Catalytic properties of benzene hydroxylation by TS-1 film reactor and Pd-TS-1 composite membrane reactor

Article abstract of DOI:10.1016/j.cattod.2010.03.034

Titanium silicalite-1 (TS-1) and mesoporous TS-1p were incorporated as catalyst for the direct hydroxylation of benzene to phenol in a palladium membrane reactor. The zeolite also served as intermediate support layer and diffusion barrier against contamin

Full text of DOI:10.1016/j.cattod.2010.03.034

Catalysis Today 156 (2010) 288–294
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Catalytic properties of benzene hydroxylation by TS-1 film reactor and
Pd–TS-1 composite membrane reactor
Xiaobin Wanga, Yu Guoa, Xiongfu Zhanga,∗, Yao Wanga, Haiou Liua,
Jinqu Wanga, Jieshan Qiua, King Lun Yeungb,1
a State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, No. 158 Zhong Shan Road, Dalian, Liao Ning 116012, PR China
b Department of Chemical and Biomolecular Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Kowloon, Hong Kong, SAR, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 14 April 2010
Titanium silicalite-1 (TS-1) and mesoporous TS-1p were incorporated as catalyst for the direct hydrox-
ylation of benzene to phenol in a palladium membrane reactor. The zeolite also served as intermediate
support layer and diffusion barrier against contaminants. Pd–TS-1 and Pd–TS-1p were characterized and
tested for hydrogen permeation before investigating their performance for the benzene direct hydrox-
ylation reaction. The effects of reactor configuration, reaction conditions and catalysts on conversion,
product yield and water production were examined. The phenol yield of Pd–TS-1p was the highest fol-
lowed by Pd–Sil-1 and Pd–TS-1 with Pd–TS-1p and Pd–TS-1 having phenol selectivity higher than 95%.
The optimum reactor configuration and H2/O2 molar feed ratio were identified.
Keywords:
Film
Composite membrane
TS-1
Pd membrane
Catalytic membrane reactor
© 2010 Elsevier B.V. All rights reserved.
1. Introduction
hydrogen peroxide. The framework titanium atom is identified
to be active site for the reaction. It is known to generate Ti per-
Phenol is mainly produced by the energy-intensive cumene pro-
cess and this motivates research into cleaner and more energy
efficient processes. A single step synthesis routes based on H2O2
and N2O as oxidant are attractive [1–4], but considerable eco-
nomic advantages could be gained if oxygen could be used. There
is growing number of reports on phenol production from H2/O2
mixtures over precious metal catalysts [5–7]. However, the oper-
ation is restricted by the flammability of the reaction mixture and
the need for acid solvent. A new synthetic route was introduced
by Niwa et al. [8] using Pd membrane to separate and control
the reaction between H2 and O2 to generate ‘in situ’ H2O2 for
the direct hydroxylation of aromatic compounds. It is a clean and
economically-attractive process. Although the benzene conversion
in the Pd membrane reactor can be high, the hydrogen efficiency
was low and the production rate of water was 500–1000 times
that of phenol [9–12]. Therefore, the phenol productivity must be
improved.
oxo species (Ti-OOH) with H2O2 before reacting with neighboring
hydroxyls to form a stable complex of five-membered ring struc-
ture [18–20]. TS-1 films and membranes were grown on various
supports and used to catalyze various reactions [21–25]. Yeung and
co-workers [26–28] investigated epoxidation and selective oxida-
tion reactions over TS-1 films and membranes in microreactors and
our recent works reported the performance of TS-1 membranes for
phenol and styrene hydroxylation reactions [29,30].
This work explores the use of titanium silicalite-1 (TS-1) zeolites
as catalyst for the direct hydroxylation of benzene to phenol using
oxygen and hydrogen as co-reactants. The zeolite also serves as
intermediate support layer and diffusion barrier for the palladium
membrane in the Pd–TS-1 composite membrane. Since TS-1 zeo-
lite stabilizes reactive oxygen species, it is expected that it would
improve the reaction for the direct hydroxylation of aromatics. The
effect of reactor configuration, reaction conditions and catalysts on
conversion, product yield and water production are examined. A
mechanism is proposed to explain the observed reaction behavior.
