
while the methanol conversion of a micronsized counterpart was only ~65% initially and
dropped to 50% during the same reaction time.
The question then arises if nanocrystal ZSM-5 zeolite is used for MTG reaction, will the
suitable SiO2/Al2O3 ratio be different from that of the conventional micronsized ZSM-5? To
address this question, nanocrystal ZSM-5 of different SiO2/Al2O3 ratio but the same crystal
size has to be used. However, synthesis of such material is challenging due to the interplay
between crystal size and SiO2/Al2O3 ratio and, it has not been reported before to the best of
our knowledge. In this work, we prepared a series of nanocrystal ZSM-5 zeolites with almost
identical crystal size but different SiO2/Al2O3 ratios of 23, 47, 107, 217 and 411. Furthermore,
we used these materials in MTG reaction to systematically investigate the intrinsic effect of
SiO2/Al2O3 ratio on the catalyst performance independent of crystal size. X-ray diffraction
(XRD), inductively coupled plasma atomic emission spectroscopy (ICP-AES), scanning
electron microscopy (SEM), transmission electron microscopy (TEM), solid-state magic
angle spinning nuclear magnetic resonance (MAS NMR), ammonia temperature programmed
desorption (NH3‒TPD) and nitrogen physisorption were employed to characterise the
structural properties of zeolites. The catalytic performance of the zeolite samples in the MTG
conversion was evaluated using a fixed-bed reactor, and coke formation on the used catalysts
was analysed using a thermogravimetric analyser (TGA).
5

2 Experimental
2.1 Reagents
All reagents were purchased from Sigma-Aldrich and used as supplied. These include sodium
aluminate (NaAlO2, anhydrous, analytic reagent) as the alumina source, tetraethyl
orthosilicate (TEOS, ≥99.0%) as the silica source, and tetrapropylammonium hydroxide
(TPAOH, 1.0 M aqueous solution) as the structure directing agent for the synthesis of
nanocrystal ZSM-5 zeolites.
2.2 Synthesis of nanocrystal ZSM-5 zeolites
A series of nanocrystal ZSM-5 zeolites with SiO2/Al2O3 ratios of 23, 47, 107, 217 and 411
(as determined by ICP-AES) were prepared using a hydrothermal synthesis method, and
denoted as NZ_23, NZ_47, NZ_107, NZ_217 and NZ_411, respectively. The samples were
synthesised from a sol-gel solution with a molar composition of x Na2O: x Al2O3: 100 SiO2:
25 TPAOH: 1157 H2O, where x is 4, 2, 1, 0.5 or 0.25, as calculated from a predetermined
amount of NaAlO2 used in the synthesis. The amount of NaAlO2 used was 0.2 g, 0.1 g, 0.05 g,
0.025 g and 0.0125 g, respectively. For each sample, the predetermined amount of NaAlO2
and 8 g of 1.0 M aqueous TPAOH solution were added to a 100 mL Schott-Duran bottle and
stirred at room temperature for 0.5 h before 6.4 g of TEOS was added drop-wise. The
resulting gel was continuously stirred at ambient temperature for 0.5 h, then in an oil bath at
80 °C for 1 h, and finally back to ambient temperature for another 20 h. The prepared gel was
transferred to a Teflon-lined autoclave to allow for the hydrothermal reaction at 180 °C for 48
h. After cooling, the sample was collected by centrifugation and washed thoroughly using
deionised water three times.
All samples synthesised were dried at 100 °C for 20 h before calcination at 550 °C in air for 5
h to remove the structure directing agent. Ion exchange was then performed to replace Na+
+
with NH4 by mixing 1.0 g of sample in 10 mL of 1.0 M NH4NO3 solution at 50 °C for 24 h.
6

The procedure was repeated twice to ensure complete ion exchange. The ion-exchanged
samples were again collected by centrifugation and washed in deionised water three times,
dried at 100 °C for 20 h and calcined at 500 °C for 5 h to obtain the final H+ type ZSM-5
samples.
2.3 Characterisation
Powder XRD was used to verify the structure of the synthesized ZSM-5 zeolite. The XRD
patterns were obtained at ambient temperature using a PANalytical Empyrean X-ray
diffractometer (Cu Kα, λ=1.54439 Å) at 40 kV and 40 mA, with a scan rate of 2.2° min-1 over
a 2θ range of 5-50°. ICP-AES was used to determine the total SiO2/Al2O3 ratio of the
synthesised samples using a Thermo Scientific iCAP 6500 Duo ICP-AES. SEM and TEM
images were used to examine the morphology and the sample crystal sizes, with a Zeiss 1555
VP-FESEM instrument (15 kV) and a JEOL 2100 TEM (120 kV), respectively. Solid state
MAS NMR was conducted on a Varian 400 wide-bore NMR spectrometer equipped with two
radio frequency channels to investigate the coordination of Al in the framework for all
27
samples. Al MAS NMR spectra were recorded at 104.2 MHz at a spinning frequency of 5
3+
kHz and 2.05 s intervals between successive accumulations. [Al(H2O)6 ] was used as the
standard reference for 27Al.
In-situ ammonia temperature programmed desorption (NH3‒TPD) was also carried using a
Quantachrome ChemBET PULSAR TPR/TPD. Specifically, under high purity helium at a
flow rate of 15 mL min-1, approximately 0.1 g of each sample was heated from room
temperature at a heating rate of 10 °C min-1 to 500 °C and maintained at this temperature for
60 min to remove any absorbed species. The sample was then cooled down to 120 °C and
flushed with a mixture of 5% NH3 (v/v) in helium at a flow rate of 15 mL min-1 for 40 min.
The gas atmosphere was then changed to helium at a flow rate of 15 mL min-1 to remove the
7

