Catalytic performance of metal oxide modified SiMcM-41 catalysts in diphenyl carbonate synthesis

Article abstract of DOI:10.1134/S0023158410030055

Decomposition of CCl₄ into diphenyl carbonate (DPC) was examined over metal oxides modified SiMCM-41. ZnO/SiMCM-41 and Fe₂O ₃/SiMCM-41 showed high activity in DPC synthesis. Although many other metal oxides, such as La2O3, CuO, Al2O3 and alkali or alkaline earth oxide, were success in destruction of CCl₄, they displayed nearly no activity on DPC synthesis. ZnO/SiMCM-41 and Fe₂O₃/SiMCM-41 were characterized by X-ray diffraction (XRD), UV-Raman, ²⁹Si MAS NMR and N₂ adsorption-desorption isotherms, and results showed that ferric and zinc oxide were supported onto SiMCM-41. The well ZnO dispersion in SiMCM-41 channels and the weak electrostatic interaction between chlorine anion and Zn²⁺ play an important role for the high activity of ZnO/SiMCM-41 in decomposition of CCl₄ into DPC. Pleiades Publishing, Ltd., 2010.

Full text of DOI:10.1134/S0023158410030055

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2010, Vol. 51, No. 3, pp. 359–363. © Pleiades Publishing, Ltd., 2010.
Catalytic Performance of Metal Oxide Modified SiMCMꢀ41
Catalysts in Diphenyl Carbonate Synthesis1
K. Su, Z. Li, B. Cheng, Y. Ren, L. Yu, and F. Wang
College of Materials and Chemical Engineering, and Tianjin Key Laboratory of Fiber Modification and Functional Fiber,
Tianjin Polytechnic University, Tianjin, 300160 China
eꢀmail: zhenhuanli1975@yahoo.com.cn, Bowen15@tjpu.edu.cn
Received February 10, 2009
Abstract—Decomposition of CCl4 into diphenyl carbonate (DPC) was examined over metal oxides modified
SiMCMꢀ41. ZnO/SiMCMꢀ41 and Fe2O3/SiMCMꢀ41 showed high activity in DPC synthesis. Although
many other metal oxides, such as La2O3, CuO, Al2O3 and alkali or alkaline earth oxide, were success in
destruction of CCl4, they displayed nearly no activity on DPC synthesis. ZnO/SiMCMꢀ41 and
Fe2O3/SiMCMꢀ41 were characterized by Xꢀray diffraction (XRD), UVꢀRaman, 29Si MAS NMR and N2
adsorptionꢀdesorption isotherms, and results showed that ferric and zinc oxide were supported onto SiMCMꢀ
41. The well ZnO dispersion in SiMCMꢀ41 channels and the weak electrostatic interaction between chlorine
anion and Zn2+ play an important role for the high activity of ZnO/SiMCMꢀ41 in decomposition of CCl4
into DPC.
DOI: 10.1134/S0023158410030055
δ+
Diphenyl carbonate (DPC) is a key raw material [7], and
could be formed from CCl4 decompoꢀ
CCl3
for producing aromatic polycarbonates, and it is
mainly synthesized by using severely toxic phosgene as
a raw material [1], the oxidative carbonylation of pheꢀ
nol [2] and transesterification [3–5]. Recently, the
studies on the chemical degradation of CCl4 by pheꢀ
noxide over ZnCl2 were carried out, and it was found
that the major DPC and minor phenyl salicylate (PS)
were synthesized [6]. In this process PhOCOO– was
sition over Lewis acid [8, 9]. Therefore trichloromeꢀ
thyl phenyl carbonate was formed from CO2 and pheꢀ
noxide in CCl4 over ZnCl2, and DPC can be obtained
through transesterification of trichloromethyl phenyl
carbonate with phenoxide (Scheme 1) [10]. The phosꢀ
gene produces from Cl3COK in situ during the reacꢀ
tion and reacts with phenoxide to produce another
produced from the reaction of CO2 with phenoxide molecular DPC [7].
