Liquid-phase catalytic oxidation of p-chlorotoluene to p-chlorobenzaldehyde over manganese oxide octahedral molecular sieves

Article abstract of DOI:10.1016/j.apcata.2013.07.015

Manganese oxide octahedral molecular sieves (OMS-2) show much higher catalytic activities in the liquid-phase catalytic oxidation of p-chlorotoluene than MnO₂, Mn₃O₄, Mn(OAc)₂ as well as than some other transition metal oxides such as Co₂O₃, V₂O₅ and Fe₂O₃. Based on temperature-programmed desorption and thermogravimetric results, it is deduced that the catalytic activity of OMS-2 can be ascribed to the abundance and mobility of lattice oxygen. We investigated the effects of reaction temperature, reaction time, catalyst amount, and initial water amount on the reaction. Under optimal reaction conditions, p-chlorotoluene conversion and p-chlorobenzaldehyde selectivity are respectively 86.0% and 68.7%. The catalysts can be easily recovered by centrifugation, and reused. According to the results of X-ray diffraction and N₂ adsorption/desorption analyses, there is no significant change in major characteristics of the catalyst after 4 cycles of reaction.

Full text of DOI:10.1016/j.apcata.2013.07.015

Applied Catalysis A: General 467 (2013) 117–123
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Liquid-phase catalytic oxidation of p-chlorotoluene to
p-chlorobenzaldehyde over manganese oxide
octahedral molecular sieves
Yi-Qiang Deng , Teng Zhang , Chak-Tong Aua,b, Shuang-Feng Yina,∗
a
a
a
State Key Laboratory of Chemo/Biosensing and Chemometrics, College of Chemistry and Chemical Engineering, Hunan University, Changsha 410082,
Hunan, China
b
Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Hong Kong
a r t i c l e i n f o
a b s t r a c t
Article history:
Manganese oxide octahedral molecular sieves (OMS-2) show much higher catalytic activities in the liquid-
phase catalytic oxidation of p-chlorotoluene than MnO2, Mn3O4, Mn(OAc)2 as well as than some other
transition metal oxides such as Co2O3, V2O5 and Fe2O3. Based on temperature-programmed desorption
and thermogravimetric results, it is deduced that the catalytic activity of OMS-2 can be ascribed to the
abundance and mobility of lattice oxygen. We investigated the effects of reaction temperature, reaction
time, catalyst amount, and initial water amount on the reaction. Under optimal reaction conditions,
p-chlorotoluene conversion and p-chlorobenzaldehyde selectivity are respectively 86.0% and 68.7%. The
catalysts can be easily recovered by centrifugation, and reused. According to the results of X-ray diffraction
and N2 adsorption/desorption analyses, there is no significant change in major characteristics of the
catalyst after 4 cycles of reaction.
Received 27 February 2013
Received in revised form 16 June 2013
Accepted 8 July 2013
Available online xxx
Keywords:
Manganese oxide
OMS-2
p-Chlorotoluene
p-Chlorobenzaldehyde
Catalytic oxidation
Lattice oxygen
© 2013 Elsevier B.V. All rights reserved.
1
. Introduction
in gas phase (with O ) or in liquid phase (with H O or
2
2
2
O ). For example, 28% of p-chlorotoluene was converted to
2
◦
The selective oxidation of p-chlorotoluene to p-
p-chlorobenzaldehyde with 61.6% selectivity at 325 C using gas-
chlorobenzaldehyde is important because p-chlorobenzaldehyde
is an essential intermediate in the production of dyes, pharma-
ceutical drugs, optical brighteners, and agricultural chemicals [1].
The commercial production of p-chlorobenzal chloride is mainly
by side-chain chlorination of p-chlorotoluene followed by acid
hydrolysis [2]. The classical method to produce aromatic aldehyde
involves multi-steps that are conducted non-catalytically in liquid
phase, and falls short of the principle of sustainable chemistry
phase O2 over Fe/Mo/borosilicate molecular sieves [6], similar
values (25% conversion and 66% selectivity) at 400 C was reported
◦
in a US patent when VCs0.5FeOx was used as catalyst [7]. Like-
wise, Bautista et al. described the preparation, structural and
acid characterization of vanadium oxides supported on TiO2-
sepiolite and sepiolite, as well as their behavior in selective
oxidation of chloro- and methoxy-substituted toluenes [8,9], and
obtained 13% p-chlorobenzaldehyde yields when TiO -sepiolite
2
[
3]. In industry, soluble acetates of transition metals are used as
supported vanadium oxides were used as catalysts [3]. On the
other hand, it is easy to operate liquid-phase oxidation and to
achieve high selectivity under relatively mild reaction conditions.
catalysts but both conversion and selectivity are unsatisfactory.
