Selective conversion of dimethyl ether to propylene and light olefins over modified H-ZSM-5

Article abstract of DOI:10.1246/cl.2005.970

Light olefins especially propylene was formed selectively with a propylene/ethylene ratio of 10, from direct conversion of dimethyl ether over ZrO₂ and H₃PO₄-modified H-ZSM-5 catalyst at atmospheric pressure. The improved propylene selectivity was due to adjusted acid property and shape-selective microspores of the modified H-ZSM-5. Copyright

Full text of DOI:10.1246/cl.2005.970

970
Chemistry Letters Vol.34, No.7 (2005)
Selective Conversion of Dimethyl Ether to Propylene and Light Olefins over Modified H-ZSM-5
Tian-Sheng Zhao,y;yy Tomokazu Takemoto, Yoshiharu Yoneyama, and Noritatsu Tsubakiꢀ
Department of Applied Chemistry, School of Engineering, Toyama University, Gofuku 3190, Toyama 930-8555
yKey Laboratory of Energy Resources & Chemical Engineering, Ningxia University, Yinchuan 750021, P. R. China
yyVenture Business Laboratory of Toyama University, Gofuku 3190, Toyama 930-8555
(Received April 7, 2005; CL-050469)
Table 1. DME conversion over modified H-ZSM-5a
Light olefins especially propylene was formed selectively
with a propylene/ethylene ratio of 10, from direct conversion
of dimethyl ether over ZrO2 and H3PO4-modified H-ZSM-5 cat-
alyst at atmospheric pressure. The improved propylene selectiv-
ity was due to adjusted acid property and shape-selective micro-
spores of the modified H-ZSM-5.
Run
Temp. (K)
DME conv. (%)
Product distribution (C-mol %)
1
2
3
4
723
100
523
8.60
723
100b
723
100c
C1
C2
C2
C3
C3
0.77
4.57
0.06
45.4
0.04
0
2.91
25.0
0
47.3
0
0.26
2.74
0.03
44.5
0
0
15.9
11.2
63.1
15.5
0.42
13.4
0.16
19.6
10.1
11.2
7.13
7.88
40.1
10.8
5.30
4.51
7.32
2.18
¼
¼
Dimethyl ether (DME) can be directly synthesized from
syngas with the thermodynamic advantage over methanol and
several new DME industrial processes are under development.
DME can be used as LPG (liquefied petroleum gas) alternative,
clean diesel engine fuel without soot. Besides these energy-
related utilizations, chemical conversion is expected. On the
other hand, propylene need increases dramatically, resulting in
very high price over ethylene. Based on these facts, it is very
significant to develop a catalyst to selectively convert DME to
propylene.
It is known that there exists fast conversion equilibrium
among methanol, DME and water. Although methanol-to-hydro-
carbons has been extensively studied,1 for example, by changing
acid strength and topology of catalysts,2 or using various shape-
selective catalysts or two-stage reactor,3 controlling of the prod-
uct distribution remains different. Methanol to olefin (MTO)