physisorbed NH3. After 40 min, the sample was heated again to 650 °C at a heating rate of
10 °C min-1 to remove NH3 absorbed on the acid sites by desorption. The desorbed NH3 was
recorded every 5 s using a thermal conductivity detector (TCD).
The nitrogen adsorption behaviours of the samples were carried out at -196 °C with a
Micromeritics Tristar II instrument. Typically, 0.2 g of each sample was degassed under
vacuum at 200 °C for 12 h prior to the measurement. The Brunauer-Emmett-Teller (BET)
method was used to determine the specific surface area using the adsorption data collected in
the pressure range of p/p0=0.05-0.2 [27, 28]. The t-plot method was adopted to calculate the
external surface area, micropore area and micropore volume using the adsorption data in the
range of p/p0=0.2-0.5 [28]. The total pore volume was estimated from the amount of nitrogen
adsorbed at p/p0=0.995.
2.4 Evaluation of Catalytic Performance in MTG
The catalytic performance of the synthesised nanocrystal ZSM-5 catalysts was evaluated in
MTG conversion using a high pressure fixed-bed reactor, a schematic of the reactor system is
shown in the Supporting Information Figure S1. The as-synthesised ZSM-5 catalyst was
pelletised, crushed and sieved into pellets of ~1 mm in diameter. In a typical run, 0.24 g of the
catalyst pellets were loaded into the middle section of the reactor – a quartz tube equipped
with a stainless steel tube jacket. The reactor was placed in a 400 mm vertical furnace and
aligned to ensure the catalyst sit at the centre of the isothermal zone. Prior to the start of the
MTG reaction, the catalyst was activated at 375 °C for 2 h under high purity nitrogen purge at
a flow rate of 44 mL min-1(STP). Then, liquid methanol (Sigma-Aldrich, HPLC grade, 99.9%)
at a flow rate of 0.01 mL min-1 was vaporised at 375 °C in a preheater, and then introduced
into the reactor using high purity nitrogen as carrier gas at a flow rate of 44 mL min-1, giving
a 10% (v/v) methanol in nitrogen mixture. The reactions were allowed to continue for 24 h
8

under a set of constant conditions of temperature 375 °C, pressure 1 MPa and weight hourly
speed velocity (WHSV) 2 h-1.
A split sample of the reactor exit stream was directed to an Agilent 7890A gas chromatograph
(GC) and analysed online at a 1 h interval for its composition. To avoid condensation, while
the reactor’s entire downstream tubes were maintained at 180 °C. The GC was equipped with
two capillary columns, VF-1 ms and PoraBOND Q, connected to a thermal conductivity
detector (TCD), and three additional capillary columns, CP Ms 5A, CP-Sil 8 CB and
PoraBOND Q, connected to a flame ionisation detector (FID). The GC results allowed the
calculation of methanol conversion and product selectivity [15, 29, 30].
High catalytic performance should be indicated by high methanol conversion, high gasoline
yield and low coke formation. The following naming conventions were used when describing
the reaction products: ‘C1-C4’ for the sum of gaseous hydrocarbons with carbon numbers
between 1 and 4, namely, C1-C4 alkanes and C2=C4 alkenes, ‘C5+ aliphatics’ for the sum of
liquid nonaromatics, ‘aromatics’ for the sum of all aromatic products including durene.. The
liquid products, namely, all products except C1-C4 were considered as gasoline. As C1-C4
cannot be used as gasoline and durene affects negatively on gasoline quality, in this study,
more attention was paid to the fraction of C1-C4 and durene when discussing the product
selectivity.
2.5 Coke Analysis
Coking in zeolite catalysed MTG is inevitable. In an effort to peek into the effect of the
SiO2/Al2O3 ratio, thus the acidity, of the ZSM-5 zeolite on coke formation and characteristics,
TGA analysis was carried out using a SDT600 TGA (TA Instrument) to quantify the coke
formed on each spent ZSM-5 zeolite sample after MTG reaction. The TGA measurements
were performed with a temperature-programmed heating profile, and were first conducted in
9

