Characterization and catalytic performance of modified nano-scale ZSM-5 for the acetone-to-olefins reaction

Article abstract of DOI:10.1016/j.apcata.2014.01.031

In the acetone to olefins (ATO) reaction using ZSM-5 zeolites, isobutylene is initially produced via decomposition of an acetone dimer, followed by dimerization of the isobutylene and its cracking to form ethylene and propylene. Nano-scale ZSM-5 zeolite exhibited stable activity during the ATO reaction compared with that of macro-scale ZSM-5. In order to further improve the catalytic stability of the nano-scale ZSM-5, control of the acidity of nano-scale ZSM-5 zeolites through SiO₂ unit formation was examined using phenyl silane and triphenyl silane as new deactivators. This modification led to an increase in the olefins yield and an improvement in the catalyst lifetime. In particular, the regioselective deactivation of acid sites located on the external surface of the zeolite inhibits the formation of aromatics. Moreover, the acidity control within the pores significantly improved the catalyst lifetime. Notably, the modified nano-scale ZSM-5 zeolite exhibited a stable olefins yield above 55 C-mol% for 180 h.

Full text of DOI:10.1016/j.apcata.2014.01.031

Applied Catalysis A: General 475 (2014) 127–133
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Characterization and catalytic performance of modified nano-scale
ZSM-5 for the acetone-to-olefins reaction
Hiroki Konno, Teruoki Tago∗, Yuta Nakasaka, Gaku Watanabe, Takao Masuda
Division of Chemical Process Engineering, Faculty of Engineering, Hokkaido University, N13 W8, Kita-ku, Sapporo, Hokkaido 060-8628, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
In the acetone to olefins (ATO) reaction using ZSM-5 zeolites, isobutylene is initially produced via decom-
position of an acetone dimer, followed by dimerization of the isobutylene and its cracking to form ethylene
and propylene. Nano-scale ZSM-5 zeolite exhibited stable activity during the ATO reaction compared with
that of macro-scale ZSM-5. In order to further improve the catalytic stability of the nano-scale ZSM-5, con-
trol of the acidity of nano-scale ZSM-5 zeolites through SiO2 unit formation was examined using phenyl
silane and triphenyl silane as new deactivators. This modification led to an increase in the olefins yield
and an improvement in the catalyst lifetime. In particular, the regioselective deactivation of acid sites
located on the external surface of the zeolite inhibits the formation of aromatics. Moreover, the acidity
control within the pores significantly improved the catalyst lifetime. Notably, the modified nano-scale
ZSM-5 zeolite exhibited a stable olefins yield above 55 C-mol% for 180 h.
Received 22 November 2013
Received in revised form 10 January 2014
Accepted 15 January 2014
Available online 20 January 2014
Keywords:
Acidity modification
Nano-zeolite
ZSM-5
Acetone conversion
Light olefins
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
dimer (diacetone alcohol) and trimer (isophorone), followed by
production of propylene, ethylene, and aromatics [21–27]. These
Light olefins are important basic raw materials for the petro-
chemical industry, and demand for such products is increasing
every year [1–3]. Ethylene, propylene, and isobutylene are key
building blocks for the production of chemical products, such as
polyethylene (LDPE, HDPE), polypropylene (PP), polyvinyl chlo-
ride (PVC), polyacrylonitrile (PAN), ethylene oxide, propylene
oxide, methyl methacrylate (MMA), ethyl tertiary-butyl ether
(ETBE), and so on. These light olefins are mainly produced by
steam cracking of naphtha. However, because this process con-
sumes more than 30% of the total amount of energy required
for petrochemical refinement, new efficient processes for light
olefins production are required. Promising alternative processes
for light olefins production include naphtha catalytic cracking
(NCC) [4–11], methanol-to-olefins (MTO) [12–14], ethanol-to-
olefins (ETO) [15–17], and propane dehydrogenation (PDH)
[18,19].
Because a large amount of acetone is obtained as a by-product
in the cumene process [20], we investigated the possibility of
light olefins synthesis from acetone (acetone-to-olefins, ATO) over
zeolite catalysts, and reported that a high light olefins yield
was achieved using ZSM-5 zeolite [21] and an alkali metal ion-
exchanged Beta zeolite [22]. In this reaction, isobutylene was
first produced from acetone via the formation of an acetone
intermediates are involved in a series of reactions that result in the
generation of coke as the terminal product.