process usually produces mixed C2–C4 lighter olefins4–8 while
high selectivity to ethylene can be obtained over modified SAPO
catalyst.9 ZSM-5 exhibits high selectivity to aromatics and
paraffins in methanol conversion.10 Improved selectivity to light
olefins can be achieved using modified ZSM-5 and suitable proc-
ess conditions.4–6 But the selectivity to propylene over ZSM-5-
related catalysts is extremely low until now.
i-C4
¼
0
C4
n-C4
14.6
2.94
9.05
75.2
6.70
4.15
1.95
0
10.8
64.6
15.0
¼
d
¼
C2 –C4 (C-mol %)
C5
C6
C7
C8
C9
aCatalyst, H3PO4/ZrO2 (12.5%, wt.)/H-ZSM-5, DME/N2
(mol), 1:5; W/F (DME), 10 gꢁhꢁmolꢂ1; 0.1 MPa; 2 h; bTime-on-
stream, 30 h; cH-ZSM-5 catalyst; dC5–C9, including all forms of
hydrocarbons.
d
5.23
5.96
d
2.81
0.72
0
2.02
1.15
0.74
d
d
0
H3PO4/12.5%ZrO/H-ZSM-5
2
12.5%ZrO/H-ZSM-5
2
Recently using ZrO2 and H3PO4-modified H-ZSM-5 cata-
lyst, we found that propylene was selectively formed from direct
conversion of DME. This was attributable to the adjusted acid
property and shape-selective microspores of H-ZSM-5.
H-ZSM-5
Table 1 shows the results of DME conversion over ZrO2
and H3PO4-modified H-ZSM-5.11 For comparison, the result
of DME conversion over pure H-ZSM-5 is listed. It is clear that
DME converted completely over H3PO4/12.5%-ZrO2/H-ZSM-
5 at 723 K (Run 1). The products terminated at C9 hydrocarbons
and centered at propylene with selectivity of 45.4%. The ratio of
propylene/ethylene was near 10 and the C2–C4 olefins selectiv-
ity was 64.6%. The formation of lighter olefins was significantly
improved, compared with that using H-ZSM-5, where the ratio
of propylene/ethylene was 1.5 and the selectivity to C2–C4
olefins was 40.1% (Run 4). Moreover, DME showed very low
conversion at 523 K while the selectivity of C2–C4 olefins was
75.2% and the ratio of propylene/ethylene was 1.9 (Run 2).
Thirty hours of continuous activity test of H3PO4/12.5%-
ZrO2/H-ZSM-5 at the same conditions as shown in Run 3
10
20
30
2 Theta
40
50
60
70
Figure 1. XRD patterns of modified H-ZSM-5.
demonstrated that DME conversion was stable. The ratio of
propylene/ethylene was even 16.
XRD results (Figure 1)12 indicated that after ZrO2 and
H3PO4 modification, both 12.5%-ZrO2/H-ZSM-5 and H3PO4/
12.5%-ZrO2/H-ZSM-5 showed similar diffraction peaks as
H-ZSM-5, except slight weakness in intensity. The crystallite
particle sizes of H-ZSM-5 by SEM at the amplifying times of
5000 were about 120–300 nm. The particles changed in appear-
ance and morphology, tending to cohere after the modification.
Pore size distribution of the modified H-ZSM-5 is shown
in Figure 2.13 Compared with H-ZSM-5, the pore size of the
Copyright Ó 2005 The Chemical Society of Japan