an inert nitrogen atmosphere to vaporise absorbed organic volatiles then followed by air for
the coke combustion to quantify coke content in the used catalyst. Typically, under inert
nitrogen atmosphere at a flow rate of 100 mL min-1, about 10 mg of a used sample was placed
in an alumina crucible. The sample was then heated to 105 °C at a heating rate of 20 °C min-1
and maintained at this temperature for 15 min to remove any surface moisture. The sample
was then further heated to 550 °C at a heating rate of 10 °C min-1 and kept at this temperature
for 15 min before switching the gas from nitrogen to air, again at a flow rate of 100 mL min-1.
The sample was held under air for 40 min to allow for the complete combustion of the coke.
The mass loss above 105 °C in nitrogen was ascribed to loss of absorbed volatile residues on
the catalyst. The mass loss after switching the gas atmosphere to air was considered to be
associated with the combustion of the hard coke formed during the MTG reaction.
3. Results and Discussion
3.1. Strategy of Synthesising nanocrystal ZSM-5 samples with various SiO2/Al2O3 ratios
The synthesis of nanocrystal ZSM-5 samples has been extensively studied and many synthesis
methods have been developed. These studies have shown that the crystal size of resultant
ZSM-5 zeolite products depends greatly on the precursor composition and synthesis
conditions, including factors such as silica and alumina sources, temperature, alkalinity and
aging duration of the gel, hydrothermal treatment conditions and SiO2/Al2O3 ratio [31, 32].
Changing SiO2/Al2O3 ratio was observed to have great impact on the crystal size of ZSM-5
zeolite. Generally, an increase in the SiO2/Al2O3 ratio leads to a decrease in the crystal size
under otherwise identical conditions [33-35], therefore making it difficult to prepare ZSM-5
zeolites of various SiO2/Al2O3 ratios and similar crystal size from the same method. On the
other hand, it is well known that crystal size of the ZSM-5 zeolite can have significant impact
on its catalytic performance, particularly on the catalyst deactivation [25, 36]. Thus it is
10

challenging to investigate the effect of SiO2/Al2O3 ratio alone on the catalytic performance of
the zeolite and in the past such study has been often performed by ignoring the effect of
crystal size [26, 37, 38].
In this study, a simple method was developed to synthesise a series of nanocrystal ZSM-5
samples with various SiO2/Al2O3 ratios but almost identical crystal sizes. Changing the
SiO2/Al2O3 ratios without affecting the crystal sizes of the samples was achieved by varying
the amount of the alumina source (NaAlO2) and using excessive amount of TPAOH , while
keeping the amount of other precursor components unchanged. The amount of TPAOH used
in the synthesis was relatively high, thus ensuring there was sufficient OH- in the precursor
solutions for all samples. This overrode the effect of varying Al contents on the nuclei
formation, and the alkalinity of the gel solutions was almost unaffected. Consequently,
nucleation and growth rate of framework structure was not significantly affected during the
synthesis when SiO2/Al2O3 ratio was varied, enabling the production of ZSM-5 samples with
almost identical crystal sizes but of various SiO2/Al2O3 ratios.
3.2. Characteristics of nanocrystal ZSM-5 samples with various SiO2/Al2O3 ratios
The powder XRD patterns of five synthesised samples are shown in Figure 1. Evidently, these
diffraction patterns are almost identical and consistent with that of typical ZSM-5 zeolite
reported in the literature [39], confirming that highly crystalline ZSM-5 were produced at all
the five different SiO2/Al2O3 ratios.
Representative SEM and TEM images of the five samples are shown in Figure 2. All samples
consisted of crystals of almost identical size of ~100 nm with a narrow size distribution. The
lattice fringes observed along different axis under high resolution TEM further confirmed that
each nanocrystal seen in Figure 3 was a high crystalline single crystal with micropore
structures of typical ZSM-5 zeolite.
11

The chemical environment of the framework aluminium in these five samples was verified
27
through Al MAS NMR spectra. As shown in Figure 3, the spectrum was dominated with a
main resonance peak at ~55.5 ppm for all samples and a minor resonance at ~0 ppm for the
samples with SiO2/Al2O3 ratio ≤ 107. According to the literature [40], the former one can be
attributed to the tetrahedrally coordinated Al within the framework structure, whereas the
latter one is associated with the octahedrally coordinated Al on the extra framework.
Therefore, the majority of the Al in the framework of the synthesised samples was
tetrahedrally coordinated, with only a very minor proportion on the extra framework for the
samples with low SiO2/Al2O3 ratio. The resonance at ~ 0 ppm dropped substantially with
increasing SiO2/Al2O3 ratio suggesting ZSM-5 zeolites with high Al content (thus low
SiO2/Al2O3 ratio) are more prone to develop extra framework Al [10, 41].
The acid site distributions of the five samples obtained via NH3‒TPD are shown in Figure 4.
As can be seen, all samples exhibited two well resolved desorption peaks typical of ZSM-5
zeolite, with the low temperature peaks at 150-320 °C corresponding to weak acid sites and
the high temperature peaks at 320-550 °C corresponding to strong acid sites, respectively [42-
44]. The intensity of a desorption peak representing the amount of acid sites decreased with
increasing SiO2/Al2O3 ratio, indicating the amount of acid sites was indeed proportional to the
Al content in the catalyst. Clearly, NZ_23 had the highest Al content and thus the highest
amount of acid sites, in good agreement with the literature [45, 46]. In this study, a low
SiO2/Al2O3 ratio was indicative of a high acid site density, and vice versa.
The nitrogen physisorption isotherms of all five samples presented in Figure 5 show all
samples exhibited very similar adsorption/desorption behaviour. The steep uptake observed
for all isotherms in low relative pressure region (p/p0 < 0.01) reflects a high microporosity in
the samples, indicating an intact microporous framework and high crystallinity, consistent
with the results of XRD and TEM. The sudden rise in the adsorption capacity at high relative
12