O
O
O
ZnCl2CCl4
PhOK
OCOK
OCOCCl3
OCO
120°C
OK
(DPC)
O
CO2
OH
OH
OH
O
O
C O
ZnCl2CCl4
PhOK
COK
COCCl3
120°C
(PhOK)
(PS)
Scheme 1. The synthesis of DPC from CO2 and phenoxide in CCl4 over ZnCl2.
Although ZnCl2/SiMCMꢀ41 and FeCl3/SiMCMꢀ be figured out. Herein, the decomposition of CCl4 into
DPC over metal oxide modified MCMꢀ41 catalysts
was carried out, and results showed the well ZnO disꢀ
persion in MCMꢀ41 play an important role in decomꢀ
position of CCl4 into DPC. The electrostatic interacꢀ
tion between chlorine anion and Zn2+ appeared to be
41 displayed high activity in decomposing CCl4 into
DPC, the main reasons for Zn2+ showing higher activꢀ
ity than Fe3+ in dispose of CCl4 into DPC was still not
1
The article is published in the original.
359

360
SU et al.
Table 1. The activity of different catalysts with metal oxide
loaded on SiMCMꢀ41
ASAP2000 (Micromeritics Instrument Co. USA),
respectively. Prior to the measurement, all samples
were outgassed at 200°C and 10–6 Torr over night. The
surface areas of the mesoporous samples were calcuꢀ
lated using the multipleꢀpoint Brunauer–Emmett–
Teller (BET) method, and the pore size distribution
curves of samples were calculated from the adsorption
branch of the isotherms using the Barrett–Joyner–
Halenda (BJH) method. The pore size distributions of
microporous samples were determined from the
adsorption branch of the isotherms employing the regꢀ
ularization method according to density functional
theory (DFT) method. 29SiMAS NMR spectra were
recorded at 79.5 MHz using Varian Infinity Plus specꢀ
trometer. UVꢀRaman spectra were measured on a
homemade VꢀRaman spectrometer [11, 12]. The
244 nm line from a Coherent Innova 300 Fred was
used as the excitation source. The power of the UV
laser lines was less than 2.0 mW. The spectral resoluꢀ
tion was estimated to be 4.0 cm–1. XRD patterns charꢀ
acterization of the catalyst samples were measured on
Selectivity, mol. %
Yield,
mol. %
Catalyst
DPC
PS
Others
Fe2O3/SiMCMꢀ41
NiO/SiMCMꢀ41
ZnO/SiMCMꢀ41
RZnO/SiMCMꢀ41
RFe2O3/SiMCMꢀ41
CdO/SiMCMꢀ41
La2O3/SiMCMꢀ41
Al2O3/SiMCMꢀ41
CuO/SiMCMꢀ41
Y2O3/SiMCMꢀ41
RhO2/SiMCMꢀ41
MgO/SiMCMꢀ41
MnO/SiMCMꢀ41
Cr2O3/SiMCMꢀ41
67.2
33.2
75.4
73.2
66.5
32.7
22.9
30.9
10.4
31.7
25.6
40.8
22.9
30.2
16.8
31.5
10.7
11.8
14.7
25.5
29.4
28.5
14.6
26.5
28.4
59.2
51.4
27.4
16.0
35.3
13.9
15.0
18.8
41.8
47.7
40.7
75.0
41.8
46.0
–
12.1
1.1
21.5
19.7
10.4
0.6
0.3
0.5
0.8
0.5
0.4
a Bruker AXS (Germany) using Cu
K radiation. The
α
0.6
date were recorded from 1
° to 10° and from 10° to 70°
(2 . The Zn content in fresh ZnO/SiMCMꢀ41 and
θ
)
25.7
42.4
0.9
refreshed RZnO/SiMCMꢀ41 was measured on ICP
(TJA Corporation, Astomscan 16).
0.7
The structure of products was confirmed by GCꢀ
MS method (HP5972, capillary column: 30 m SEꢀ30,
0.25 mm inner diameter, and 0.25 μm film thickness).
