Study of active sites on the MFI zeolite catalysts for the transformation of ethanol into propylene

Article abstract of DOI:10.1016/j.molcata.2010.06.008

Industrial demand for propylene is increasing. Alternative sources of propylene other than petroleum are of wide interest. The catalytic transformation of bioethanol is a promising avenue for the production of propylene. However, the reaction mechanism and the nature of active sites on the catalyst remained unclear. In this work, MFI-type zeolite catalysts with various Si/Al₂ ratios (73, 97, 128, 176, 207) were prepared by hydrothermal synthesis under atmospheric pressure. We characterized these MFI catalysts by BET surface area measurements, X-ray fluorescence, X-ray diffraction, scanning electron microscopy, temperature programmed desorption of ammonia, and ²⁷Al MAS NMR to understand the effect of the Si/Al ₂ ratio. The ethanol conversion reaction was systematically investigated using these catalysts according to the contact time and reaction temperature. The conversion of ethanol yielded similar product distributions over these catalysts at different contact times, indicating that the product distribution has no relationship with the Si/Al₂ ratio. Also, the propylene production rates per acidic site of these catalysts were similar. These results strongly imply that the acidic sites of the different MFI-type zeolite catalysts have the same nature, as well as the same reactivity. The catalytic activity was proportional to the number of acidic sites.

Full text of DOI:10.1016/j.molcata.2010.06.008

Journal of Molecular Catalysis A: Chemical 328 (2010) 114–118
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Study of active sites on the MFI zeolite catalysts for the transformation of ethanol
into propylene
Wei Xia, Atsushi Takahashi, Isao Nakamura, Hiromichi Shimada, Tadahiro Fujitani∗
Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced Industrial Science and Technology (AIST),
Tsukuba West 16-1 Onogawa, Tsukuba, Ibaraki 305-8569, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Industrial demand for propylene is increasing. Alternative sources of propylene other than petroleum
are of wide interest. The catalytic transformation of bioethanol is a promising avenue for the production
of propylene. However, the reaction mechanism and the nature of active sites on the catalyst remained
unclear. In this work, MFI-type zeolite catalysts with various Si/Al2 ratios (73, 97, 128, 176, 207) were
prepared by hydrothermal synthesis under atmospheric pressure. We characterized these MFI catalysts
by BET surface area measurements, X-ray fluorescence, X-ray diffraction, scanning electron microscopy,
temperature programmed desorption of ammonia, and 27Al MAS NMR to understand the effect of the
Si/Al2 ratio. The ethanol conversion reaction was systematically investigated using these catalysts accord-
ing to the contact time and reaction temperature. The conversion of ethanol yielded similar product
distributions over these catalysts at different contact times, indicating that the product distribution has
no relationship with the Si/Al2 ratio. Also, the propylene production rates per acidic site of these catalysts
were similar. These results strongly imply that the acidic sites of the different MFI-type zeolite catalysts
have the same nature, as well as the same reactivity. The catalytic activity was proportional to the number
of acidic sites.
Received 16 April 2010
Received in revised form 3 June 2010
Accepted 3 June 2010
Available online 11 June 2010
Keywords:
Ethanol
Propylene
MFI
Si/Al2 ratio
Active sites
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
in industrial application. Only a few literatures reported the con-
version of ethanol or ethylene to olefin. These literatures reported
As a major chemical intermediate, the production of propy-
lene always attracts broad attention in the industrial and academic
research field. Propylene is generally produced by steam thermal
cracking of the naphtha from the petroleum. However, due to the
global energy crisis and the increasing demand for propylene in the
industrial field, much more attention has been paid to alternative
resources to petroleum.
Methanol can be obtained in large quantities from fossil fuels
like petroleum, nature gas and coal. The process for methanol to
gasoline (MTG) and methanol to olefins (MTO), which can be used
to produce liquid fuels for the automotive sector or to make olefins
suitablefor polymerization, has beenwellestablishedandcommer-
cialized since 1980s [1,2]. Ethanol can be produced from renewable
sources such as biomass by fermentation. It also has great poten-
tial to be a candidate for the source of light olefins such as ethylene
and propylene. If the conversion of ethanol to propylene is selective
and efficient, propylene can be produced without the utilization
of petroleum resources. Therefore, ethanol to olefin (ETO) process
has recently received much attention but has not yet been used
the conversion of ethanol to hydrocarbons over unmodified and
metal-modified ZSM-5 catalysts [3–9]. The conversion of ethy-
lene to propylene has also been investigated over SAPO-34 [10]
and nickel ion-loaded mesoporous silica catalysts [11]. Recently,
LEV zeolite catalyst was also used for the ethanol to olefin reac-
tion [12]. However, as mentioned above, much attention has been
paid to mainly synthesize ethylene, gasoline, or aromatic hydrocar-
bons from ethanol. There are few reports concerning the ethanol to
propylene process [8].