pressures (p/p0>0.9) along with the hysteresis loops suggests the existence of mesopores in
the samples. These mesopores were created by intercrystal spaces and are commonly
observed in zeolites with nanosized crystals [25, 47]. The adsorption properties of the samples
extracted from the isotherms are summarised in Table 1. The BET surface areas, micropore
areas and volumes were found to increase slightly (by ~5%) with increasing SiO2/Al2O3 ratio,
consistent with existing understanding on the aluminium content effect [48, 49]. The total
pore volume for all the samples was determined to be ~0.35 cm3g-1, in good agreement with
that of typical nanocrystal ZSM-5 zeolite with high crystallinity [47].
It is clear that for the five ZSM-5 samples produced, although their SiO2/Al2O3 ratios varied
significantly from 23 to 411, their crystal sizes and structural properties remained almost
identical. This thus provided a unique opportunity to independently investigate the effect of
SiO2/Al2O3 ratio on the catalytic performance of these samples in MTG conversion which is
discussed in the following.
3.3 Catalytic performance of nanocrystal ZSM-5 zeolite catalysts
3.3.1 Effect of SiO2/Al2O3 ratio on methanol conversion and product distribution
The effect of SiO2/Al2O3 ratio on methanol conversion versus time on stream is shown in
Figure 6. With the exception of NZ_411, all other ZSM-5 samples showed complete methanol
conversion (100%) during the first 5 h reaction period. After that, a decrease in methanol
conversion was observed in the samples with low SiO2/Al2O3 ratios (NZ_107, NZ_47,
NZ_23), and this reduction became increasingly pronounced with decreasing SiO2/Al2O3 ratio,
indicating the samples having higher aluminium content thus higher acidity were deactivated
more rapidly. In addition, the extra framework Al observed for those three samples might also
contribute to the fast deactivation [41, 50, 51]. The methanol conversion remained complete
and steady for NZ_217 over the 24h time period studied. However, further increasing
SiO2/Al2O3 ratio to 411 resulted in incomplete methanol conversion (~94%) even from the
13

very beginning of the reaction. This was most likely due to the reduced number of active sites
in NZ_411. As the number of active sites is proportional to the aluminium content in the
ZSM-5 zeolite, catalyst with highly increased SiO2/Al2O3 ratio of 411 led to insufficiency in
active sites and hence incomplete methanol conversion under the reaction conditions enforced.
Note that methanol conversion for NZ_217 and NZ_411 remained steady over the reaction
time studied, suggesting a sustained catalyst activity on ZSM-5 with high SiO2/Al2O3 ratios
(due to less coking as elaborated in next section).
The product distributions and gasoline yield of the catalysts with different SiO2/Al2O3 ratios
are shown in Figure 7. Aromatics decreased with increasing SiO2/Al2O3 ratio. The zeolite
catalyst with the highest aluminium content (NZ_23) had the highest aromatics selectivity,
suggesting a high amount of acid sites clearly promoted the aromatisation reaction. An
important effect of increasing SiO2/Al2O3 ratio was a significant decrease in the durene
selectivity, from 10.8% to 3.9% as SiO2/Al2O3 ratio increased from 23 to 411. It was also
evident that increasing SiO2/Al2O3 ratio facilitated the formation of C5+ aliphatics,
consequently contributed to an increase in the gasoline yield. The gaseous products C1-C4 that
cannot be considered as gasoline, was found to decrease with increasing SiO2/Al2O3 ratio.
The term of C1-C4 in this study includes C1-C4 alkanes (paraffins) and C2=C4 alkenes (olefins).
The latter, particularly ethylene (C2H4) and propylene (C3H6), is believed to be the primary
products for methanol conversion to hydrocarbons (MTH) catalysed by zeolitic catalysts [52,
53]. Under the reaction conditions for MTG, as demonstrated in the reaction path shown in
Figure 8, methanol (CH3OH) is first catalytically dehydrated to form an equilibrium mixture
of dimethyl ether (DME), the mixture undergoes further reactions to form the primary
products. These primary products subsequently undergo oligomerisation, cyclisation,
aromatisation, aromatic methylation, hydrogen transfer and aromatic dealkylation to form the
final product.
14