Zeolites can catalyze many reactions and several works have
been published reporting the performance of zeolite films and
membranes for reactions and separations [13–17]. The TS-1 zeolite
is an active catalyst for the selective oxidation of organics using
2. Experimental
2.1. Materials
The ␣-Al2O3 tubes from Fo Shan (Guangdong, China) with
inner and outer diameters of 9 and 13 mm respectively and a
length of 75 mm were used for membrane support. The tubes have
pores of 3–5 ␮m and a porosity of 40%. The chemicals for mem-
∗
Corresponding author. Tel.: +86 411 39893605; fax: +86 411 39893605.
E-mail addresses: xfzhang@dlut.edu.cn (X. Zhang), kekyeung@ust.hk
(K.L. Yeung).
1
Tel.: +852 2358 7123; fax: +852 2358 0054.
0920-5861/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.cattod.2010.03.034

X. Wang et al. / Catalysis Today 156 (2010) 288–294
289
Fig. 1. Schematic of experiment apparatus used in this work.
Fig. 2. TEM micrographs of TS-1 zeolite before (a) and after (b) post-treatment; SEM images of TS-1 films before ((c) surface and (e) cross-section) and after ((d) surface and
(f) cross-section) post-treatment.

290
X. Wang et al. / Catalysis Today 156 (2010) 288–294
a sequential immersion in SnCl2 and PdCl2 solutions for 10 times.
Each dipping lasted for 4 min and was followed by a rinsing step
in distilled water. The palladium was deposited from an electro-
less plating solution containing PdCl2, EDTA·2Na, NH3·H2O and
N2H4 at 318 K. The membrane was rinsed with deionized water
and dried in the oven at 393 K. The Pd membranes deposited
on the TS-1 and TS-1p zeolite layer were respectively designated
as Pd–TS-1 and Pd–TS-1p composite membranes. Comparisons
were made with Pd–Sil-1 membrane prepared by depositing 5 ␮m
Pd on 2 ␮m Sil-1. The membranes were examined by KYKY-
2008B scanning electron microscopy (SEM) and characterized by
D/Max 2400 Rigaku X-ray diffractometer with a Cu-K␣ X-ray radi-
ation (XRD, ꢀ = 0.1542 nm). The TS-1 and TS-1p were analyzed
by Bruker EQUINOX55 Fourier transformed infrared spectrome-
ter and JASCO V-550 UV-Vis spectrometer. Nitrogen physisorption
(Autosorb-1) were performed on the powder samples from the
membrane.
2.3. Hydroxylation of benzene to phenol
Fig. 3. XRD patterns of the support (a) and TS-1 film before (b) and after (c) TPAOH
treatment.
The direct hydroxylation of benzene to phenol was performed
with hydrogen and oxygen gases as co-reactants in the membrane
reactor shown in Fig. 1. The hydrogen permeation at different trans-
membrane pressures and temperatures was measured prior to the
reaction. The reaction was conducted at 473 K and the H2 and O2
brane preparation were purchased from Tianjian Kermel Chemical
Reagent Ltd. Co. and included tetraethyl orthosilicate (TEOS), tetra-
butyl orthotitanate (TBOT), palladium chloride (PdCl2), tin chloride
dehydrate (SnCl2·2H2O), Na2EDTA·2H2O (C10H14N2Na2O8·2H2O),
hydrazine (N2H4), benzene (C6H6). The 17 wt.% tetrapropylammo-
nium hydroxide (TPAOH) was prepared in the laboratory.