Previously, we reported that the use of ZSM-5 zeolite was
effective for the suppression of aromatics formation because
the consecutive reactions are mainly routed via diacetone
alcohol formation/decomposition, rather than isophorone forma-
tion/decomposition, which is an aromatics precursor, due to the
spatial limitations of ZSM-5 zeolite with its 10-membered rings
[21]. However, the undesirable excessive reactions that produce
aromatics and coke due to light olefins consumption do readily
occur on acid sites near the internal cross-sectional spaces and on
the external surfaces of the zeolite crystals. Therefore, a method is
desired for the regioselective deactivation and acidity control of the
acid sites of zeolite crystals in order to prevent these undesirable
excessive reactions.
We previously developed a method for the modification of the
acidity of zeolites using organic silane compounds (catalytic crack-
ing of silane, or the CCS method) [28–30]. In the CCS method, SiO2
units are formed on the acid sites where the silane compounds are
chemically adsorbed, thus resulting in modification of the acidity
of the zeolite. Because the molecular diameter of the silane com-
pounds depends on the types of organic groups bound to the Si
atom, it was anticipated that deactivation of the acid sites located
on the external surface of the zeolite crystals could be achieved by
utilizing the molecular sieving effect of the zeolites. Furthermore,
when silane compounds those are comparable to or smaller in size
than the size of the zeolite pores, SiO2 formation may occur on the
∗
Corresponding author. Tel.: +81 117066551; fax: +81 117066552.
E-mail address: tago@eng.hokudai.ac.jp (T. Tago).
0926-860X/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2014.01.031

128
H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
acid sites located at the cross-sectional spaces within the crystals,
leading to further modification of the acidity [30].
PS treatment, it is expected that SiO2 units formed on the acid sites
located both on the internal and external surfaces of the crystal,
leading to a decrease in acidic strength of the zeolite.
Herein, we report the results obtained when triphenyl silane
(TPS) and phenyl silane (PS) were used as new deactivators. The
modified nano-scale ZSM-5 zeolites prepared using TPS and PS
were extensively characterized, and their catalytic performance for
the ATO reaction was investigated. The type of Si-compounds used
for modification of the acidity was observed to influence the prod-
uct selectivity. Notably, the PS-treated, nano-scale ZSM-5 showed
both high selectivity for light olefins and stable activity during the
ATO reaction. The relationship between the acidity modification
and the catalytic performance is also discussed.
2.3. Characterization
The morphology and crystallinity of the obtained samples were
analyzed using field emission scanning electron microscopy (FE-
SEM; JSM-6500F, JEOL Co. Ltd.) and X-ray diffraction analysis (XRD;
JDX-8020, JEOL Co. Ltd.), respectively. The surface areas of the
obtained samples were calculated using the BET-method based on
N2 adsorption isotherms (Belsorp mini, BEL JAPAN Co. Ltd.). The
Si/Al ratios of the samples were determined based on X-ray flu-
orescence measurements (XRF; Supermini, Rigaku Co. Ltd.), and
the acidity of the obtained samples was evaluated using the ac-
NH3–TPD method [33]. In the TPD experiment, the carrier gas was
1.0% NH3 (balance He), the heating rate was 5 K min−1, and the tem-
perature range was 373–823 K. The desorption of NH3 molecules
from the acid sites of the zeolite was measured under complete
adsorption equilibrium conditions (1.0% NH3–He atmosphere).
Pyridine adsorption on the obtained samples was observed using
a diffuse reflectance infrared Fourier transform (DRIFT) spectrom-
eter equipped with a mercury cadmium telluride (MCT) detector
(FT/IR-4100, JASCO Co. Ltd.). A total of 200 scans were averaged
for each spectrum. Pre-treatment was conducted in vacuo at 723 K
for 12 h, and then pyridine was introduced and adsorbed onto the
sample at 373 K for 2 h. The physically adsorbed pyridine was then
removed in flowing N2 at 373 K for 0.5 h, and the remaining species
were subsequently measured via Fourier transform infrared (FT-IR)
analysis at 373 K.
2. Experimental
2.1. ZSM-5 zeolite preparation
Nano-scale ZSM-5 zeolite was prepared via hydrothermal syn-
thesis using a water/surfactant/organic solvent (emulsion method)
[31,32]. An aqueous solution containing the Si and Al source
materials was obtained by hydrolyzing each metal alkoxide in
a dilute tetrapropyl ammonium hydroxide (TPAOH)/water solu-
tion. The water solution (10 ml) thus obtained was added to
the surfactant/organic solvent (70 ml, surfactant concentration
of 0.5 mol/l). Polyoxyethylene-(15)-oleyl ether and cyclohexane
were employed as the surfactant and organic solvent, respectively.