The zeolite H-ZSM-5 has been widely used as the active cat-
alyst for various reactions owing to its special pore structure,
thermal and hydrothermal stability, acidity, and reactivity. It has
long been used for MTO conversion in industry [1,2]. Furthermore,
this catalyst was found to be the alternative catalyst for ethanol
transformation into hydrocarbons. Understanding the reaction
mechanism and active sites of H-ZSM-5 zeolite catalysts is crucial
for improving catalytic performance in ethanol conversion; how-
ever, the reaction mechanism and the nature of the active sites
remain unclear.
Recently, we studied the conversion reaction of ethanol to
propylene on H-ZSM-5 zeolite catalysts. In those works, we spec-
ulated that a series of MFI-type zeolite catalysts with various
Si/Al2 ratios could be used to explore both the catalytic perfor-
∗
Corresponding author. Tel.: +81 29 861 8454; fax: +81 29 861 8374.
E-mail address: t-fujitani@aist.go.jp (T. Fujitani).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.06.008

W. Xia et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 114–118
115
mance of these catalysts under various operating conditions and
the nature of active sites for the conversion of ethanol to propy-
lene. In this work, we successfully synthesized a series of MFI-type
zeolite catalysts with various Si/Al2 ratios (73, 97, 128, 176, 207)
and characterized the catalysts by BET surface area measurements,
X-ray fluorescence (XRF), X-ray diffraction (XRD), scanning elec-
tron microscopy (SEM), temperature programmed desorption of
ammonia (NH3-TPD), and 27Al MAS NMR to understand the effects
of Si/Al2 ratios. We systematically converted ethanol to propylene
over these catalysts at various contact times and reaction temper-
atures.
2. Experimental
2.1. Preparation of catalyst
A series of MFI-type zeolite catalysts with various Si/Al2 ratios
was synthesized hydrothermally. Colloidal silica (SNOWTECH O)
and Al(NO3)3·9H2O were used as Si and Al sources, respec-
tively. Tetra-n-propylammonium bromide was used as a template.
Al(NO3)3·9H2O was first dissolved in the Colloidal Silica, and
then NaCl and NaOH were added to get solution A. The tem-
plate was dissolved in water to get solution B. Solution B was
added slowly to solution A with stirring. After the addition was
completed, the gel was stirred for at least 30 min to homogenize
while being heated to 373 K. Then the hydrothermal conver-
sion was conducted under reflux for 14 days at atmospheric
pressure. Finally, the Na-type zeolites were filtered off, washed
thoroughly with deionized water until no Cl− remained in the fil-
trate, dried at 383 K for 16 h, and calcined at 873 K for 4 h in an
air stream to burn off the template remaining in the zeolite crys-
tals.
Fig. 1. XRD patterns of the synthesized MFI series of catalysts with various Si/Al2
ratios.
adsorption of NH3/He (30 mL/min) for 1 h took place at 373 K, and
then the catalyst was flushed with He (30 mL/min) at the same tem-
perature for 1 h to remove NH3 that was physically adsorbed on the
sample surface. TPD measurements were done from 373 to 873 K
with a heating rate of 10 K/min, with He as the carrier gas. The des-
orbed NH3 was quantified with a thermal conductivity detector.
The 27Al MAS NMR spectra were obtained on a Varian VXR-400
spectrometer at 104.3 MHz with a ca. 4.5 kHz spinning speed and
1.73 ␮s pulses for 4000 scans. The chemical shift was referenced to
3+
an external standard of Al (H2O)6
.
2.3. Catalytic apparatus and ethanol transformation reaction
H-type zeolite catalysts were prepared from Na-type catalysts
by the ion-exchange method. The Na-type zeolite catalysts were
repeated ion-exchange for 4 times with a 1 M HCl aqueous solution
(500 mL) under reflux at 373 K for 7 days. Finally, the product was
filtered off, washed with deionized water until no Cl− remained in
the filtrate and dried at 383 K for 16 h. After that, the catalysts were
calcined in air at 873 K for 4 h with a heating rate of 4.5 K/min.