Furthermore, some of the processes are polluting and expensive
mainly because the separation of soluble catalyst is always tedious
and costly [4,5].
Since solid catalysts have advantages over soluble cata-
lysts in terms of recovery and stability, efforts are put in to
develop heterogeneous catalysts for the generation of aromatic
aldehydes by selective oxidation of alkylaromatic hydrocarbons
Using H O2 as oxidants and vanadium silicate molecular sieves
2
as catalyst, Singh and Selvam observed p-chlorotoluene conver-
sion of 13.4% and p-chlorobenzaldehyde selectivity of 65.0% [10]. It
was reported that similar p-chlorobenzaldehyde selectivity could
be attained over a supported Co(II) catalyst together with a sup-
ported N-hydroxyphthalimide promoter [11], but the conversion
of p-chlorotoluene was only 11.0%. In these cases, the initial oxi-
dation products are often more susceptible to oxidation than the
starting material. Once a methyl group is attacked, it is likely to
be oxidized to carboxylic acid. While such reactions readily give
p-chlorobenzoic acids in high yields, it is rather difficult to stop the
∗ Corresponding author. Tel.: +86 731 88821171; fax: +86 731 88821171.
E-mail address: sf yin@hnu.edu.cn (S.-F. Yin).
0
926-860X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2013.07.015

1
18
Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
Scheme 1. Liquid-phase catalytic oxidation of p-chlorotoluene with O2 over OMS-2.
◦
reactions at the aldehyde stage. To convert p-chlorotoluene catalyt-
ically giving aldehyde as the main product (i.e. with high aldehyde
selectivity) using various oxidants is a big challenge, and it is still
an attractive research area [3].
5–60 range. The surface area of OMS-2 was measured by the
Brunauer–Emmett–Teller (BET) method on a Tristar 3000 instru-
◦
ment; before each measurement, the sample was heated to 300 C
and kept at this temperature for 5 h. The BET specific surface
area (SBET) was calculated from the adsorption data in the rela-
tive pressure range of 0.04–0.20. External surface area (SEXT) was
estimated by the t-plot method [20]. Thermal stability of cata-
lysts was studied by means of thermogravimetric analysis (TGA,
PerkinElmer Diamond TG/DTA instrument) with ca. 10 mg of sam-
ple being heated from 30 to 850 C (ramp rate = 10 C/min) under a
N2 atmosphere. H2 temperature-programmed reduction (H -TPR)
and temperature-programmed desorption of O2 (TPD-O ) exper-
With special electronic configurations and being variable in
valence state, transition metal oxides show high catalytic activity
for selective oxidation of hydrocarbons [12,13]. Manganese oxide is
technologically important in catalytic and electrochemical indus-
tries [14,15]. Nanoscale manganese oxide octahedral molecular
sieve (OMS-2) is a prominent oxidation catalyst under thermal
conditions [16]. It has a 2 × 2 tunnel structure and has a chemi-
cal composition of KMn O , with charge-balancing K+ ions and
◦
◦
2
8
16
2
H O residing in the tunnels [17]. Such structure leads to interesting
2
iments were conducted over a Micromeritics 2920 II apparatus
physicochemical properties such as porosity and high adsorption
capacity that are commonly related to good catalytic performance
using a thermal conductivity detector (TCD). For TPD-O , a catalyst
2
◦
(100 mg) was heated in He flow (20 mL/min) at a rate of 10 C/min
◦
[
17]. The unique redox activity of OMS-2 is attributed to the pres-
from 50 to 900 C. Before TPD-O analysis, the catalyst was degassed
2
3
+
4+
ence of Mn and Mn ions, the long and open structure, and the
formation of OH groups on the surface [18].
in He flow for 2 h. H -TPR experiment was carried out with the
2
sample kept under a 20 mL/min flow of 10% H in Ar. The temper-
2
◦
◦
In this paper, we report for the first time the liquid-phase cat-
alytic oxidation of p-chlorotoluene to p-chlorobenzaldehyde with
O2 in a reflux system using manganese oxide octahedral molecular
sieve as catalyst and hydrogen bromide as reaction initiator (see
Scheme 1). Adopting this atmospheric system, we achieve high p-
chlorotoluene conversion (86.0%) and good p-chlorobenzaldehyde
selectivity (68.7%). We believe that the process is practical for the
conversion of aromatic hydrocarbons to the corresponding alde-
hydes using O2 as oxidant.
ature was increased from 50 to 800 C (10 C/min). A cold trap was
installed before the TCD to stop H O interference.