The water/surfactant/organic solvent mixture thus obtained was
poured into a Teflon-sealed stainless steel bottle and heated to
423 K for 72 h. In order to obtain macro-scale ZSM-5 as a compara-
tive catalyst, hydrothermal synthesis was also carried out without
the surfactant/organic solvent (conventional method). The precip-
itates thus obtained were washed with alcohol, dried at 373 K
for 12 h, and calcined at 823 K for 3 h in an air stream. Physically
adsorbed and/or ion-exchanged sodium ions on the zeolite sur-
face were removed and exchanged with NH4+ using a conventional
ion exchange technique with a 10% NH4NO3 aqueous solution. The
powdered NH4+-zeolite described above was pelletized, crushed,
and sieved to yield samples ca. 0.3 mm in diameter that were sub-
sequently heated to 823 K to yield an H-ZSM-5 zeolite for the ATO
reaction.
2.4. Cumene and TIPB cracking
In order to evaluate the regioselective deactivation of the acid
sites, the catalytic cracking of isopropyl benzene (cumene) and
1,3,5-triisopropyl benzene (TIPB) was carried out using a fixed-bed
reactor at a reaction temperature of 573 K under an N2 flow at atmo-
spheric pressure [34,35]. The W/F ratio (W: amount of catalyst/g, F:
feed rate/g h−1) and the feed rate of cumene and TIPB were 0.13 h
and 0.18 h, and 1.0 ml/h and 1.33 ml/h, respectively. The catalyst
weight was 0.15 g. The composition of the exit gas was measured
via on-line gas chromatography (GC-2014, Shimadzu Co. Ltd.) with
a Unipak-S column for the FID detector.
2.2. ZSM-5 zeolite acidity control
2.5. ATO reaction
A ZSM-5 zeolite was exposed to an organic silane compound
vapor at 373 K in a nitrogen stream for 2 h, after which time the feed
of the organic silane compound was then halted in order to enable
the removal of any silane compounds physically adsorbed on the
zeolite surface. The sample was then heated to 873 K in order to
chemically decompose the silane molecules adsorbed on the acid
sites and deposit a silicon-containing carbonaceous residue. The
sample was then calcined in an air stream at 873 K, during which
time the silicon-containing carbonaceous residue was converted
into SiO2 unit on each acid site, resulting in deactivation. In other
words, the SiO2 units were formed on the acid sites where the silane
compounds were chemically adsorbed. Previously, we showed that
SiO2 units can be formed on the acid sites of zeolites using this
CCS method, and that the regioselective deactivation of the acid
sites is achieved using silane compounds of different molecular
sizes [28–30]. In this study, two types of silane compounds were
employed for the first time: triphenyl silane (TPS) and phenyl silane
(PS). The order for the molecular size of the silane compounds and
pore size of an MFI-type zeolite is TPS > pore diameter of MFI ≈ PS.
Therefore, in the TPS treatment, because the molecular size of TPS
is larger than the pore size of the ZSM-5 zeolite, the acid sites on
the outer surface can be selectively deactivated. In contrast, during
The ATO reaction over ZSM-5 zeolite catalysts was carried out
using a fixed-bed reactor at a reaction temperature of 723 K under
an N2 flow at atmospheric pressure. The W/F ratio, feed rate of the
acetone, and catalyst weight were 0.5 h, 1.8 ml/h, and 0.71 g, respec-
tively. The composition of the exit gas was measured via on-line
gas chromatography (GC-8A, Shimadzu Co. Ltd.) with a Porapak-
Q column as the FID detector. The amount of coke deposited on
the catalyst was determined via thermogravimetric analysis (TG;
TGA-50, Shimadzu Co. Ltd.).
3. Results and discussion
3.1. Preparation of ZSM-5 zeolites with different crystal sizes
Fig. 1 shows X-ray diffraction patterns of samples obtained using
conventional hydrothermal and emulsion preparation methods.
The patterns of the samples contained peaks corresponding to MFI-
type zeolites. Figs. 2 and 3 show FE-SEM micrographs and NH3–TPD
profiles of the obtained samples, respectively. As can be seen in
Fig. 2, the crystal sizes of the macro- and nano-ZSM-5 zeolites were

H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
129
5
macro-ZSM-5
Reference MFI
(a)
Desorption from
strong acid sites
4
3
2
1
0
nano-ZSM-5
(b)
5
10
15
20
25
30
35
40
45
50
2 / degree
350
450
550
650
750
850
Temperature [K]
Fig. 1. XRD patterns of (a) macro- and (b) nano-ZSM-5 zeolites.