The reactions involved in MTH are catalysed by the acid sites arose from the bridging
hydroxyl group (Si‒OH‒Al), shown in Figure 8. Apparently, high amount of acid sites would
direct the reactions towards to formation of high hydrocarbons, in agreement with the high
aromatic selectivity for NZ_23 discussed above. It also can be inferred that C2=C4 alkenes,
being the primary products, should have low selectivity at low SiO2/Al2O3 ratio. This is
revealed in the selectivity of C2=C4 alkenes as a function of SiO2/Al2O3 ratio displayed in
Figure 7b, increasing SiO2/Al2O3 ratio led to an increase in C2=C4 alkenes. Therefore, a high
SiO2/Al2O3 ratio is preferred for the production of light olefins, and is widely reported in the
literature for methanol/ethanol to olefins (M/ETO) [37, 54-57].
In contrary, C1-C4 alkanes were found to decrease with increasing SiO2/Al2O3 ratio, shown in
Figure 7b. This can be understood by the unique shape selectivity of ZSM-5 zeolite [58]. It is
well known that ZSM-5 zeolite has an intermediate pore size of ~ 0.55nm, only small and
intermediate organic molecules can traverse through these pores [16]. The growth of the
molecules formed during the reaction is regulated by the pores, leading to the occurrence of
aromatic dealkylation and thus the formation of light hydrocarbons (C1-C4). C2=C4 alkenes
can further react to form other hydrocarbons; however, C1-C4 alkanes are saturated and thus
relatively stable, remaining as effluent gas. Low SiO2/Al2O3 ratio promoted the further
reaction of C2=C4 alkenes and also aromatic deakylation, resulted in C1-C4 alkanes being the
predominant light hydrocarbons at low SiO2/Al2O3 ratio and the overall C1-C4 selectivity
decreasing with increasing SiO2/Al2O3 ratio.
Figure 9 shows the product distributions as a function of time on stream for the five catalysts.
All the catalysts showed a decrease in the selectivity of C1-C4, aromatics and durene, and an
increase in C5+ aliphatics, reflecting a similar phenomenon as the effect of increasing
SiO2/Al2O3 ratio on product distribution. This can be understood by the active sites in the
catalysts being progressively covered by coke formed during the reaction, and consequently,
15

the effective amount of active sites decreased with time on stream. Furthermore, the changes
in product distributions with time on stream were found to be more stable for NZ_217 and
NZ_411, and more pronounced with decreasing SiO2/Al2O3 ratio, indicating a faster coverage
of active sites at lower SiO2/Al2O3 ratios, in line with the more rapid catalyst deactivation
with decreasing SiO2/Al2O3 ratio aforementioned.
Overall, NZ_217 had the highest gasoline yield through the course of the test, as a
consequence of its complete and steady methanol conversion and relatively low C1-C4
selectivity. It is worth noting that NZ_217 had an acceptable durene selectivity of 4.6%, as
durene concentration in the gasoline higher than 5% will cause rough engine operation [3].
3.3.2 Effect of SiO2/Al2O3 ratio on coke formation
Deactivation of catalysts is mainly caused by the formation of carbonaceous residues and
subsequent deposition on the catalyst surfaces and/or pores, thus covering the active sites.
These carbonaceous residues are commonly known as coke, constituting polycyclic aromatic
hydrocarbons [22]. During the catalytic reaction, coke is gradually formed and retained in the
catalyst. A small proportion of volatile compounds are also trapped in the catalyst thus not
considered as “real coke” but the precursors [59]. Therefore, specifically quantifying the “real
coke” was studied by thermal gravimetric analysis conducted first in an inert nitrogen
atmosphere below 550 °C to remove volatile molecules and then followed by in air for the
combustion of the “real coke”.
The weight loss curves of the spent catalysts are shown in Figure 10. It can be found the
amount of coke on the spent catalysts was 3.3, 3.2, 1.7, 0.4 and 0.1 wt% for NZ_23, NZ_47,
NZ_107, NZ_217 and NZ_411, respectively, after subtracting the trapped volatile
components. It is also noticed that the proportion of volatile compounds relative to the total
removable organics/cokes was more significant for catalysts with higher SiO2/Al2O3 ratios,
16

e.g. around 80% of the removals in the spent NZ_411 sample was volatile compounds
whereas only about 25% in NZ_23. Evidently, catalysts with lower SiO2/Al2O3 ratios were
more prone to coke formation, suggesting samples with high aluminium content would
deactivate more rapidly, which was clearly reflected by the time-dependent methanol
conversion results shown in Figure 6. In addition, typical SEM and TEM images (Figure S2)
of the spent catalysts revealed virtually identical morphologies as their pristine counterparts,
suggesting coking was the dominant cause of catalyst deactivation.NZ_217 showed higher
sustained activity towards methanol conversion, desirable product selectivity and also the
highest gasoline yield. It was also expected that NZ_217 would have longer life span due to
its relatively low coke formation. Collectively, the analysis of the five samples showed the
one with relatively high SiO2/Al2O3 ratio (217) had the best overall catalytic performance. In
comparison, previous studies reported a preferred SiO2/Al2O3 ratio of 30-70 for zeolitic
catalysts in MTG conversion [14, 15]. This difference is actually a reflection of the effect of
crystal size on catalytic performance of ZSM-5 zeolite. The earlier studies were based on
conventional ZSM-5 zeolites that had crystal size in the micrometre range, in which a certain
proportion of active sites were inaccessible to reactants due to its long diffusion path. Thus
high Al content (low SiO2/Al2O3 ratio) was required for micron-sized ZSM-5 zeolite to
provide sufficient accessible active sites to enable good methanol conversion. In contrast, by
reducing the crystal size of the zeolite from micron to nanometre scale, the number of
accessible active sites is greatly increased leading to higher activity of nanocrystal ZSM-5
zeolite than its conventional counterpart. In other words, nanocrystal ZSM-5 zeolite with the
same aluminium content as a micron-sized counterpart would show better catalytic
performance [36, 60].
4. Conclusions
17