Ethanol (99.5%, Wako) was used without further purification.
All the MFI zeolite samples were powdered, pressed, crushed, and
sieved with a 14–22 mesh. The conversion reaction was carried
out at atmospheric pressure using a continuous-flow fixed-bed
quartz tubular reactor with 0.1–0.8 g catalyst. Reaction tempera-
ture (673–873 K) was monitored by a thermocouple in the catalyst
bed. Prior to each reaction, the catalyst was pretreated in a flow
2.2. Characterization
Surface areas and pore volumes were determined by the N2
adsorption-desorption method (BET method) at 77 K by using a
volumetric unit (Micromeritics ASAP 2020). Prior to the adsorp-
tion measurements, each sample was degassed at 623 K for 10 h
under reduced pressure. XRD patterns of the solid products were
measured with a powder X-ray diffractometer (Rigaku, RINT 2000)
with Cu K␣ monochromatized radiation (ꢀ = 0.154178 nm) at 40 kV
and 40 mA. Si/Al2 ratios of these catalyst samples were determined
by XRF (RIGAKU EDXL 300). The crystal morphology was observed
by SEM (Jeol-JSM-5600). NH3-TPD was also used (BEL-CAT-32) to
investigate the acidic site properties. Catalyst sample (0.1 g) was
placed in a small quartz tube and dried in flowing He (99.99%,
30 mL/min) at 773 K for 1 h. The sample was cooled to 373 K, the
Table 1
Characteristics of MFI-type zeolite catalysts.
Sample no.
Si/Al2 ratio
BET surface
area (m2 g−1
Pore volume
)
(cm3 g−1
)
1
2
3
4
5
73
97
128
176
207
371
371
418
379
414
0.212
0.209
0.202
0.175
0.212
Fig. 2. 27Al MAS NMR spectra of the series of synthesized MFI catalysts with various
Si/Al2 ratios.

116
W. Xia et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 114–118
Fig. 3. SEM images of the series of synthesized MFI catalysts with various Si/Al2 ratios.
of N2 (flow rate: 100 mL/min) at 873 K for 1 h. Reactants (ethanol
concentration 50 vol% in a flow of N2) were fed into the reactor
by means of a micro pump and were vaporized before entering
the reactor. The reaction products were analyzed by using on-line
gas chromatography (hydrogen-flame ionization detector with RT-
alumina PLOT (Restek USA) for C1–C4 hydrocarbon, and thermal
conductive detector with Shincarbon ST (Shinwa Chem. Ind. Ltd.,
Japan) for N2 and H2).
3. Results and discussion
3.1. Characterization of MFI zeolite catalysts
Table
1
summarizes the XRF and nitrogen adsorp-
tion/desorption (BET) results for these MFI zeolite catalysts.
The BET surface areas of all the catalysts were between 370 and
420 m2 g−1. The pore volumes of the catalysts were mostly about
0.2 cm3 g−1
.
These MFI zeolites were characterized for their crystalline
nature by XRD. The XRD patterns of the synthesized MFI zeolites are
shown in Fig. 1. All samples’ XRD patterns presented similar diffrac-
tion peaks, which were consistent with typical X-ray patterns of
MFI-type zeolite structure, and no other peak could be observed.
The results indicated that the MFI-type catalysts with high purity
Fig. 4. NH3-TPD profiles of the series of synthesized MFI catalysts with various Si/Al2
ratios.

W. Xia et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 114–118
117
Fig. 5. Product distribution over the series of synthesized MFI catalysts: (᭹) ethylene, (ꢀ) ethane, (ꢀ) propylene, (ꢁ) propane, (ꢁ) butene, (♦) butane and (ꢂ) others. Reaction
conditions: time-on-stream, 30 min; reaction temperature, 823 K; P = 0.1 MPa.
were successfully synthesized. As shown in Fig. 1, the XRD pat-
terns have different relative peak intensity. The MFI catalysts with
the higher Si/Al2 ratio showed higher peak intensity.
alyst framework. Furthermore the intensity of the signal at 56 ppm
decreased with increasing Si/Al2 ratio.