2
2.4. Catalytic reactions
All experiments were carried out in a three-necked round-
bottom flask equipped with a reflux condenser. Oxygen was
introduced into the solution at atmospheric pressure at a desired
flow rate, and the flask was placed in an isothermal paraffin oil bath
with a magnetic stirrer. In a typical reaction, 1 mL p-chlorotoluene,
2 mL hydrobromic acid–water mixture, 10 mL acetic acid (as sol-
vent), and 50 mg catalyst were place in the flask. The OMS-2 catalyst
2
. Experimental
◦
2.1. Materials
was heated at 110 C overnight before being used. With stirring, the
reaction system was heated to a desired temperature. Then oxy-
gen (50 mL/min) was introduced into the bottom of the reaction
mixture. After the reaction was over, the catalyst was removed
by means of centrifugation, and the products were dissolved in
acetonitrile and analyzed by gas chromatography (GC). To find
out whether there was leaching of catalyst into the solution, the
solid substance was removed after 4 h of reaction by centrifuga-
tion, and the reaction was allowed to continue for another 10 h.
The products were identified by a Shimadzu GCMS-QP2010 ultra
mass spectrometer and quantified by Shimadzu GC2014 chromato-
graph equipped with a SFID1 detector and a RTX-1 capillary column
The chemical reagents were of analytical grade and were used
without further purification. No impurities were found in p-
chlorotoluene by GC analysis. Pure gaseous O2 was used as the
oxygen source.
2
.2. Catalyst preparation
OMS-2 was prepared by refluxing a mixture of potassium per-
manganate and manganese sulfate in an acidic medium according
to procedures described by Makwana et al. [19]. A 0.4 M solution
(30 m × 0.25 mm × 0.25 ␮m). The internal standard method was
of KMnO (6.5 g in 113 mL of deionized water) was added to a mix-
4
employed with biphenyl being the internal standard. The outlet
gases were also analyzed on an Agilent 6890 N GC with TCD and
Agilent GS-GASPRO capillary column (30 m × 0.32 mm × 0.25 ␮m).
We detected no formation of gas products.
ture of a 1.75 M solution of MnSO ·H O (9.9 g in 34 mL of deionized
4
2
water) and 3.4 mL of concentrated HNO (68 wt%). The resulting
3
solution with black precipitate was stirred vigorously (600 rpm)
◦
and subject to reflux at 100 C for 24 h. Then the precipitate was
filtered out and washed with deionized water until neutral pH and
◦
dried overnight at 110 C.
3. Results and discussion
2.3. Catalyst characterization
3.1. Catalyst characterization
Powder X-ray diffraction (XRD) experiment was conducted
The XRD patterns of as-synthesized and used (four times) OMS-
2 are shown in Fig. 1 The OMS-2 exhibits a pattern of tetragonal
cryptomelane-type corresponding to KMn O (JCPDS 29-1020).
on a Bruker D8 Advance diffractometer using Cu K␣ radiation.
◦
−1
in the
The data were recorded at a scan rate of 0.02 (2ꢀ) s
8
16

Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
119
1
00
(
a)
9
5
(
(
b)
c)
(
b)
a)
90
8
8
5
(
0
5
15
25
35
45
55
65
1
00
200
300
400
500
600
700
800
2
Theta (degree)
Temperature (ºC)
Fig. 1. X-ray diffraction patterns of (a) fresh OMS-2, and (b) used (4 cycles) OMS-2.
Fig. 2. TGA profiles of (a) commercial Mn3O4, (b) OMS-2, and (c) commercial MnO2.
There is no obvious difference between the patterns of the fresh
and used catalysts. It is hence deduced that there is no change
in catalyst structure during the reaction. Furthermore, there is
no impurity phase observed over the two samples. The scanning
electron micrographs of fresh and used catalyst show a fibrous
needle-like morphology, typical of OMS-2 materials (Fig. S1).