Quantitative analysis was carried on a gas chromatoꢀ
graph (Agilent 6890N GC with a FID detector, HPꢀ
5/DBꢀ5 capillary column) with cetane as interior
standard.
A definite quantity of catalyst and reactants were
firstly placed in an autoclave of 250 ml volume, the
autoclave was then sealed and flushed with CO2 to
remove any air in it. After that, CO2 was pressurized in
the autoclave to given pressure and the autoclave was
heated to given reaction temperature. After reactions
lasted for certain duration, the autoclave was cooled
down to room temperature. The contents were disꢀ
charged to separate the catalyst by simple filtration.
The loading of metal oxide on the support is 4 mmol/g. Reaction
condition: phenol 30 mmol, K CO 15 mmol, catalyst 1.0 g, CCl
4
2
3
40 ml, PCO = 1.0 MPa, 120°C, 6 h.
2
significantly weak which made chlorine anion much
easier to divorce from zinc cation to make activity sites
recycle.
EXPERIMENTAL
Pure silica MCMꢀ41 was prepared according to the
documented procedure in [10]. Metal oxides were
loaded on supports by the wet impregnation of correꢀ
sponding metal nitrate. After solvent was removed by
evaporation at 80°C for 8 h, the catalysts were calcined
at 550°C for 3 h. RZnO/SiMCMꢀ41 and
RFe2O3/SiMCMꢀ41 are the secondly used catalyst
and they are activated at 550°C for 3 h.
RESULTS AND DISCUSSION
DPC is the main product of the CCl4 transformaꢀ
N2 adsorptionꢀdesorption isotherms of the mesopoꢀ tion, and the main byꢀproduct is PS. The more precise
rous and microporous samples were measured at 77 K stoichiometry of the DPC and PS synthesis was listed
on Micromeritics Tristar 3000 sorptometer and as the following equations:
(I)
4PhOH + 2K2CO3 + CCl4 = 2(PhO)2CO + 4KCl + 2H2O + CO2,
OH
O
(II)
4PhOH + 2K2CO3 + CCl4 =
+ (PhO) CO + 4KCl + 2H O + CO .
2 2 2
C OPh
Some data on the selectivity of DPC formation from CCl4 in reaction are presented in Table 1.
KINETICS AND CATALYSIS Vol. 51
No. 3
2010

CATALYTIC PERFORMANCE OF METAL OXIDE MODIFIED SiMCMꢀ41 CATALYSTS
361
(a)
(b)
1
2
3
4
1
1
2
2
5
3
3
1
2
3
2
4 5
, deg
6
7
10 20 30 40 50 60 70
2θ, deg
200
400
600
800
1000
1200
θ
300
500
700
900
1
1100
Raman shift, cm−
Fig. 2. UVꢀRaman spectra of fresh and refreshed catalysts
with 4 mmol/g loading fresh ZnO/SiMCMꢀ41 ( ),
refreshed ZnO/SiMCMꢀ41 ( ), fresh Fe O /SiMCMꢀ41
), refreshed Fe O /SiMCMꢀ41 (4), and SiMCMꢀ41 (5).
1
Fig. 1. Small angle XRD (a) and extensive XRD (b) specꢀ
trums of different catalysts: —SiMCMꢀ41,
Fe O /SiMCMꢀ41 —ZnO/SiMCMꢀ41.
1
2—
2
2
3
(
3
,
3
2
3
2 3
ference in pore diameter between ZnO/SiMCMꢀ41
and Fe2O3/SiMCMꢀ41 was only 0.1 nm, but this charꢀ
acterized results suggested that ZnO was well dispersed
in the pore channel of SiMCMꢀ41. The well ZnO disꢀ
persion over SiMCMꢀ41 might result in
ZnO/SiMCMꢀ41 showing high activity.
Metal oxides, such as La2O3 [8, 13], CuO [14],
Al2O3 [15] and alkali or alkaline earth oxide [16], were
success in destruction of CCl4, but they displayed
nearly no activity in decomposition of CCl4 into DPC.