2.2. Preparation of TS-1 and Pd–TS-1 composite membranes
The detailed procedure for support pretreatment, seeding and
growth of TS-1 films were described in a prior work [29]. The tubes
were end-sealed leaving a 30 mm porous section and a membrane
area of 12.2 cm2. The support was seeded to eliminate unwanted
contribution from the support [31] and to control the growth of
the zeolite [32–36]. The TS-1 was grown from a synthesis solution
of 1 SiO2:0.02–0.03 TiO2:0.18 TPA2O:250 H2O under hydrother-
mal conditions. Post-treatment of TS-1 film (TS-1p) in 0.5 M TPAOH
solution introduced mesopores and increased site accessibility. The
treatment was carried out in an autoclave containing 60 ml TPAOH
solution (17 wt.%) and 42 ml distilled water at 448 K for 24 h. The
prepared sample was then washed with distilled water, dried in an
oven at 383 K and calcined in air at 823 K for 6 h.
The palladium membrane was deposited on the TS-1 film by
electroless plating method [37–41]. The TS-1 was first seeded by
Fig. 4. Nitrogen adsorption/desorption isotherms of TS-1 zeolite before (a) and after
Fig. 5. The FT-IR (a), UV–Vis (b) spectra of the TS-1 zeolite (Ti/Si = 0.02: full line and
(b) post-treatment.
Ti/Si = 0.03: dotted line) before and after TPAOH treatment.

X. Wang et al. / Catalysis Today 156 (2010) 288–294
291
3. Results and discussion
3.1. Supported TS-1 film
Fig. 2 compares the TS-1 before and after treatment in TPAOH.
Ogura et al. [42] reported that ZSM-5 treated with sodium hydrox-
ide exhibited increased inter-grain porosity and numerous surface
cracks and defects. The transmission electron microscope images
in Fig. 2a and b show the TS-1 crystals before and after TPAOH
treatment respectively. The formation of internal pores and voids
are evident in the TS-1p zeolites as shown in Fig. 2b, but the
original size and shape of the crystals (cf. Fig. 2a) remained
unchanged. It is believed that the intra-particle voids are formed by
a dissolution–recrystallization process [43], where the nonuniform
framework composition leads to a preferential etching as shown
in Fig. 2b. The effects of TPAOH treatment of TS-1 films are not
apparent from the SEM images of the film surface (Fig. 2c and d)
and cross-sections (Fig. 2e and f). The film morphology, grain inter-
growth and crystal orientation are identical for the TS-1 and TS-1p
films.
The X-ray diffraction data in Fig. 3 show that TS-1p has weaker
diffraction signal with a significant decrease in peak intensity
of (0 3 3) that could be evident of selective etching. The X-ray
diffraction peak at 2ꢁ = 29.3◦ is generally attributed to titanium
incorporated in MFI framework and is present in both TS-1 and
TS-1p. The nitrogen physisorption isotherms of TS-1 and TS-1p
powders are plotted in Fig. 4. TS-1 displays a Type-I isotherm
(Fig. 4a) that is typical of zeolites, while TS-1p exhibits a H2 hys-
teresis loop from mesopores created after the TPAOH treatment
(Fig. 4b). The FT-IR spectra of the TS-1 and TS-1p display all the
characteristic signals (i.e., 1220, 1100, 800, 550 and 450 cm−1) of
MFI zeolites as well as the peak at 965 cm−1 attributed to substi-
tuted titanium in the zeolite framework. The intensity of the 965
Fig. 6. The flux of reactants through TS-1 film layer before and after TPAOH treat-
ment.
reactants were metered by mass flow controllers. The benzene was
fed to reactor by bubbling nitrogen carrier gas through benzene
saturator. The reaction was monitored by two online gas chro-
matographs (GC-7890T; GC-7890F). The inorganic products were
analyzed by thermal conductivity detector (TCD) with a 13X molec-
ular sieve and a GDX molecular sieve packed columns, while the
organic products were separated by 50 m SE-30 capillary column
and analyzed by the FID.