An unbiased study of the effect of SiO2/Al2O3 ratio on the catalytic performance of
nanocrystal ZSM-5 in the MTG conversion at a high pressure (1 MPa) was systematically
conducted. Five nanocrystal ZSM-5 samples with different SiO2/Al2O3 ratios ranging from 23
to 411 but of similar crystal sizes and structural properties were successfully synthesised. This
allowed for the investigation into the intrinsic effect of SiO2/Al2O3 ratio on catalytic
performance of ZSM-5 zeolite in MTG.
The catalytic performance of the synthesised samples in MTG reaction was examined in a fix-
bed reactor. It was shown that decreasing SiO2/Al2O3 ratio facilitated higher methanol
conversion whereas the catalyst activity was better sustained with increasing SiO2/Al2O3 ratio.
C2=C4 alkenes decreased and C1-C4 alkanes increased with decreasing SiO2/Al2O3 ratio,
leading to an increase in the overall C1-C4 selectivity and thus low gasoline yield. Decreasing
SiO2/Al2O3 ratio also promoted aromatisation reaction, thus high durene selectivity and coke
formation, leading to rapid catalyst deactivation. The nanocrystal catalyst with a moderate
SiO2/Al2O3 ratio of 217 showed a sustained 100% methanol conversion, high gasoline yield
and low coke formation under the reaction conditions studied. By using nanocrystal ZSM-5
catalyst in MTG at high pressures, the preferred SiO2/Al2O3 ratio is much higher than that for
the microcrystal counterparts.
Acknowledgment
Partial financial support has been received for this research from the Australian Research
Council under the ARC Linkage Projects Scheme (LP100200136) and Chevron Energy
Technology Pty Ltd. The authors also acknowledge the facilities, and the scientific and
technical assistance of the Australian Microscopy & Microanalysis Research Facility at the
Centre for Microscopy, Characterisation & Analysis, The University of Western Australia, a
facility funded by the University, State and Commonwealth Governments. Zhijian Wan also
acknowledges the ARC for providing an APAI scholarship.
18

References
[1] C.D. Chang, A.J. Silvestri, J. Catal. 47 (1977) 249-259.
[2] N.R.C.F. Machado, V. Calsavara, N.G.C. Astrath, A. Medina, M.L. Baesso, Appl. Catal.
A. 311 (2006) 193-198.
[3] C.D. Chang, J.C. Kuo, W.H. Lang, S.M. Jacob, J.J. Wise, A.J. Silvestri, Ind. Eng. Chem.
Process Des. Dev. 17 (1978) 255-260. 17 (1978) 255-260.
[4] G.A. Olah, Angew. Chem. Int. Ed. 44 (2005) 2636-2639.
[5] Z.X. Di, C. Yang, X.J. Jiao, J.Q. Li, J.H. Wu, D.K. Zhang, Fuel. 104 (2013) 878-881.
[6] M. Inaba, K. Murata, I. Takahara, K.i. Inoue, J. Chem. Technol. Biot. 86 (2011) 95-104.
[7] A.G. Gayubo, A. Alonso, B. Valle, A.T. Aguayo, J. Bilbao, Appl. Catal. B. 97 (2010)
299-306.
[8] G.A. Olah, A. Goeppert, G.K.S. Prakash, J. Org. Chem. 74 (2009) 487-498.
[9] A. Corma, J. Catal. 216 (2003) 298-312.
[10] A. Primo, H. Garcia, Chem. Soc. Rev. 43 (2014) 7548-7561.
[11] M. Yamamura, K. Chaki, T. Wakatsuki, H. Okado, K. Fujimoto, Zeolites. 14 (1994) 643-
649.
[12] M. Li, Y. Zhou, C. Ju, Y. Fang, Appl. Catal. A. 512 (2016) 1-8.
[13] D.M. Bibby, R.F. Howe, G.D. Mclellan, Appl. Catal. A. 93 (1992) 1-34.
[14] C. Chang, S. Grover, US Patent 4,058,576, 1977.
[15] N.L. Michels, S. Mitchell, J. Perez-Ramirez, Acs. Catal. 4 (2014) 2409-2417.
[16] A.W. Chester, E.G. Derouane, Zeolite characterization and catalysis, Springer, Berlin,
2009.
[17] D.H. Olson, G.T. Kokotailo, S.L. Lawton, W.M. Meier, J. Phys. Chem. 85 (1981) 2238-
2243.
[18] R.C. Wei, H. Yang, D.K. Zhang, Rsc. Adv. 5 (2015) 63765-63776.
19