However, when ZnO/SiMCMꢀ41 acted as catalyst,
DPC yield amounted to 21.5%, and DPC selectivity
achieved 75.4%. When reaction was carried out over
Fe2O3/SiMCMꢀ41, DPC yield and selectivity
declined to 12.1 and 67.2%, respectively. The used
ZnO/SiMCMꢀ41 was refreshed, and it displayed high
activity for DPC synthesis, which indicated
ZnO/SiMCMꢀ41 could be recycled.
Figure 2 showed UVꢀRaman spectra of the fresh
and refreshed catalysts. In the spectrum of SiMCMꢀ
41, the bands at 490 and 613 cm–1 were assigned to the
three and four siloane rings, and the band at 810 cm–1
was the symmetric stretching mode of the [SiO4] tetraꢀ
hedron [17–20], which was the unit of the MCMꢀ41
framework. Fe2O3/SiMCMꢀ41 exhibited new bands at
524, 1126, and 1152 cm–1. The chemical analysis of
SiMCMꢀ41 and Fe2O3SiMCMꢀ41 showed that the
iron was the only difference in chemical composition
of these samples, and the new bands must be attributed
to the iron species [21]. A new Raman band at
Although ZnO/SiMCMꢀ41 and Fe2O3/SiMCMꢀ
41 displayed high activity in decomposing CCl4 into
DPC, the main reasons for ZnO showing higher activꢀ
ity than Fe2O3 in dispose of CCl4 into DPC was still
not be figured out. Here, the XRD patterns of
ZnO/SiMCMꢀ41 and Fe2O3/SiMCMꢀ41 were listed
in Fig. 1. The peaks (100) intensity at 2° still existed
when ZnO and Fe2O3 were loaded (Fig. 1a), which
suggests that the mesoporous structure of SiMCMꢀ41
1
283 cm⎯ was observed in the spectrum of
ZnO/SiMCMꢀ41 which must be caused by zinc speꢀ
1
existed in ZnO/SiMCMꢀ41 and Fe2O3/SiMCMꢀ41
.
cies. The bands at 283, 491, 810, and 1046 cm⎯ still
No metal oxide crystallite diffraction peaks were
detected and only one strong SiO2 diffraction peak
appeared in extensive XRD spectra of different cataꢀ
lysts (Fig. 1b), which indicated the metal oxides were
well dispersed on SiMCMꢀ41.
existed in refreshed ZnO/SiMCMꢀ41, which implied
that the structure of SiMCMꢀ41 was not destroyed
1
during reaction. However, the band at 283 cm⎯ in
refreshed ZnO/SiMCMꢀ41 declined greatly, which
The physical properties of catalysts are listed in
Table 2. The properties of catalysts
Table 2. The BET surface (
Fe2O3/SiMCMꢀ41 was 546 and 703 m2/g, respecꢀ
tively. The sequence of pore volume ( ) of catalyst
declined as follows: SiMCMꢀ41 > Fe2O3/SiMCMꢀ41
ZnO/SiMCMꢀ41. The fact that the pore diameter (
S) of ZnO/SiMCMꢀ41 and
Catalyst
SiMCMꢀ41
S
, m2/g
ν
, cm3/g
d, nm
ν
>
)
820
546
703
1.06
0.63
0.76
2.7
2.6
2.7
d
of Fe2O3/SiMCMꢀ41 did not decrease could sugꢀ
gested that most Fe2O3 did not fill the pore channels of
SiMCMꢀ41, but rather cover the pore mouth. The difꢀ
ZnO/SiMCMꢀ41
Fe2O3/SiMCMꢀ41
KINETICS AND CATALYSIS Vol. 51
No. 3
2010

362
SU et al.