Fig. 3. NH3–TPD profiles of macro- and nano-ZSM-5 zeolites.
approximately 1500 nm and 150 nm, respectively. In addition, as
can be seen in Fig. 3, the ZSM-5 zeolites exhibited nearly the same
amounts of strong acid sites regardless of the crystal size. Table 1
presents the Si/Al ratios determined via XRF and the surface areas
of the obtained samples. Notably, the Si/Al ratios were nearly the
same as the Si/Al ratio calculated from the Si and Al concentrations
present in the synthetic solutions. Furthermore, the nano-ZSM-5
zeolite exhibited nearly the same BET surface area of 400 m2/g as
the macro-ZSM-5 zeolite. As a result, the macro- and nano-ZSM-5
zeolites with nearly the same types, strengths and concentrations of
acid sites were obtained. Next, in order to investigate the effects of
crystal size and the regioselective deactivation of the ZSM-5 zeo-
lites on their catalytic performance, these zeolites were used as
catalysts for the ATO reaction.
Table 1
BET surface areas and Si/Al ratios of macro- and nano-ZSM-5 zeolites.
Zeolite
SBET (m2 g−1
)
Si/Almeasured by XRF
Untreated macro-ZSM-5
Untreated nano-ZSM-5
TPS-treated nano-ZSM-5
PS-treated nano-ZSM-5
405
400
391
370
89
81
–
–
formation of SiO2 units [28–30]. In this study, the CCS method was
carried out using two different silanes: triphenyl silane (TPS) and
phenyl silane (PS).
The FT-IR spectra of the adsorbed pyridine and ac-NH3–TPD pro-
files of the nano-ZSM-5 zeolite before and after CCS treatment can
be seen in Figs. 4 and 5, respectively. As shown in Fig. 4, the bands
near 1450 and 1540 cm−1 are attributed to Lewis acid sites (LAS)
and Brønsted acid sites (BAS), respectively [36]. Although the IR
absorbance was qualitatively measured, it can be clearly seen that
the intensity of the peaks corresponding to the acid sites decreased
after CCS treatment. These changes in the acidity due to the CCS
treatment could also be clearly observed in the NH3–TPD profiles,
as shown in Fig. 5. With both silane compounds, the amount of
NH3 desorbed from the acid sites (mainly BAS) decreased after CCS
treatment due to the formation of SiO2 units on the acid sites. In
addition, the TPD profiles indicated that the quantity of acid sites for
the PS-treated nano-ZSM-5 zeolite was less than that for the TPS-
treated nano-ZSM-5 zeolite. The greater decrease in the amount of
strong acid sites for the PS-treated nano-ZSM-5 zeolite is attributed
to the relationship between the pore diameter of the ZSM-5 and the
molecular size of the silane compound. In the case of TPS, because
3.2. Preparation of ZSM-5 zeolites with different acid site
locations
Although the nano-ZSM-5 zeolite exhibited stable catalytic
activity, the lifetime of this catalyst was insufficient and the prod-
uct selectivity was unstable [21] due to the formation of coke on
the external surface of the zeolite crystal and non-volatile car-
bonaceous compounds on the pore surfaces within the crystal.
These compounds inhibit diffusion and adsorption of the reactant
within the zeolite crystal. In order to overcome these problems,
suppression of coke formation due to excessive reaction of the
desired products is indispensable. Thus, the regioselective deac-
tivation of acid sites located on the external surface of the zeolite
crystal has been examined. One promising approach is the CCS (cat-
alytic cracking of silane) method, which can achieve control and/or
modification of the acidity of zeolites through the regioselective
Fig. 2. SEM micrographs of (a) macro- and (b) nano-ZSM-5 zeolites.

130
H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
BAS
BAS+LAS
LAS
200
(a)
150
(b)
100
untreated nano-ZSM-5
TPS-treated nano-ZSM-5
PS-treated nano-ZSM-5
(c)
50
1550
1500
1450
1400
Wavenumber [cm-1]
0
Fig. 4. FT-IR spectra of pyridine-adsorbed nano-ZSM-5 zeolites. (a) Before CCS treat-
ment, (b) after TPS treatment, and (c) after PS treatment.
0
0.2
0.4
0.6
0.8
1.0
P / P0 [-]
the molecular size of TPS is larger than the pore diameter of the
ZSM-5, the acid sites located on the external surface were selec-
tively deactivated. In contrast, because the molecular size of PS is
slightly smaller than the pore diameter of the ZSM-5 zeolite, not
only the acid sites on the external, but those on the internal sur-
face were deactivated. Thus, the difference in the molecular size
of these silane compounds influenced the number of acid sites that
were deactivated, leading to the difference in the NH3–TPD profiles.