Fig. 7. SEM images of TS-1 films and Pd–TS-1 composite membrane ((a and b) surface and (c and d) cross-section).

292
X. Wang et al. / Catalysis Today 156 (2010) 288–294
Scheme 1. Modes of reaction scheme with Pd–TS-1 composite membrane reactor.
Table 1
The catalytic performances of benzene hydroxylation on different conditions.
Reaction condition
Benzene conversion (%)
Product distribution (%)
Phenol
Cyclohexane
Cyclohexanone
Dihydroxybenzene
Mode 1
Mode 2
Mode 3
Mode 4
Mode 4-2
Pd–Sil-1
0.69
0
4.26
0.12
5.39
4.10
0
0
100
0
8.79
0
0
0
0
1.68
2.85
0.62
0
72.95
97.15
94.79
62.45
16.58
0
2.61
22.32
0
1.98
15.23
Mode 4-2: is similar with mode 4 except TS-1p instead of TS-1. Pd–Sil-1: is the same as the mode 3.
and 800 cm−1 peak increased after the TPAOH treatment, but the
I965/I800 ratio corresponding remained unchanged [44]. The UV–Vis
spectra in Fig. 5b indicated that TPAOH treatment could reincorpo-
rate some of the extra-framework titanium species (Fig. 5, dotted
line), but did not lead to a change in the tetra-coordinated titanium
(ca. 210 nm).
It is possible to conclude that the TPAOH treatment resulted
mainly in creating mesoporosity in TS-1 without significant chang-
ing the zeolites morphology (cf. Fig. 2). The overall titanium content
remained unchanged after the treatment, but a redistribution of
titanium species was observed. The impact of the treatment is more
clearly seen in the gas permeation data in Fig. 6. The flux of N2, O2
and benzene/N2 in TS-1p is about 100 times higher than TS-1 over
the entire range of trans-membrane pressures studied, which is
consistent with the increased porosity of the TS-1p.
3.2. Pd–TS-1 and Pd–TS-1p composite membranes
Palladium was deposited by electroless plating on the zeo-
lites surface after seeding. Fig. 7 displays the membrane surface
(Fig. 7a and b) and cross-sections (Fig. 7c and d) before and
after palladium deposition. A 4 ␮m thick Pd was deposited on
the 5 ␮m thick TS-1 layer as shown in Fig. 7d. The hydro-
gen and nitrogen permeance of TS-1 at 473 K were respectively
3.3 × 10−4 and 1.81 × 10−4 mol m−2 s−1 Pa−0.5, and after deposition
of Pd a comparable H2 permeance of 4.4 × 10−4 mol m−2 s−1Pa−0.5
was obtained but the Pd–TS-1 displayed a low N2 permeance
of 6.5 × 10−6 mol m−2 s−1 Pa−0.5. The low nitrogen permeance of
Pd–TS-1 suggests low number of defects, while the identical
hydrogen permeance of TS-1 and Pd–TS-1 indicates that trans-
port resistance across the 5 ␮m thick TS-1 is not negligible. The
hydrogen and H2/N2 ideal selectivity were plotted in Fig. 8 as a func-
tion of trans-membrane pressures and temperatures. The Pd–TS-1
gave sufficiently high hydrogen permeance even at 473 K with good
H2/N2 selectivity that is adequate for the current application.
3.3. Direct hydroxylation of benzene to phenol over Pd–TS-1
membrane
The direct hydroxylation of aromatic compounds in palladium
Fig. 8. Hydrogen flux (a) and nitrogen (b) flux H2/N2 ideal selectivity (b) through Pd
membrane as a function of pressure at different temperatures.
membrane reactor is considered an attractive alternative for the
production of phenol from benzene [8–12]. The purpose of this

X. Wang et al. / Catalysis Today 156 (2010) 288–294
293
Fig. 9. The reaction properties of benzene conversion (a), phenol yield (b), hydrogen efficiency (c) and water generation (d) as a function of H2/O2 molar ratio on the different
membrane reactors.