SEM was also used to investigate the crystal size and morphol-
ogy of these MFI catalyst samples. The micrographs were presented
in Fig. 3. Most of these crystals showed the MFI-typical hexago-
nal morphology. The crystals of the catalyst with Si/Al2 = 73 were
smaller than 1 ␮m. The small crystals of the catalyst with Si/Al2 = 97
tended to form agglomerates around 5–10 ␮m. The catalyst with
Si/Al2 = 128 included some small crystals (smaller than 2.5 ␮m)
together with large crystals (about 10 ␮m), all with the same shape.
The crystals of the catalysts with Si/Al2 = 176 and 207 were similar
Fig. 2 displays the 27Al MAS NMR spectra of the catalyst sam-
ples with various Si/Al2 ratios. The peak at 56 ppm is commonly
assigned to four-coordinated framework aluminum, while the peak
at 0 ppm is referred to non framework octahedral aluminum [13].
As shown in Fig. 2, all the catalyst samples exhibited the same
framework aluminum (tetrahedrally coordinated) peaks at about
56 ppm. No nonframework aluminum peak could be observed. The
results indicate that all the Al atoms are incorporated into the cat-
Table 2
Ethanol conversion over synthesized MFI catalysts.
Si/Al2 ratio
W/F
Ethylene
Ethane
Propylene
Propane
Butene
Butane
Others
73
97
128
176
207
0.0125
0.0125
0.03
0.05
0.05
34.1
37.6
40.7
35.6
31.6
1.0
0.8
0.9
1.1
1.0
24.5
28.4
25.0
25.5
28.6
9.4
8.5
7.2
7.6
8.9
9.7
11.0
9.3
9.8
11.4
4.3
2.4
1.8
1.9
2.5
16.7
12.2
15.0
16.5
16.7
Reaction conditions: time-on-stream, 30 min; reaction temperature, 823 K; P = 0.1 MPa.

118
W. Xia et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 114–118
with high selectivity for ethanol transformation to propylene, it is
very important to inhibit the transformation from propylene and
butene to over C5 products, which will be further investigated in
our future work.
We also further compared and summarized the above product
distribution results. At various contact times (W/F), similar product
distributions were obtained on all five catalysts (Table 2). This result
strongly implies that the product distribution is independent of the
Si/Al2 ratio. The nature of acidic sites among H-ZSM-5 catalysts
should be the same.
The high-temperature peak acid sites were reported to be con-
nected with the active center for the conversion of methanol [14].
We calculated the number of strong acid sites corresponding to
the desorption amount of NH3 at high temperature from the NH3-
TPD spectra of our zeolite catalyst series; also, we calculated the
initial propylene space-time yield (STY) of these MFI-type cata-
lysts. Fig. 6 showed the relationship between STY and the NH3
desorption amount at high temperature (673 K). The relationship
of STY to NH3 desorption amount at the high temperature (673 K)
was almost linear; that is, STY increased linearly with the num-
ber of strong acid sites. So the strong acid sites are the active sites
for the ethanol conversion to propylene. The turnover frequency
on each active site was same on the different Si/Al2-ratio cata-
lysts. This indicates that the active sites are the same in nature on
the different H-ZSM-5 catalysts. For the ethanol conversion reac-
tion, only the number of active sites on the MFI catalysts changed
with Si/Al2 ratio. On the basis of the product distribution and TPD
results, we propose that the nature and reactivity of the acidic
sites are the same on MFI-type zeolite catalysts with different
Si/Al2 ratios.
Fig. 6. Correlation between space-time yield and the NH3 desorption amount at
high temperature (673 K).
in shape, but the latter were a little larger. Clearly, the Si/Al2 ratio
affected the morphology of these MFI catalysts. The crystals size
increased with Si/Al2 ratio, a conclusion that is also supported by
the XRD peak intensities.
We investigated the amount and strength of acidic sites of the
zeolite catalyst series by using the NH3-TPD method (Fig. 4). All
the catalyst samples exhibited the same two well-resolved NH3
desorption peaks: the low-temperature peak at about 473 K and
the high-temperature peak at about 673 K, corresponding to the
weak and strong acid sites, respectively. This result means that the
strengths of the acid sites of these MFI catalysts were the same.
The spectral peak area decreased with increasing Si/Al2 ratio, which
indicates that the density and total amount of acidic sites decreased
with increasing Si/Al2 ratio.
4. Conclusions
A series of MFI-type zeolite catalysts with various Si/Al2 ratios
was successfully synthesized. To understand the effects of Si/Al2
ratios on the catalysts, we characterized them by BET surface
area measurements, XRF, XRD, SEM, NH3-TPD and 27Al MAS NMR.