Listed in Table 1 are the textural properties of the nanomateri-
obvious weight loss of about 5.8% is observed over MnO2, mainly
due to the removal of chemisorbed water [24]. On the contrary, the
weight loss in the 200–400 C range observed over OMS-2 material
◦
is only ca. 0.65%. The results suggest that compared with MnO2,
OMS-2 is more hydrophobic and the adsorption of organic com-
pounds is more favorable on OMS-2. The major weight loss of
2
◦
als. The BET surface area of fresh OMS-2 is 71.8 m /g, slightly higher
OMS-2 and MnO2 occurs between 400 and 680 C and there is a
2
◦
than that of used catalyst (70.5 m /g). The SEXT values are very close
less remarkable one above 680 C (Figs. 2(b) and (c)); the former
to the SBET values, indicating that the measured surface area is pre-
dominantly external. The micropore volume of the materials only
contributes to less than 1% of the total pore volume. It is noted that
similar findings were reported by Tang et al. [21] and Luo et al.
is due to the reduction of MnO2 to Mn2O3 and the latter is due to
the transformation of Mn2O3 to Mn3O4 [22]. It is worth pointing
out that the weight losses above 680 C occur at different tempera-
tures with that of OMS-2 lower than that of MnO2 by ca. 90 degrees.
Referring to the small weight loss of Mn3O4 (about 0.06%, Fig. 2a), it
is apparent that there is thermal decomposition of MnO2 and OMS-
2 to Mn3O4. The weight losses due to release of lattice oxygen from
OMS-2 and MnO2 are 7.55% and 10.38%, respectively. Clearly, there
are plenty of lattice oxygen species in OMS-2 and MnO2 that would
become mobile upon thermal excitation.
◦
[
22]. According to the adsorption isotherms of adsorbates reported
by Wang et al. [23], the average crystalline pore openings of cryp-
tomelane materials is about 0.46 nm. Nonetheless, it is plausible
that the effective pore openings of the material are in the range
of 0.265–0.330 nm. It is known that the dynamic diameter of N2 is
0.326 nm. In other words, a great majority of N2 are not adsorbed
inside the microporous tunnels of cryptomelane.
The O2-TPD profiles of commercial Mn3O4, OMS-2, and com-
mercial MnO2 are shown in Fig. 3A There is only one small peak at
530 C in the profile of Mn3O4, which is attributable to the desorp-
The lattice oxygen content of OMS-2 was studied by thermo-
gravimetric analysis. Fig. 2 shows the TGA curves of commercial
Mn O , OMS-2, and commercial MnO . Generally, weight losses of
◦
tion of chemisorbed oxygen species (Fig. 3A(a)). Over OMS-2, there
3
4
2
◦
◦
◦
this kind in the 30–850 C range can be divided into four categories:
I) 30–200 C attributable to desorption of physisorbed water, (II)
00–400 C attributable to desorption of chemisorbed water and
oxygen, (III) 400–680 C ascribable to release of lattice oxygen in
are two major peaks (one at 490 C and the other at 670 C), and
◦
◦
(
2
there is a small shoulder at 320–460 C (Fig. 3A(b)). According to
◦
Genuino et al. [25] and Santos et al. [26], the shoulder peak can
be ascribed to the desorption of chemisorbed oxygen, whereas the
◦
◦
◦
the “MnO → Mn O ” process, and (IV) 680–850 C ascribable to
peaks at 490 and 670 C can be attributed to the desorption of lat-
2
2
3
the evolution of O in the “Mn O → Mn O ” process [24,25]. As
tice oxygen due to the conversion of cryptomelane to Mn2O3 and
that of Mn2O3 to Mn3O4, respectively, in agreement with the TGA
results (Fig. 2b). Over commercial MnO2, two major peaks at 520
2
2
3
3
4
shown in Fig. 2a, Mn O4 only displays a small weight loss (about
3
◦
0
.06% between 30 and 200 C due to removal of physisorbed water)
◦
◦
within the entire temperature range, indicating that it is thermally
stable. As for MnO2 and OMS-2, the contents of physisorbed water
are ca. 3.85% and 1.64%, respectively. In the 200–400 C range, an
(more intense) and 770 C, and a small one at 580 C are observed
◦
(Fig. 3A(c)). The peak at 520 C is ascribed to desorption of surface
oxygen species and labile oxygen species, about 30 C higher than
that of OMS-2 (at 490 C). The other two peaks are due to desorption
◦
◦
◦
of lattice oxygen [25]. Comparing the O -TPD profiles of MnO2 and
2
Table 1
OMS-2, the oxygen species of the former are thermally more stable
than that of the latter. Luo et al. reported that for cryptomelane-
type octahedral molecular sieves, desorption of oxygen could lead
to the formation of oxygen vacancies which might be active sites for
catalytic oxidation [27]. Wang and Li demonstrated that the cata-
lysts with weaker Mn O bonds are usually higher in lattice defects
and labile lattice oxygen, and exhibit better performance [28]. It
was confirmed that labile lattice oxygen has a promotional effect
Physical characteristics of catalysts.