work is to improve hydrogen utilization and product yield by
adding a zeolites catalyst layer to the Pd membrane. The effects
of reactor configuration (see Scheme 1) were examined.
membrane where the TS-1 zeolite was pretreated with TPAOH to
introduce mesopores and decrease the mass transport resistance
in the zeolites layer (cf. Fig. 6). A benzene conversion of 5.4% and
phenol selectivity of 95% were obtained. Fig. 9 plots the benzene
conversion (Fig. 9a), phenol yield (Fig. 9b), hydrogen conversion
(Fig. 9c) and water generation (Fig. 9d) as a function of H2/O2 molar
feed ratio for Pd–TS-1, Pd–TS-1p and Pd–Sil-1 membranes operated
under reactor modes 3 and 4. The benzene conversions of Pd–TS-1
and Pd–Sil-1 operated in reactor mode 3 are comparable and dis-
play a maximum at H2/O2 molar feed ratio of 4.7. This suggests that
the zeolites layer did not participate in the catalytic conversion of
benzene which is also indicated by the plot of phenol yield in Fig. 9b.
This is reasonable as the hydroxylation reaction happens on the Pd
surface in this reactor configuration (see Scheme 1c).
The hydroxylation reaction occurs in the zeolite layer under
reaction mode 4 as shown in Scheme 1d. In this case, the nature
of the zeolite catalyst is important. Slow diffusion in the TS-1 zeo-
lite results in low benzene conversion (Fig. 9a) and phenol yield
(Fig. 9b) regardless of the H2/O2 molar feed ratio. Replacing the TS-1
with TS-1p in Pd–TS-1p composite membrane results in immedi-
ate increase in benzene conversion and phenol yield. The optimum
H2/O2 molar feed ratio is 4 lower than the Pd–TS-1 and Pd–Sil-1
ran under reactor mode 3. This gives strong evidence that TS-1p
layer catalyzed the reaction, which is further supported by hydro-
gen conversion and water generation results in Fig. 9c and d. The
hydrogen conversion is significantly lower for Pd–TS-1p membrane
(ca. 25% at H2/O2 = 4) as compared to Pd–TS-1 and Pd–Sil-1 mem-
branes, which have 27% hydrogen conversion at H2/O2 of 4.7. Water
production is dramatically lower in Pd–TS-1p membrane and is less
than half of Pd–Sil-1 membrane.
3.3.1. Reactor configurations
Reactor mode 1 fed the benzene/nitrogen mixture to the TS-
1 zeolite on the support side, while hydrogen was fed on the
Pd membrane-side as shown in Scheme 1a. This reactor config-
uration produced mainly cyclohexane from the hydrogenation of
benzene over the Pd–TS-1 membrane (Table 1). Reactor mode 2
co-fed benzene/N2 and oxygen to the TS-1 zeolite on the support
side as shown in Scheme 1b. No reaction was observed under this
operating condition and TS-1 is inactive for the direct oxidation
of benzene by molecular oxygen. It is clear from the results that
TS-1 cannot catalyze direct hydroxylation of benzene with either
hydrogen or oxygen gas alone. The reactor mode 3 fed hydrogen
to the TS-1 zeolite on the support side and benzene/O2/N2 mix-
ture to the Pd membrane-side as shown in Scheme 1c. Phenol
was the main product of the reaction (4.26% yield, 72.95% selec-
tivity) with dihydroxybenzene, cyclohexane and cyclohexanone as
by-products (Table 1). The reactor mode 4 fed the benzene/O2/N2
mixture to the TS-1 zeolite on the support side and hydrogen to the
Pd membrane-side (Scheme 1d). This configuration was expected
to led to higher conversion and product yield, but the results show
that for the Pd–TS-1 composite membrane that the conversion
was very low (i.e., 0.1%) although a product selectivity of 97% was
obtained.