During CCS treatment, because SiO2 units are selectively formed
only on the acid sites, the zeolitic pore properties, such as the pore
size and surface, should remain nearly unchanged before and after
treatment. In order to confirm this assumption, the N2 adsorption
and cracking activity for alkyl aromatics of nano-ZSM-5 zeolites
before and after CCS treatment were examined. Fig. 6 shows the N2
adsorption isotherms of nano-ZSM-5 zeolites before and after CCS
treatment. As anticipated, the modified zeolites exhibited nearly
the same micropore volume as the untreated nano-ZSM-5 zeolite.
It can thus be concluded that the CCS method has an insignificant
effect on the pore properties of zeolite catalysts.
Fig. 6. N2 adsorption isotherms of nano-ZSM-5 zeolites before/after CCS treatment.
gradually decreased with time on-stream, indicating that the exter-
nal acid sites became poisoned due to coke formation [34,35].
The CCS-treated nano-ZSM-5 zeolites (Fig. 7(b) and (c)), on the
other hand, showed almost no TIPB conversion despite having some
intact acid sites, as confirmed by the high cumene conversion com-
pared to the untreated ZSM-5 catalyst. The slight cracking of TIPB at
the beginning of the reaction may be due to the very small number
of acid sites remaining on the external of the zeolite. No additional
70
60
50
Cumene
40
(a)
30
20
10
TIPB
3.3. Cumene and TIPB cracking over nano-ZSM-5 with different
acid site locations
0
0
2
4
4
4
6
8
8
8
10
10
10
70
60
50
40
30
20
10
0
The results of cumene and TIPB cracking using nano-ZSM-5 with
and without CCS treatment are presented in Fig. 7. In the cumene
and TIPB cracking, it is well-known that these cracking reactions
proceed on the Brønsted acid sites (not Lewis acid sites) [34,35].
The untreated nano-ZSM-5 (Fig. 7(a)) exhibited high catalytic activ-
ity for both reactions. In this case, however, the conversion of TIPB
Cumene
(b)
(c)
TIPB
5
0
2
6
Desorption from
strong acid sites
70
60
50
40
30
20
10
0
untreated nano-ZSM-5
(before CCS treatment)
4
3
2
1
0
Cumene
TPS-treated nano-ZSM-5
TIPB
PS-treated
nano-ZSM-5
0
2
6
Time on-stream / h
350
450
550
650
750
850
Temperature [K]
Fig. 7. Catalytic cracking results for isopropyl benzene (cumene) and 1,3,5-
triisopropyl benzene (TIPB) over (a) untreated nano-ZSM-5, (b) TPS-treated
nano-ZSM-5, and (c) PS-treated nano-ZSM-5 zeolites.
Fig. 5. NH3–TPD profiles of the nano-ZSM-5 zeolites before/after CCS treatment.

H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
131
100
100
80
60
40
20
0
80
60
40
20
0
Aromatics
untreated macro-ZSM-5
untreated nano-ZSM-5
TPS-treated nano-ZSM-5
PS-treated nano-ZSM-5
Others
Paraffins
Ethylene
0
20
40
60
80
100 120 140 160 180
Time on-stream / h
Propylene
Fig. 9. Changes in acetone conversion with time on-stream over macro- and nano-
ZSM-5 with/without PS treatment.
11 C- mol%) compared with that of the nano-ZSM-5 zeolite before
CCS treatment (light olefins: 52 C-mol%, aromatics: 29 C- mol%).
The improvement in the light olefins selectivity is ascribed to the
suppression of non-selective and excessive reactions producing
aromatics. After TPS-treatment, high ethylene and propylene yields
were obtained because the generated isobutylene reacted with
itself to produce ethylene and propylene via dimerization/cracking
over the acid sites inside the zeolite crystal, where strong acid-
ity remained. In contrast, with the PS-treated catalyst, both a high
isobutylene yield and significantly reduced aromatic selectivity
were achieved. After PS treatment, in addition to the acid sites
on the external surface of the zeolite, the acid sites in the cross-
sectional spaces inside the zeolite crystal, which are wide open
spaces, were also deactivated. Accordingly, it is considered that
SiO2 formation on both the external surface and the internal pore
surfaces led to the suppression of the dimerization and cyclization
of isobutylene and subsequent formation of aromatics and coke. As
a result, both CCS-treated nano-ZSM-5 zeolites were effective cat-
alysts for the production of light olefins at high yield. Furthermore,
regioselective deactivation of the acid sites was achieved using the
CCS method with silane compounds with different molecular sizes,
leading to control of zeolite acidity and consequently the product
selectivity of the catalysts.