41. 29Si NMR spectra could be deconvolved into three
Gaussian peaks with chemical shift of –113.0(Q4),
100.1(Q3) and –90.31(Q2) ppm corresponding to
⎯
Si(OSi)4, Si(OSi)3OH, and Si(OSi)2–(OH)2, respecꢀ
tively [22]. These NMR peaks were used to monitor
the relative concentration variety of the silicon speꢀ
cies. After ZnO and Fe2O3 were loaded on the surface
of SiMCMꢀ41, the intensities of Si(O–Si)3OH and
Si(O–Si)2(OH)2 signals decreased while that of Q4 sigꢀ
1
nal increased by the contribution of Si(OSi)4–
n(OM)n
2
species ( = 0, 1, 2) [23], which suggested that a part
n
of ferric and zinc ions were fixed onto SiMCMꢀ41
framework by the reaction between metal ions and silꢀ
anol groups [24].
3
The Zn2+ and Fe3+ content on the surface of fresh
and refreshed catalyst were summarized in Table 3.
The characterized results showed that a small amount
of Zn species leached out of ZnO/SiMCMꢀ41 during
reaction. The loss of Zn species belonged to no fixed
Zn species which resulted in the appearance of UVꢀ
Raman band at 283 cm⎯1. However, nearly no Fe3+
species was leached out from Fe2O3/SiMCMꢀ41. The
refreshed ZnO/SiMCMꢀ41 displayed high activity for
DPC synthesis, which indicated the leached Zn speꢀ
cies in solution has little effect on decomposition of
−
220
−
180
−
140
−
100
−
60
−
20
−
200
−
160
−120
, ppm
−
80
−
40
δ
29
Fig. 3. Solid state Si MAS NMR spectra of catalysts
SiMCMꢀ41 ), ZnO/SiMCMꢀ41 ), and
Fe O /SiMCMꢀ41 ).
(
1
(2
(
3
2
3
displayed a part of Zn species might leach out of
ZnO/SiMCMꢀ41.
Figure 3 shows 29Si MAS NMR spectra of CCl4 into DPC, and supported ZnO was main activity
SiMCMꢀ41, ZnO/SiMCMꢀ41, and Fe2O3/SiMCMꢀ site for DPC synthesis.
KCl
Zn O Zn
CCl4
Zn O Zn
Cl ᮎ
Zn O Zn
Cl
⊕
C
K
Cl
Cl
Cl
O
CO2 PhOK
PhOCOCCl3
Zn O Zn
Clᮎ
Zn O Zn
K
O
K
O
C
Cl
⊕
C
O C
OPh
C
O
Cl
Cl
Cl
OPh
Cl
Cl
Cl
Scheme 2. Reaction mechanism for DPC synthesis over ZnO/SiMCMꢀ41.
In order to figure out the main reasons for Zn2+
form PhOCOOCCl3 which further reacted with
showing higher activity than Fe3+ in decomposition of another PhOK to form DPC [25]. After transesterifiꢀ
CCl4 into DPC, a possible reaction path was suggested
+
cation of PhOCOOCCl3 with phenoxide, the phosꢀ
gene produces from Cl3COK in situ during the reacꢀ
in Scheme 2. PhOCOOK interacted with
to
CCl3
KINETICS AND CATALYSIS Vol. 51
No. 3
2010

CATALYTIC PERFORMANCE OF METAL OXIDE MODIFIED SiMCMꢀ41 CATALYSTS
363
Table 3. The metal content on the surface of fresh and reꢀ
freshed catalyst
3. Li, Z., Cheng, B., Gu, Y., Su, K., et al., J. Mol. Catal.
A: Chem., 2008, vol. 289, p. 100.
4. Ishii, H., Goyal, M., Ueda, M., et al., Catal. Lett.
,
Metal content, wt %
Catalyst
2000, vol. 65, p. 57.
5. Ma, X., Wang, S., Gong, J., et al., J. Mol. Catal. A:
Chem., 2004, vol. 222, p. 183.
fresh
used
ZnO/SiMCMꢀ41
Fe2O3/SiMCMꢀ41
19.7
27.3
16.3
26.2
6. Li, Z., Qin, Z., Zhu, H., and Wang, J., Chem. Lett.
2006, vol. 7, p. 784.
,
7. Lindsey, A.S. and Jeskey, H., Chem. Rev., 1957, vol. 57,
p. 583.