3.3.2. H2/O2 molar feed ratio
It was speculated that this is due to slow diffusion of the reac-
tants and product through the TS-1 zeolite. The reactor mode 4
was repeated for a Pd–TS-1p (Mode 4-2 in Table 1) composite
The reaction results indicate that the direct hydroxylation of
benzene requires both hydrogen and oxygen co-reactant. The Pd

294
X. Wang et al. / Catalysis Today 156 (2010) 288–294
membrane can catalyze the conversion of benzene to phenol as
reported by Niwa et al. [8] and the reaction is sensitive to the
H2/O2 molar feed ratio. The optimum H2/O2 molar feed ratio is
governed by membrane permeation rate, reactant composition
and reaction conditions, and changes accordingly. Besides phenol,
water, cyclohexanone, cyclohexane and carbon dioxide are major
by-products when reactions occurred on the Pd membrane as in
the case of Pd–TS-1 and Pd–Sil-1 membrane operated in reactor
mode 3. Benzene conversion to phenol is also catalyzed by TS-1
and TS-1p catalysts according to the reaction results from reactor
mode 4. These catalysts are active for the hydroxylation of aro-
matics by H2O2 as reported in our recent works [29,30]. Superb
selectivity to phenol was obtained from both Pd–TS-1 and Pd–TS-
1p. The TS-1 and TS-1p catalysts also suppressed water production
and improved hydrogen utilization efficiency. The lower transport
resistance in TS-1p is responsible for the higher reaction conversion
and product yield observed in Pd–TS-1p membrane. It is postulated
that the ability of zeolite catalyst to stabilize the reactive species
generated from reactions between permeated hydrogen ions and
molecular oxygen is responsible for the enhanced reaction.
[3] B. Louis, L. Kiwi-Minsker, P. Reuse, A. Renken, Ind. Eng. Chem. Res. 40 (2001)
1454–1459.
[4] L.V. Pirutko, V.S. Chernyavsky, A.K. Uriarte, G.I. Panov, Appl. Catal. A: Gen. 227
(2002) 143–157.
[5] T. Miyake, M. Hamada, Y. Sasaki, M. Oguri, Appl. Catal. A: Gen. 131 (1995) 33–42.
[6] N.I. Kuznetsova, L.I. Kuznetsova, V.A. Likholobov, G.P. Pez, Catal. Today 99
(2005) 193–198.
[7] W. Laufer, W.F. Hoelderich, Chem. Commun. (2002) 1684–1685.
[8] S. Niwa, M. Eswaramoorty, J. Nair, A. Raj, N. Itoh, H. Shoji, T. Namba, F. Mizukami,
Science 295 (2002) 105–107.
[9] G.D. Vulpescu, M. Ruitenbeek, L.L. van Lieshout, L.A. Correia, D. Meyer, P.P.A.C.
Pex, Catal. Commun. 5 (2004) 347–351.
[10] N. Itoh, S. Niwa, F. Mizukami, T. Inoue, A. Igarashi, T. Namba, Catal. Commun. 4
(2003) 243–246.
[11] K. Sato, T. Hanaoka, S. Niwa, C. Stefan, T. Namba, F. Mizukami, Catal. Today 104
(2005) 260–266.
[12] K. Sato, T. Hanaoka, S. Hamakawa, M. Nishioka, K. Kobayashi, T. Inoue, T. Namba,
F. Mizukami, Catal. Today 118 (2006) 57–62.
[13] E.E. McLeary, J.C. Jansen, F. Kapteijn, Micropor. Mesopor. Mater. 90 (2006)
198–220.
[14] S.M. Lai, R. Martin-Aranda, K.L. Yeung, Chem. Commun. (2003) 218–219.
[15] S.M. Lai, C.P. Ng, R. Martin-Aranda, K.L. Yeung, Micropor. Mesopor. Mater. 66
(2003) 239–252.