2
3
Sample
Particle
Surfacearea(m /g)
Porevolume(cm /g)
sizea (nm)
SBETb
SEXTb
Vtotalc
Vmicc
Fresh OMS-2
Used OMS-2
11.9
11.3
71.8
70.5
67.2
66.7
0.359
0.359
0.003
0.002
a
Particle size: calculated from Scherrer’s equation.
SBET: BET surface area. SEXT: external surface area.
Vtotal: total pore volume. Vmic: micropore volume.
b
c
on catalytic activity [27,28]. Based on the TGA and O -TPD results, it
2

1
20
Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
(
A)
(B)
(
b)
c)
(
c)
(
(
(
b)
a)
(a)
200
400
600
800
100
200
300
400
500
600
700
Temperature (ºC)
Temperature (ºC)
Fig. 3. (A) O2-TPD and (B) H2-TPR profiles of (a) commercial Mn3O4, (b) OMS-2, and (c) commercial MnO2.
is deduced that OMS-2 has high provision of lattice oxygen species
that are dislodged from the framework.
p-chlorotoluene conversion reaches 49.7% and 41.1%, respectively,
but selectivity to p-chlorobenzaldehyde is only 1.1% and 25.7%.
Depicted in Fig. 3B are the H -TPR profiles of commercial Mn O ,
In addition, the p-chlorotoluene conversion over Mn O4 is only
2
3
4
3
OMS-2, and commercial MnO . In the case of Mn O , there is only
26.5% while that over commercial MnO2 is 54.5%, but selectiv-
ity to p-chlorobenzaldehyde is relatively low (44.1%). However,
using the synthesized OMS-2, we achieve a p-chlorotoluene con-
version of 86.0%, indicating that the as-synthesized OMS-2 is active
for p-chlorotoluene oxidation in the adopted liquid-phase cat-
alytic oxidation system. The selectivity to p-chlorobenzaldehyde
and that to p-chlorobenzoic acids are 68.7% and 31.3%, respec-
tively. The observations suggest that the OMS-2 is an appropriate
catalyst for the p-chlorotoluene aerobic oxidation system, and the
as-synthesized OMS-2 shows better catalytic activity than com-
mercial MnO2 and Mn O . Therefore, OMS-2 catalyst was adopted
2
3
4
◦
one weak band at 410 C corresponding to the reduction of Mn O
3
4
to MnO. Over OMS-2, there is a broad profile of two overlapping
peaks, corresponding to a two-step reduction process. Assuming
that MnO is the final state in the reduction of OMS-2 Mn species
[
24], the component at lower temperature could be assigned to the
reduction of MnO /Mn O to Mn O , and that at higher temper-
2
2
3
3
4
ature to the reduction of Mn O to MnO [25]. As for MnO , there
3
4
2
◦
◦
is one main band at 280 C and a weak band at 400 C. Accord-
ing to the results of Wang et al. [29] and Stobbe et al. [30], the
reduction process could be reasonably divided into two steps: (I)
3
4
4+
3+
3+
2+
Mn → Mn and (II) Mn → Mn . It is obvious that compared
in the rest of the study to investigate the effects of reaction condi-
tions on catalytic activity.
with MnO and Mn O , OMS-2 can be reduced at much lower tem-
2
3
4
peratures. In other words, the Mn O bonds of OMS-2 are relatively
weaker and the Mn species in OMS-2 can be reduced more eas-
ily. The hydrogen consumption was calculated based on the area
of the reduction peaks. Theoretically, the H2 consumptions for the
reduction of MnO2 to Mn O4 and that of Mn O4 to MnO are 7.67
That catalytic activity is closely related to the physical and chem-
ical properties of catalysts is known. Taking the TPD and TGA results
into account, the catalytic activity of OMS-2 could be correlated
with the nature of lattice oxygen. The abundance of lattice oxygen
could be a reason why OMS-2 is high in p-chlorotoluene conversion.