Fig. 9 shows the changes in acetone conversion with time on-
stream over the ZSM-5 zeolites with and without CCS treatment.
Although the untreated nano-ZSM-5 catalyst exhibited stable activ-
ity compared to that of the macro-ZSM-5 zeolite, the conversion
decreased to 20 C-mol% after 30 h. In contrast, the TPS-treated and
PS-treated nano-ZSM-5 maintained a high conversions above 93%
for 50 h and 70% for 180 h, respectively. This result indicates excel-
lent stability compared to that of the ZSM-5 zeolite without CCS
treatment. The amounts of coke formed on the catalysts as mea-
sured by TG analysis are listed in Table 2. Although the reaction time
of the nano-ZSM-5 zeolite after CCS treatment using PS was much
longer than that of the nano-ZSM-5 catalyst without CCS treat-
ment, the amount of coke formed on the PS-treated nano-ZSM-5
zeolite after the reaction (9.0 wt% after 180 h) was comparable to
that of the macro- and nano-ZSM-5 catalysts before CCS treatment
(macro-ZSM-5: 8.5 wt% after 6.5 h; nano-ZSM-5: 8.2 wt% after 30 h).
Isobutylene
Fig. 8. Product selectivities after 2.5 h for reactions over nano-ZSM-5 with/without
CCS treatment.
conversion of TIPB was observed, however, indicating that the
complete deactivation of the external acid sites was readily accom-
plished by the slight coke deposition that occurred during TIPB
cracking at the initial stage of the reaction. These results strongly
imply that the Brønsted acid sites on the external surface of nano-
ZSM-5 catalysts are selectively deactivated after CCS treatment. In
contrast, the ZSM-5 zeolites after CCS treatment exhibited an activ-
ity for cumene cracking as high as the untreated nano-ZSM-5. The
high catalytic activity for cumene suggests that the cumene diff-
used into the channels of the CCS-treated ZSM-5-zeolites, which in
turn indicates that the pore size of the ZSM-5 zeolites were largely
unchanged after CCS treatment, because the minimum molecular
diameter of cumene, which corresponds to the diameter of a ben-
zene ring, is nearly the same as the pore diameter of untreated
ZSM-5 zeolite. Notably, the initial cumene conversion for the PS-
treated nano-ZSM-5 (51 C-mol%) was slightly lower than that for
the TPS-treated nano-ZSM-5 (55 C-mol%), indicating that more SiO2
units were formed on the Brønsted acid sites located on the internal
and external surfaces of the PS-treated crystal, but a sufficient num-
ber of the Brønsted acid sites remained on the internal surface to the
catalytic reaction. Next, to further investigate the effects of the CCS
treatment on catalytic performance, these modified nano-ZSM-5
zeolites were used as catalysts for the ATO reaction.
3.4. Catalytic performance of ZSM-5 zeolites with different acid
site locations
In the ATO reaction, isobutylene is first produced as an aldol
condensation product of acetone, and then converted to propyl-
ene, ethylene, and aromatics over the acid sites. These light olefins
are intermediate chemicals in a series of reactions that ends with
the production of aromatics and coke. The product selectivities
and acetone conversions shortly after initiation of the reaction
(2.5 h) over the nano-ZSM-5 zeolites with and without CCS treat-
ment are shown in Fig. 8. CCS treatment of the nano-zeolite
improved the light olefin selectivity (TPS: 57 C-mol%, PS: 58 C-
mol%) and suppressed aromatics formation (TPS: 26 C-mol%, PS:
Table 2
Amount of coke produced using ZSM-5 zeolite catalysts as determined by TG
analysis.
Zeolite
Reaction time (h)
Coke amount (wt%)
Untreated macro-ZSM-5
Untreated nano-ZSM-5
PS-treated nano-ZSM-5
6.5
30.0
8.5
8.2
9.0
180.0

132
H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
70
60
50
40
30
20
10
0
Light olefins
Aromatics
Paraffins
Acetic acid
Others
0
20
40
60
80
100
120
140
160
180
Time on-stream / h
60
50
40
30
20
10
0
Isobutylene
Propylene
Ethylene
0
20
40
60
80
100
120
140
160
180
Time on-stream / h
Fig. 10. Product yields with time on-stream over the PS-treated nano-ZSM-5 zeolite.