Chemistry Letters Vol.34, No.7 (2005)
971
pared with H-ZSM-5, which reduced hydrogen transfer reaction.
In addition, the Brꢁnsted basic sites by ZrO2 benefited to depro-
tonation of surface-bound methoxy species, and improved olefin
formation. On the other hand, the modified H-ZSM-5 possessed
shape-selective micropores slightly shrinked after the modifica-
tion, which constrained larger hydrocarbon formation. These
two factors resulted in high selectivity to propylene from direct
conversion of DME.
0.05
H-ZSM-5
12.5%ZrO2/H-ZSM-5
H3PO4/12.5%ZrO2/H-ZSM-5
0.045
0.04
0.035
0.03
0.025
0.02
0.015
0.01
0.005
0
References and Notes
1
2
M. Stocker, Microporous Mesoporous Mater., 29, 3 (1999).
J. F. Haw, W. Song, D. M. Marcus, and J. B. Nicholas,
Acc. Chem. Res., 36, 317 (2003).
¨
3
3.5
4
4.5
5
5.5
6 6.5 7 7.5 8 8.5 9
Pore width /Å
3
4
L. R. M. Martens, K. H. Kuechler, and J. R. Lattner, U. S.
Patent 6,797,851 (2004).
F. A. Wunder and E. I. Leupold, Angew. Chem., Int. Ed.
Engl., 19, 126 (1980).
W. W. Kaeding and S. A. Butter, J. Catal., 61, 155 (1980).
W. J. H. Dehertog and G. F. Froment, Appl. Catal., 71, 153
(1991).
Figure 2. Pore size distributions of modified H-ZSM-5.
H-ZSM-5
12.5 % ZrO2/H-ZSM-5
1.26 mmol/g
1.12 mmol/g
5
6
H3PO4/12.5% ZrO2/H-ZSM-5 1.57 mmol/g
7
8
C. D. Chang, Catal. Rev.—Sci. Eng., 26, 323 (1984).
S. Wilson and P. Barger, Microporous Mesoporous Mater.,
29, 117 (1999).
9 M. Kang, J. Mol. Catal. A: Chem., 150, 205 (1999).
10 C. D. Chang and A. J. Silvestri, J. Catal., 47, 249 (1977).
11 Modified H-ZSM-5 catalysts were prepared as follows. A
300
400
500
600
700
800
900
suspension of H-ZSM-5 (SiO /Al O ¼ 83:7, Sud Chem.)
¨
2
2
3
Temperature/K
.
and ZrO(NO3)2 2H2O was stirred at 353 K for 2 h. Then it
was adjusted to pH 9 using ammonia and continuously stir-
red for 2 h. After filtration and washing, the cake was dried
at 393 K overnight to obtain Zr(OH)4/H-ZSM-5, and then
calcined at 773 K in air for 3 h. A mixture of Zr(OH)4/
H-ZSM-5 and 1 M H3PO4(1 g/10 mL) was stirred for 0.5 h
at room temperature. After filtration, the cake was dried at
393 K overnight and calcined at 773 K in air for 3 h to obtain
H3PO4/ZrO2/H-ZSM-5. Catalysts (20–40 mesh) tests were
conducted in a flow-type fixed-bed quartz reactor. All prod-
ucts were analyzed via on-line gas chromatographs: One
with a thermal conductivity detector (TCD) and a Porapak
N column for N2, DME analysis. Another with a flame
ionization detector (FID), a methanator and a Porapak Q
column for carbon oxides, hydrocarbons, and oxygenates
analysis.
Figure 3. NH3-TPD profiles of modified H-ZSM-5.
ꢀ
modified H-ZSM-5 slightly changed from 5.5 A to 5.3 A. The
ꢀ
pore volume changed from 0.047 mL/A/g to 0.026 mL/A/g
ꢀ
for 12.5%-ZrO2/H-ZSM-5 and to 0.025 mL/A/g for H3PO4/
ꢀ
ꢀ
12.5%-ZrO2/H-ZSM-5, respectively. This suggested that ZrO2
and H3PO4 were well dispersed on the outer and inner surface
of H-ZSM-5, and demonstrated that the slight shrink of pores
of H-ZSM-5 happened after the modification. Figure 3 shows
the NH3-TPD profiles and the corresponding total acid amount
of the catalysts.14 Compared with H-ZSM-5, H3PO4/ZrO2/
H-ZSM-5 showed a higher desorption amount of NH3 (1.57
mmol/g) while ZrO2/H-ZSM-5 a lower amount (1.12 mmol/
g). The stronger acidity distribution of H3PO4/ZrO2/H-ZSM-5
was different from that of H-ZSM-5. It should mainly originate
from Lewis acid sites (related to ZrO2) and Brꢁnsted acid sites
(related to H3PO4) as reported.15
12 XRD (X-ray diffraction) measurement was performed on a
Rigaku XRD instrument with Cu Kꢀ radiation.
For MTO process it is generally accepted that methanol is
firstly dehydrated to DME, and then to light olefins. The light
olefins react to form paraffins, aromatics, naphthenes, and higher
olefins by hydrogen transfer, alkylation and polycondensation.
In order to decrease the selectivity of aliphatics and aromatics
on ZSM-5, reducing the acidity of ZSM-5 is needed.16
From the test and characterization results, the modified H-
ZSM-5, on one hand, possessed adjusted acid property, that is,
weakened acidic strength but increased acidic sites density com-
13 Pore size distribution was measured on a Quantachrome
Autosorb-1 gas sorption system with N2 as adsorbent by
H-K model.
14 After NH3 saturation, NH3 desorbed from the catalyst as
temperature linearly increased from 293 to 873 K.
15 Y. Ikeda, M. Asadullah, K. Fujimoto, and K. Tomishige,
J. Phys. Chem. B, 105, 10653 (2001).
16 M. M. Abdillahi, U. A. El-Nafaty, and A. M. Al-Jarallah,
Appl. Catal., 91, 1 (1992).
Published on the web (Advance View) June 4, 2005; DOI 10.1246/cl.2005.970