In addition, the catalytic activity can be attributed to oxygen mobil-
ity: the higher the oxygen mobility, the better the activity [25,29].
With desorption and reduction of oxygen occurs at temperatures
3
3
and 3.83 mmol/g, respectively. In the present study, the total H2
consumption for OMS-2, MnO2 and Mn O4 are respectively 10.39,
3
1
3.29, and 4.74 mmol/g. The total H2 consumptions of MnO2 and
Mn O are quite close to the theoretical values. The results indi-
lower than those of MnO , it is considered that the OMS-2 catalyst is
3
4
2
4+
4+
3+
cate that substantial amount of Mn in MnO , and Mn and Mn
high in lattice oxygen mobility. For better understanding of the role
of lattice oxygen on the catalytic properties of OMS-2, a series of
OMS-2 differing in composition, morphology, and specific surface
area were synthesized and examined for the liquid-phase catalytic
2
in OMS-2 can be reduced to Mn2+ below 400 C. Furthermore, the
total amount of H2 consumption detected over OMS-2 suggests a
stage of mixed-valence. Being multi-valent is essential for electron
transfer in a material, and hence the co-existence of various oxida-
tion states of Mn in OMS-2 is important for electron transport. The
efficiency of catalysts, especially those for redox reaction, is usually
governed by their ability and tendency to cycle among the valence
states of manganese.
◦
Table 2
a
Results of p-chlorotoluene oxidation over different catalysts.
Entry
Catalyst
Conv. (mol%)
Product distribution (mol%)
PCBb
PCAb
PCBAb
Others
3
.2. Catalytic activity
1
2
3
4
5
6
7
8
No catalyst
OMS-2
MnO2
Mn3O4
Mn(OAc)2
Co2O3
0
0
0
0
0
0
0.7
0.9
0
0
0
0
86.0
54.5
26.5
31.1
49.7
41.1
0
68.7
44.1
45.1
33.6
1.1
31.3
42.8
36.2
19
2.5
7.6
0
To study the factors that influence p-chlorotoluene oxida-
13.1
18.7
46.7
95.5
66.7
0
tion, several manganese compounds and a number of transition
metal oxides were tested as catalysts. Table 2 lists the p-
chlorotoluene conversion and selectivity to p-chlorobenzaldehyde,
p-chlorobenzyl alcohol, and p-chlorobenzoic acids. In a blank run
V2O5
Fe2O3
25.7
0
0
(
without a catalyst) and over commercial Fe O , there is no conver-
2 3
a
Reaction conditions: catalyst = 50 mg, p-chlorotoluene = 1 mL, solvent (acetic
acid) = 10 mL, water =2 g, HBr (40 wt. %) = 0.02 g, oxygen flow rate = 50 mL/min,
sion of p-chlorotoluene. It is known that cobalt oxide [5,11,31] and
vanadium oxide [1,10] are excellent catalysts for C H bond activa-
tion in the oxidation of p-chlorotoluene to p-chlorobenzaldehyde.
We found that when Co O3 and V O5 are used as catalysts,
◦
time = 10 h, temperature = 100 C.
b
PCB: p-chlorobenzaldehyde; PCA: p-Chlorobenzyl alcohol; PCBA: p-
2
2
Chlorobenzoic acid.

Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
121
8
6
4
2
0
0
0
0
6
4
2
0
0
0
0
Conv.
Sel.
Yield
0
0
50
100
150
200
250
300
Amount of cat.(mg)
Fig. 6. Effect of catalyst amount on p-chlorotoluene oxidation (reaction conditions:
substrate 1 mL, solvent 10 mL, HBr (40 wt.%) 0.02 g, water 2 g, oxygen flow rate
Fig. 4. Reaction network during the oxidation of p-chlorotoluene using
metal/bromide catalysts [32].
◦
50 mL/min, time 10 h, temperature 100 C).
rate. Ultimately, water appears as a separate phase, and the rate
drops to almost that of the initial low value [33].
oxidation of p-chlorotoluene. We find that the catalysts higher in
oxygen mobility and with more lattice oxygen species exhibit bet-
ter catalytic activity. The related results will be published in the
near future.