Thus, the CCS treatment enabled the suppression of coke formation,
leading to the excellent stability.
summarizes the relationship between acetone conversion and light
olefins yield in the ATO reaction over macro- and nano-ZSM-5
zeolites before/after CCS treatment. The CCS-treated nano-ZSM-5
was exhibited high light olefins yield as compared with untreated
ZSM-5 zeolites. Therefore, it is believed that the CCS treatment is
effective for improving the light olefins yield. From these results,
it can be concluded that the modification of nano-ZSM-5 zeolites
using the CCS method is effective for improving the isobutylene
yield and catalytic stability when these zeolites are used as catalysts
for the ATO reaction.
Finally, Fig. 10 shows the product yields over the PS-treated
nano-ZSM-5 catalyst. It is apparent that the high yield of light
olefins (particularly isobutylene) was maintained for a long reac-
tion time (180 h), and that this catalyst exhibited remarkable
catalytic performance compared to that observed in our previous
study [21,22]. The isobutylene yield was increased with time-on
stream. This is because the consecutive reactions were suppressed
along with the decrease in the acidity due to coke formation. In
the ATO reaction, isobutylene was first produced from acetone, fol-
lowed by production of propylene, ethylene, and aromatics. These
intermediates are involved in a series of reactions that result in
the generation of coke as the terminal product. In this reaction, the
consecutive reactions were suppressed along with the decrease in
the acid amounts due to CCS treatment and coke formation. Fig. 11
4. Conclusion
The regioselective deactivation of the acid sites on nano-ZSM-5
zeolites was achieved using the CCS method with silane compounds
of different molecular sizes, leading to the control of the zeolite
acidity. These modified nano-ZSM-5 zeolites were then applied
as catalysts for the ATO reaction, and the regioselective deactiva-
tion of the acid sites due to CCS treatment resulted in remarkable
catalytic performance. In particular, the PS-treated nano-ZSM-5
zeolite exhibited a high yield of light olefins with excellent stability
compared to the results obtained with the macro- and nano-ZSM-5
zeolites without CCS treatment.
60
TPS-treated nano-ZSM-5
PS-treated nano-ZSM-5
50
40
Acknowledgements
untreated macro-ZSM-5
untreated nano-ZSM-5
This work was supported by a Research Grant Program of the
New Energy and Industrial Technology Development Organization
(NEDO) of Japan and the Global COE Program (Project No. B01:
Catalysis as the Basis for Innovation in Materials Science) from
the Ministry of Education, Culture, Sports, Science and Technology,
Japan.
30
60
70
80
90
100
Acetone conversion / C-mol%
Fig. 11. Relationship between acetone conversion and light olefins yield.

H. Konno et al. / Applied Catalysis A: General 475 (2014) 127–133
133
References
[17] D. Goto, Y. Harada, Y. Furumoto, A. Takahashi, T. Fujitani, Y. Oumi, M. Sadakane,
T. Sano, Appl. Catal. A: Gen. 383 (2010) 89–95.
[18] C. Yu, Q. Ge, H. Xu, W. Li, Appl. Catal. A: Gen. 315 (2006) 58–67.
[19] L. Liu, Q. Deng, Y. Liu, T. Ren, Z. Yuan, Catal. Commun. 16 (2011) 81–85.
[20] R.J. Schmidt, Appl. Catal. A: Gen. 280 (2005) 89–103.
[21] T. Tago, H. Konno, M. Sakamoto, Y. Nakasaka, T. Masuda, Appl. Catal. A: Gen 403
(2011) 183–191.
[22] T. Tago, H. Konno, S. Ikeda, S. Yamazaki, W. Ninomiya, Y. Nakasaka, T. Masuda,
Catal. Today 164 (2011) 158–162.
[23] S.H. McAllister, W.A. Bailey, C.M. Bouton, J. Am. Chem. Soc. 62 (1940)
3210–3215.
[1] Y. Yoshimura, N. Kijima, T. Hayakawa, K. Murata, K. Suzuki, F. Mizukami,
K. Matano, T. Konishi, T. Oikawa, M. Saito, T. Shiojima, K. Shiozawa, K.
Wakui, G. Sawada, K. Sato, S. Matsuo, N. Yamaoka, Catal. Surv. Jpn. 4 (2000)
157–167.
[2] O. Bortnovsky, P. Sazama, B. Wichterlova, Appl. Catal. A: Gen. 287 (2005)
203–213.