[19] M. Bjorgen, F. Joensen, M.S. Holm, U. Olsbye, K.P. Lillerud, S. Svelle, Appl. Catal. A.
345 (2008) 43-50.
[20] M. Inaba, K. Murata, M. Saito, I. Takahara, React. Kinet. Catal. Lett. 88 (2006) 135-142.
[21] G.F. Froment, J. Demeyer, E.G. Derouane, J. Catal. 124 (1990) 391-400.
[22] P.L. Benito, A.G. Gayubo, A.T. Aguayo, M. Olazar, J. Bilbao, Ind. Eng. Chem. Res. 35
(1996) 3991-3998.
[23] F. Schmidt, C. Hoffmann, F. Giordanino, S. Bordiga, P. Simon, W. Carrillo-Cabrera, S.
Kaskel, J. Catal. 307 (2013) 238-245.
[24] M. Choi, K. Na, J. Kim, Y. Sakamoto, O. Terasaki, R. Ryoo, Nature. 461 (2009) 246-249.
[25] A.A. Rownaghi, J. Hedlund, Ind. Eng. Chem. Res. 50 (2011) 11872-11878.
[26] T. Tago, H. Konno, M. Sakamoto, Y. Nakasaka, T. Masuda, Appl. Catal. A. 403 (2011)
183-191.
[27] J.C. Groen, L.A.A. Peffer, J. Perez-Ramirez, Micropor. Mesopor. Mat. 60 (2003) 1-17.
[28] S. Lowell, Characterization of porous solids and powders: surface area, pore size and
density, Springer, Berlin, 2004.
[29] Z.J. Wan, W. Wu, W. Chen, H. Yang, D.K. Zhang, Ind. Eng. Chem. Res. 53 (2014)
19471-19478.
[30] K. Kim, R. Ryoo, H.D. Jang, M. Choi, J. Catal. 288 (2012) 115-123.
[31] C.S. Cundy, P.A. Cox, Chem. Rev. 103 (2003) 663-701.
[32] C.S. Cundy, P.A. Cox, Micropor. Mesopor. Mat. 82 (2005) 1-78.
[33] R. Van Grieken, J.L. Sotelo, J.M. Menendez, J.A. Melero, Micropor. Mesopor. Mat. 39
(2000) 135-147.
[34] Y. Wang, L.F. Guo, Y. Ling, Y.M. Liu, X.A.H. Li, H.H. Wu, P. Wu, Appl. Catal. A. 379
(2010) 45-53.
20

[35] Y.Y. Hu, C. Liu, Y.H. Zhang, N. Ren, Y. Tang, Micropor. Mesopor. Mat. 119 (2009)
306-314.
[36] A.A. Rownaghi, F. Rezaei, J. Hedlund, Catal. Commun. 14 (2011) 37-41.
[37] F.L. Bleken, S. Chavan, U. Olsbye, M. Boltz, F. Ocampo, B. Louis, Appl. Catal. A. 447
(2012) 178-185.
[38] D.B. Lukyanov, Zeolites. 12 (1992) 287-291.
[39] M.M. Treacy, J.B. Higgins, R. von Ballmoos, I.Z. Association, S. Commission,
Collection of simulated XRD powder patterns for zeolites, Elsevier, New York, 1996.
[40] O.H. Han, C.S. Kim, S.B. Hong, Angew. Chem. Int. Ed. 41 (2002) 469-472.
[41] K. Barbera, F. Bonino, S. Bordiga, T.V.W. Janssens, P. Beato, J. Catal. 280 (2011) 196-
205.
[42] Z.X. Song, A. Takahashi, I. Nakamura, T. Fujitani, Appl. Catal. A. 384 (2010) 201-205.
[43] M. Inaba, K. Murata, I. Takahara, K. Inoue, Adv. Mater. Sci. Eng. 2012 (2012).
[44] M. Inaba, K. Murata, M. Saito, I. Takahara, Green Chem. 9 (2007) 638-646.
[45] D. Ma, W. Zhang, Y. Shu, X. Liu, Y. Xu, X. Bao, Catal. Lett. 66 (2000) 155-160.
[46] A.S. Al-Dughaither, H. de Lasa, Ind. Eng. Chem. Res. 53 (2014) 15303-15316.
[47] H.B. Zhang, Y.C. Ma, K.S. Song, Y.H. Zhang, Y. Tang, J. Catal. 302 (2013) 115-125.
[48] L. Shirazi, E. Jamshidi, M.R. Ghasemi, Cryst. Res. Technol. 43 (2008) 1300-1306.
[49] A.F.P. Ferreira, M.C. Mittelmeijer-Hazeleger, J.V.D. Bergh, S. Aguado, J.C. Jansen, G.
Rothenberg, A.E. Rodrigues, F. Kapteijn, Micropor. Mesopor. Mat. 170 (2013) 26-35.
[50] Z.X. Qin, L. Lakiss, L. Tosheva, J.P. Gilson, A. Vicente, C. Fernandez, V. Valtchev,
Adv. Funct. Mater. 24 (2014) 257-264.
[51] K.S. Triantafyllidis, A.G. Vlessidis, L. Nalbandian, N.P. Evmiridis, Micropor. Mesopor.
Mat. 47 (2001) 369-388.
[52] S. Ilias, A. Bhan, Acs. Catal. 3 (2013) 18-31.
21