8. Van der Avert, P., Podkolzin, S.G., Manoilova, O.,
tion and reacts with phenoxide to produce another
molecular DPC. The electrostatic interaction between
chlorine anion and Zn2+ appeared to be significantly
weaker, which resulted in chlorine anion being easier
to divorce from Zn2+ than from Fe3+, i.e., the electroꢀ
static interaction between chlorine anion and Zn2+
appeared to be significantly weak which made chlorine
anion much easier to divorce from zinc cation to make
activity sites recycle.
et al., Chem. Eur. J., 2004, vol. 10, p. 1637.
9. Farrokhnia, A., Sakakini, B., and Waugh, K.C., Catal.
Lett., 2001, vol. 76, p. 241.
10. Su, K., Li, Z., Cheng, B., and Ren, Y., Catal. Commun.
2008, vol. 9, p. 1666.
,
11. Li, C., Xiong, G., Xin, Q., Liu, J., et al., Angew. Chem.,
Int. Ed. Engl., 1999, vol. 38, p. 2220.
12. Xiong, G., Li, C., Feng, Z., et al., J. Catal., 1999,
vol. 186, p. 234.
Thus, ZnO/SiMCMꢀ41 displayed higher activity
than Fe2O3/SiMCMꢀ41 in DPC synthesis because
chlorine anion was easier to divorce from Zn2+ than
from Fe3+ to keep catalyst from deactivation. Metal
ions were bound to the wall of SiMCMꢀ41, which
resulted in leaching of a small amount of metal species
from catalyst during reaction. The well ZnO disperꢀ
sion in SiMCMꢀ41 pore channels also played an
important role in decomposition of CCl4 into DPC.
13. Van der Avert, P. and Weckhuysen, B.M., Angew. Chem.,
Int. Ed. Engl., 2002, vol. 41, p. 4730.
14. Chien, Y., Wang, H.P., and Yang, Y.W., Environ. Sci.
Technol., 2001, vol. 35, p. 3259.
15. Khaleel, A. and Dellinger, B., Environ. Sci. Technol.
2002, vol. 36, p. 1620.
,
16. Weckhuysen, B.M., Mestal, G., Rosynek, M.P., et al.,
J. Phys. Chem. B, 1998, vol. 102, p. 3773.
17. Kornatowski, J., Wichterlova, B., Jitkovsky, J., et al.,
J. Chem. Soc., Faraday Trans., 1996, vol. 92, p. 1067.
18. Dutta, P.K. and Puri, M., J. Phys. Chem., 1987, vol. 91,
ACKNOWLEDGMENTS
p. 4329.
19. Dutta, P.K., Rao, K.M., and Park, J.Y., J. Phys. Chem.
1991, vol. 95, p. 6654.
20. Kornatowski, J., Sychev, H., Kuzenkov, S., et al.,
J. Chem. Soc., Faraday Trans., 1995, vol. 91, p. 2217.
21. Li, Y., Feng, Z.C., Lian, Y.X., et al., Microporous Mesoꢀ
porous Mater., 2005, vol. 84, p. 41.
,
The authors are grateful for the financial support of
the Natural Science Foundation of Tianjin Technoꢀ
loge and Science Committee (09JCYBJC03000) and
the Natural Science Foundation of Tianjin Education
Committee (20070607).
22. Chao, K.J., Wu, C.N., Chang, H., et al., J. Phys. Chem.
B
, 1997, vol. 101, p. 6341.
REFERENCES
23. Moudrakovski, I.L., Sayari, A., Ratcliffe, C.I., et al.,
J. Phys. Chem., 1994, vol. 98, p. 10895.
24. Li, C., J. Catal., 2003, vol. 216, p. 203.
1. Shaikh, A.G. and Sivaram, S., Chem. Rev., 1996,
vol. 96, p. 951.
2. Yasuda, H., Maki, N., Choi, J.C., and Sakakura, T., 25. Li, Z. and Qin, Z.F., J. Mol. Catal. A: Chem., 2007,
J. Organomet. Chem., 2003, vol. 682, p. 66. vol. 264, p. 255.
KINETICS AND CATALYSIS Vol. 51
No. 3
2010