[16] X.F. Zhang, E.S.M. Lai, R. Martin-Aranda, K.L. Yeung, Appl. Catal. A: Gen. 261
(2004) 109–118.
[17] W.N. Lau, K.L. Yeung, R. Martin-Aranda, Micropor. Mesopor. Mater. 115 (2008)
156–163.
[18] G. Bellussi, A. Carati, M.G. Clerici, G. Maddinelli, R. Millini, J. Catal. 133 (1992)
220–230.
4. Conclusion
[19] M.G. Clerici, P. Ingallina, J. Catal. 140 (1993) 71–83.
[20] P. Wu, T. Komatsu, T. Yashima, J. Phys. Chem. B 102 (1998) 9297–9303.
[21] P. Chen, X.B. Chen, X.S. Chen, H. Kita, J. Membr. Sci. 330 (2009) 369–378.
[22] X.D. Wang, B.Q. Zhang, X.F. Liu, J.Y.S. Lin, Adv. Mater. 18 (2006) 3261–3265.
[23] X.D. Wang, P.P. Zhang, X.F. Liu, B.Q. Zhang, Appl. Surf. Sci. 15 (2007) 544–547.
[24] P. Chen, X.S. Chen, K. Tanaka, H. Kita, Chem. Lett. 36 (2007) 1078–1079.
[25] Q. Zhao, P. Li, D.Q. Li, X.G. Zhou, W.K. Yuan, X.J. Hu, Micropor. Mesopor. Mater.
108 (2008) 311–317.
[26] Y.S.S. Wan, J.L.H. Chau, A. Gavriilidis, K.L. Yeung, Chem. Commun. (2002)
978–979.
[27] Y.S.S. Wan, J.L.H. Chau, A. Gavriilidis, K.L. Yeung, J. Catal. 223 (2004) 241–249.
[28] Y.S.S. Wan, K.L. Yeung, A. Gavriilidis, Appl. Catal. A: Gen. 281 (2005) 285–293.
[29] X.B. Wang, X.F. Zhang, H.O. Liu, K.L. Yeung, J.Q. Wang, Chem. Eng. J. 156 (2010)
562–570.
This work demonstrates that it is possible to improve the direct
hydroxylation of benzene in Pd membrane reactor by incorporat-
ing TS-1 catalyst. Reactor configuration, reactant composition and
reaction conditions are shown to be important to the reaction. Large
intra-crystalline voids and mesopores in TS-1p were obtained by
simply treating the TS-1 with TPAOH. The TS-1p had better cat-
alytic activity for benzene hydroxylation by H2O2 compared to
untreated TS-1 due to improved diffusion and elimination of extra-
framework titanium. The Pd–TS-1 and Pd–TS-1p membrane were
prepared by electroless plating Pd and display high selectivity for
phenol (>95%) compared to Pd–Sil-1 (ca. 65%). Faster diffusion in
mesoporous TS-1p results in higher benzene conversion in Pd–TS-
1p (ca. 5.4%) as compared to Pd–TS-1 (ca. 0.1%). The TS-1 and TS-1p
catalysts suppressed water production and increased hydrogen uti-
lization efficiency. Although the benzene conversion and phenol
yield remain low compared to the literature report, we believe that
this is mainly due to poorer mass transfer in the large diameter
membrane used in this study compared to capillary membranes
employed in prior works [8–13], and indeed the use of MEMS-based
Pd microreactor gave benzene conversion of up to 54% and phenol
yield of 20% [45]. Miniaturization can significantly improve mass
transfer and results in better reaction performance [46–53].
[30] F.R. Qiu, X.B. Wang, X.F. Zhang, H.O. Liu, S.Q. Liu, K.L. Yeung, Chem. Eng. J. 147
(2009) 316–322.
[31] J.L.H. Chau, C. Tellez, K.L. Yeung, K.C. Ho, J. Membr. Sci. 164 (2000) 257–275.
[32] S.M. Lai, L.T.Y. Au, K.L. Yeung, Micropor. Mesopor. Mater. 54 (2002) 63–77.