The conversion of ethanol over these catalysts at various contact
times yielded similar product distributions, which indicates that
the product distribution is independent of the Si/Al2 ratio. Also, the
production rates of propylene per acidic site of these catalysts were
identical. These results strongly imply that the acidic sites on the
different MFI-type zeolite catalysts have the same nature and reac-
tivity; the catalytic activity should be proportional to the number
of acidic sites.
3.2. Transformation of ethanol to propylene
Over each of the synthesized MFI-type catalysts, we carried out
the ethanol conversion reactions under atmospheric pressure at
673, 723, 773, 823, and 873 K. Under these reaction conditions on
these catalysts, ethanol was completely converted to hydrocarbons,
such as ethylene, propylene, and other higher hydrocarbons. We
compared catalytic performance at various contact times and reac-
tion temperatures. The optimum yield for propylene (at various
reaction temperatures) was 24–28%. Fig. 5 shows the production
distribution on these catalysts at 823 K. The contact time is defined
as W/F, where W denotes the catalyst weight (g) and F denotes
the total flow rate (mL/min). As shown in Fig. 5, over all catalysts,
ethylene was the main product under very low W/F conditions. As
W/F increased, ethylene yield decreased, and propylene and butene
yields increased. Paraffins (ethane, propane, and butane) increased
with the decrease of olefins (ethylene, propylene, and butene),
which were produced by hydrogenerating olefin. On the basis of
these results, we proposed the reaction mechanism. First, ethylene
was produced from ethanol. Then the ethylene was converted in
parallel to propylene and butene. Finally, propylene, and butene
were converted to propane, butane, and other higher hydrocarbons
(olefin, paraffins, and aromatics). The details of the reaction mech-
anism will be presented in a future report. As shown in Fig. 5, the
contact time for optimum propylene yield increased with increas-
ing Si/Al2 ratio, perhaps due to variations in MFI catalyst acidic site
density with Si/Al2 ratio: the acidic site density of MFI catalyst with
high Si/Al2 ratio was lower than that of catalyst with low Si/Al2
ratio. So the contact time when the optimum propylene yield was
obtained was delayed. In addition, in order to develop the catalyst
References
[1] C.D. Chang, A.J. Silvestri, J. Catal. 47 (1977) 249–259.
[2] P. Dejaifve, J.C. Vedrine, V. Bolis, E.G. Derouane, J. Catal. 63 (1980)
331–345.
[3] J. Schulz, F. Bandermann, Chem. Eng. Technol. 17 (1994) 179–186.
[4] A.T. Aguayo, A.G. Gayubo, A.M. Tarrío, J. Bilbao, J. Chem. Technol. Biotechnol.
77 (2002) 211–216.
[5] M. Inaba, K. Murata, M. Saito, I. Takahara, Green Chem. 9 (2007) 638–646.
[6] N.R.C. Fernandes Machado, V. Calsavara, N.G.C. Astrath, A. Medina Neto, M.L.
Baesso, Appl. Catal. A 311 (2006) 193–198.
[7] V. Calsavara, M.L. Baesso, N.R.C. Fernandes Machado, Fuel 87 (2008)
1628–1636.
[8] K. Murata, M. Inaba, I. Takahara, J. Jpn. Petrol. Inst. 51 (2008) 234–239.
[9] R. Johansson, S.L. Hruby, J. Rass-Hansen, C.H. Christensen, Catal. Lett. 127 (2009)
1–6.
[10] H. Oikawa, Y. Shibata, K. Inazu, Y. Iwase, K. Murai, S. Hyodo, G. Kobayashi, T.
Baba, Appl. Catal. A: Gen. 312 (2006) 181–185.
[11] M. Iwamoto, Y. Kosugi, J. Phys. Chem. C 111 (2007) 13–15.
[12] T. Inoue, M. Itakura, H. Jon, Y. Oumi, A. Takahashi, T. Fujitani, T. Sano, Microp-
orous Mesoporous Mater. 122 (2009) 149–154.
[13] W. Zhang, X. Bao, X. Guo, X. Wang, Catal. Lett. 60 (1999) 89–94.
[14] H. Itoh, C.V. Hidalgo, T. Hattori, M. Niwa, Y. Murakami, J. Catal. 85 (1984)
521–526.