[53] J. Kim, M. Choi, R. Ryoo, J. Catal. 269 (2010) 219-228.
[54] Z.X. Song, A. Takahashi, N. Mimura, T. Fujitani, Catal. Lett. 131 (2009) 364-369.
[55] C.D. Chang, C.T.W. Chu, R.F. Socha, J. Catal. 86 (1984) 289-296.
[56] D. Goto, Y. Harada, Y. Furumoto, A. Takahashi, T. Fujitani, Y. Oumi, M. Sadakane, T.
Sano, Appl. Catal. A. 383 (2010) 89-95.
[57] K. Murata, M. Inaba, I. Takahara, J. Jpn. Pet. Inst. 51 (2008) 234-239.
[58] V. Calsavara, M.L. Baesso, N.R.C. Fernandes-Machado, Fuel. 87 (2008) 1628-1636.
[59] H. Schulz, Catal. Today. 154 (2010) 183-194.
[60] S.K. Saxena, N. Viswanadham, A.a.H. Al-Muhtaseb, J. Ind. Eng. Chem. 20 (2014) 2876-
2882.
22

Figure Captions
Figure 1 Typical XRD patterns of nanocrystal ZSM-5 samples with various SiO2/Al2O3
ratios ranging from 23 – 411
Figure 2 Typical SEM and TEM images of nanocrystal ZSM-5 samples with various
SiO2/Al2O3 ratios ranging from 23 – 411
Figure 3 27Al MAS NMR spectra of nanocrystal ZSM-5 samples with various SiO2/Al2O3
ratios ranging from 23 – 411
Figure 4 NH3‒TPD profiles of nanocrystal ZSM-5 samples with various SiO2/Al2O3 ratios
ranging from 23 – 411
Figure 5 Nitrogen physisorption isotherms of nanocrystal ZSM-5 samples with various
SiO2/Al2O3 ratios ranging from 23 – 411. Solid symbols denote adsorption and
hollow ones for desorption
Figure 6 Effect of SiO2/Al2O3 ratio on methanol conversion of nanocrystal ZSM-5 zeolite
catalysts tested at 375 °C, 1 MPa and WHSV of 2 h-1
Figure 7 Effect of SiO2/Al2O3 ratio on product distribution and gasoline yield of
nanocrystal ZSM-5 zeolite catalysts tested at 375 °C, 1 MPa and WHSV of 2 h-1
Figure 8 Proposed reaction pathways for methanol conversion on ZSM-5 catalyst
Figure 9 Product selectivity as a function of time for the nanocrystal ZSM-5 zeolite
catalysts with various SiO2/Al2O3 ratios tested at 375 °C, 1 MPa and WHSV of 2
h-1
Figure 10 TGA curves of the spent catalysts with different SiO2/Al2O3 ratios tested at 375 °C,
1 MPa and WHSV of 2 h-1
23

Figure 1
NZ_411
NZ_217
NZ_107
NZ_47
NZ_23
5
10
15
20
25
30
35
40
45
50
2
Figure 1 Typical XRD patterns of nanocrystal ZSM-5 samples with various SiO2/Al2O3
ratios ranging from 23 – 411.
24

Figure 2
Figure 2 Typical SEM and TEM images of nanocrystal ZSM-5 samples with various
SiO2/Al2O3 ratios ranging from 23 – 411.
25

Figure 3
NZ_411
NZ_217
NZ_107
NZ_47
NZ_23
120
100
80
60
40
20
0
-20
Chemical Shift (ppm)
Figure 3 27Al MAS NMR spectra of nanocrystal ZSM-5 samples with various SiO2/Al2O3
ratios ranging from 23 – 411.
26

Figure 4
NZ_23
NZ_47
NZ_107
NZ_217
NZ_411
100
200
300
400
500
600
Temperature (C)
Figure 4 NH3‒TPD profiles of nanocrystal ZSM-5 samples with various SiO2/Al2O3 ratios
ranging from 23 – 411.
27

Figure 5
300
NZ_411
NZ_217
150
300
150
300
NZ_107
NZ_47
150
300
150
300
NZ_23
0.4
150
0.0
0.2
0.6
0.8
1.0
Relative Pressure (p/p0)
Figure 5 Nitrogen physisorption isotherms of nanocrystal ZSM-5 samples with various
SiO2/Al2O3 ratios ranging from 23 – 411. Solid symbols denote adsorption and
hollow ones for desorption.
28

Figure 6
100
98
96
94
92
NZ_411
NZ_217
NZ_107
NZ_47
NZ_23
0
5
10
15
20
25
Time on stream (h)
Figure 6 Effect of SiO2/Al2O3 ratio on methanol conversion of nanocrystal ZSM-5 zeolite
catalysts tested at 375 °C, 1 MPa and WHSV of 2 h-1.
29