[33] W.C. Wong, L.T.Y. Au, C. Tellez, K.L. Yeung, J. Membr. Sci. 191 (2001) 143–163.
[34] L.T.Y. Au, W.Y. Mui, P.S. Lau, C. Tellez, K.L. Yeung, Micropor. Mesopor. Mater. 47
(2001) 203–216.
[35] L.T.Y. Au, K.L. Yeung, J. Membr. Sci. 194 (2001) 33–55.
[36] W.C. Wong, L.T.Y. Au, P.P.S. Lau, C. Tellez, K.L. Yeung, J. Membr. Sci. 193 (2001)
141–161.
[37] K.L. Yeung, R. Aravind, J. Szegner, A. Varma, Stud. Surf. Sci. Catal. 101 (1996)
1349–1358.
[38] K.L. Yeung, S.C. Christiansen, A. Varma, J. Membr. Sci. 159 (1999) 107–122.
[39] Y.S. Cheng, K.L. Yeung, J. Membr. Sci. 182 (2001) 195–203.
[40] Y.S. Cheng, K.L. Yeung, J. Membr. Sci. 158 (1999) 127–141.
[41] Y.S. Cheng, M.A. Pen˜a, J.L. Fierro, D.C.W. Hui, K.L. Yeung, J. Membr. Sci. 204
(2002) 329–340.
[42] M. Ogura, S.Y. Shinomiya, J. Tateno, Y. Nara, M. Nomura, E. Kikuchi, M. Mat-
sukata, Appl. Catal. A: Gen. 219 (2001) 33–43.
Acknowledgements
[43] Y.R. Wang, M. Lin, A. Teul, Micropor. Mesopor. Mater. 102 (2007) 80–85.
[44] D.P. Serrano, M.A. Uguina, G. Ovejero, R. Van Grieken, M. Camacho, Micropor.
Mater. 4 (1995) 273–282.
[45] S.-Y. Ye, S. Hamakawa, S. Tanaka, K. Sato, M. Esashi, F. Mizukami, Chem. Eng. J.
155 (2009) 829–837.
The authors gratefully acknowledge financial support pro-
vided by National Natural Science Foundation of China (Grant
No. 20673017), PetroChina Innovation Foundation (Grant No.
2008D-5006-05-06), China National 863 Program (Grant No.
2007AA05Z137) and Universities Science & Research Project of
Liaoning Province Education Department (Grant No. 2009S019).
[46] J.L.H. Chau, Y.S.S. Wan, A. Gavriilidis, K.L. Yeung, Chem. Eng. J. 88 (2002)
187–200.
[47] J.L.H. Chau, K.L. Yeung, Chem. Commun. (2002) 960–961.
[48] J.L.H. Chau, A.Y.L. Leung, K.L. Yeung, Lab Chip 3 (2003) 53–55.
[49] A.Y.L. Leung, K.L. Yeung, Chem. Eng. Sci. 59 (2004) 4809–4817.
[50] S.M. Kwan, K.L. Yeung, Chem. Commun. (2008) 3631–3633.
[51] Y.S.S. Wan, A. Gavriilidis, K.L. Yeung, Chem. Eng. Res. Des. 81 (2003) 753–759.
[52] K.L. Yeung, X.F. Zhang, W.N. Lau, R. Martin-Aranda, Catal. Today 110 (2005)
26–37.
References
[1] M. Stockmann, F. Konietzni, J.U. Notheis, J. Voss, W. Keune, W.F. Maier, Appl.
Catal. A: Gen. 208 (2001) 343–358.
[2] D. Bianchi, R. Bortolo, R. Tassinari, M. Ricci, R. Vignola, Angew. Chem. Int. Ed.
39 (2000) 4321–4323.
[53] K.L. Yeung, S.M. Kwan, W.N. Lau, Top. Catal. 52 (2009) 101–110.