[3] C. Mei, P. Wen, Z. Liu, Y. Wang, W. Yang, Z. Xie, W. Hua, Z. Gao, J. Catal. 258
(2008) 243–249.
[4] A. Corma, J. Planelles, J. Sanchezmarin, F. Tomas, J. Catal. 93 (1985) 30–37.
[5] S.M. Babitz, B.A. Williams, J.T. Miller, R.Q. Snurr, W.O. Haag, H.H. Kung, Appl.
Catal. A: Gen. 179 (1999) 71–86.
[6] H. Mochizuki, T. Yokoi, H. Imai, R. Watanabe, S. Namba, J.N. Kondo, T. Tatsumi,
Micropor. Mesopor. Mater. 145 (2011) 165–171.
[7] K. Kubo, H. Iida, S. Namba, A. Igarashi, Micropor. Mesopor. Mater. 149 (2012)
126–133.
[8] S. Inagaki, S. Shinada, Y. Kaneko, K. Takechi, R. Komatsu, Y. Tsuboi, H. Yamazaki,
J.N. Kondo, Y. Kubota, ACS Catal. 3 (2013) 74–78.
[9] H. Konno, T. Okamura, Y. Nakasaka, T. Tago, T. Masuda, J. Jpn. Petrol. Inst. 55
(2012) 267–274.
[10] H. Konno, T. Okamura, Y. Nakasaka, T. Tago, T. Masuda, Chem. Eng. J. 207–208
(2012) 460–467.
[11] H. Konno, T. Tago, Y. Nakasaka, R. Ohnaka, J. Nishimura, T. Masuda, Micropor.
Mesopor. Mater. 175 (2013) 25–33.
[12] M. Stocker, Micropor. Mesopor. Mater. 29 (1999) 3–48.
[13] F.J. Keil, Micropor. Mesopor. Mater. 29 (1999) 49–66.
[14] Y. Kumita, J. Gascon, E. Stavitski, J.A. Moulijn, F. Kapteijn, Appl. Catal. A: Gen.
391 (2011) 234–243.
[15] T. Meng, D. Mao, Q. Guo, G. Lu, Catal. Commun. 21 (2012) 52–57.
[16] Y. Furumoto, Y. Harada, N. Tsunoji, A. Takahashi, T. Fujitani, Y. Ide, M. Sadakane,
T. Sano, Appl. Catal. A: Gen. 399 (2011) 262–267.
[24] J. Cejka, L. Kubelkova, P. Jiru, Collect. Czech. Chem. Commun. 54 (1989)
2054–2062.
[25] J. Cejka, P. Jiru, Collect. Czech. Chem. Commun. 54 (1989) 2998–3002.
[26] L. Kubelkova, J. Cejka, J. Novakova, Zeolite 11 (1991) 48–53.
[27] A.G. Panov, J.J. Fripiat, J. Catal. 178 (1998) 188–197.
[28] T. Tago, K. Iwakai, K. Morita, K. Tanaka, T. Masuda, Catal. Today 105 (2005)
662–666.
[29] T. Tago, M. Sakamoto, K. Iwakai, H. Nishihara, S.R. Mukai, T. Tanaka, T. Masuda,
J. Chem. Eng. Jpn. 42 (2009) 162–167.
[30] T. Masuda, N. Fukumoto, M. Kitamura, S.R. Mukai, K. Hashimoto, T. Tanaka, T.
Funabiki, Micropor. Mesopor. Mater. 48 (2001) 39–245.
[31] T. Tago, M. Nishi, Y. Kouno, T. Masuda, Chem. Lett. 33 (2004) 1040–1041.
[32] T. Tago, H. Konno, Y. Nakasaka, T. Masuda, Catal. Surv. Asia 16 (2012) 148–163.
[33] T. Masuda, Y. Fujikata, S.R. Mukai, K. Hashimoto, Appl. Catal. A: Gen. 165 (1997)
57–72.
[34] C. Anand, I. Toyama, H. Tamada, S. Tawada, S. Noda, K. Komura, Y. Kubota, S.W.
Lee, S.J. Cho, J.H. Kim, G. Seo, A. Vinu, Y. Sugi, Ind. Eng. Chem. Res. 51 (2012)
12214–12221.
[35] S. Inagaki, S. Shinoda, Y. Kaneko, K. Takechi, R. Komatsu, Y. Tsuboi, H. Yamazaki,
J.N. Kondo, Y. Kubota, ACS Catal. 3 (2013) 74–78.
[36] E.P. Parry, J. Catal. 2 (1963) 371–379.