As shown in Fig. 5, when the amount of water added ini-
tially to the system is lower than 2 g, there is enhancement of
catalytic activity. When the initial water amount exceeds 2 g,
there is decline in p-chlorotoluene conversion plausibly due to the
strong deactivation effect of water on reaction VI (Fig. 4). In par-
allel to the increase of p-chlorotoluene conversion, there is rise
of p-chlorobenzaldehyde selectivity. Also, when the initial water
amount exceeds 2 g, there is decrease in p-chlorobenzaldehyde
selectivity. It is because with obvious decrease of catalytic activity,
it is difficult for p-chlorobenzyl alcohol and p-chlorobenzyl acetate
to be further oxidized to p-chlorobenzaldehyde (reactions III and V,
Fig. 4). Moreover, an excessive amount of water may have an inhibi-
tion effect on reaction I (Fig. 4). Taking p-chlorobenzaldehyde yield
into consideration, the most suitable initial water amount should
be 2 g in the present reaction system.
3.3. Effect of reaction conditions
3.3.1. Effect of water addition
The sequence of reactions that occurs during metal/bromide
autoxidation of p-chlorotoluene is given in Fig. 4 [32]. The effect
of water addition on p-chlorotoluene conversion is depicted in
Fig. 5 One can see that water can promote the oxidation of p-
chlorotoluene, and there is a point that water would be in excess
and the catalysis is inhibited. In the absence of water, the autoxida-
tion of p-chlorotoluene takes place at very low rate. It is plausible
that under mild conditions, the extent of peroxide decomposi-
tion for the generation of radicals is low. After the addition of a
small quantity of water, there is enhanced radical concentration
because the protons of water molecules interact with the peroxide
molecules, facilitating the splitting of O O bond of H-bonded per-
oxide complexes [32]. When water is in excess, the concentration
of active water starts to decrease due to self-association, and there
is decline in radical production and diminution of overall reaction
3.3.2. Effect of catalyst amount
The catalyst amount was varied between zero and 300 mg.
In Fig. 6, one can see that even a small amount (20 mg) of
as-synthesized OMS-2 can lead to significant p-chlorotoluene con-
version (74.8%). Between catalyst amounts of 40 and 100 mg,
p-chlorotoluene conversion varies only slightly from 83.4% to
8
5.4%. The phenomenon suggests that the conversion is limited by
the availability of dissolved oxygen which is rather constant under
the adopted reaction conditions [34]. Furthermore, we studied the
effect of mass transfer limitation by varying the stirring rate from
8
6
4
2
0
0
0
0
6
4
2
0
0
0
0
5
00 to 1200 rpm. We observed that there is little change in conver-
sion as well as in selectivity. The results suggest that mass transfer
limitation has little effect on the reaction under the adopted con-
ditions.
However, when the amount of catalyst is above 100 mg, there
is significant decline in p-chlorotoluene conversion. It was pointed
out by Black [35] that transition metal salts or some biomimetic
catalytic systems function as catalysts at low loadings but as
inhibitors at high loadings. If more radicals are formed due to higher
loading of catalyst, there is actually a decline in activity due to
self-termination of radicals. It is believed that due to the combina-
tion of radicals, there is decline in radical concentration and hence
less conversion of p-chlorotoluene (Fig. 6). The variation of catalyst
amount between 40 and 300 mg does not affect the selectivity to
p-chlorobenzaldehyde significantly (59.7–68.9%). In terms of max-
imum p-chlorobenzaldehyde yield, a catalyst amount of 50 mg is
Conv.
Sel.
Yield
0
0
0.5
1
1.5
2
2.5
3
Initial water amount(g)
Fig. 5. Effect of water addition on p-chlorotoluene oxidation (reaction conditions:
catalyst 50 mg, substrate 1 mL, solvent 10 mL, HBr (40 wt.%) 0.02 g, oxygen flow rate
0 mL/min, time 10 h, temperature 100 C).
◦
5

1
22
Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
1
00
of p-chlorobenzaldehyde selectivity and rise of selectivity to p-
chlorobenzoic acid.
6
4
0
0
8
0
0
0
0
3.4. Leaching experiments and catalyst stability
6
4
2
To verify that there was no leaching of catalyst during the reac-
tion, OMS-2 was used under the adopted conditions but removed
after 4 h, and the mixture was left to react for another 10 h. It was
found that there is almost complete suspension of reaction upon
OMS-2 removal. In other words, the leach of catalyst into the liquid
phase is insignificant.
Conv.
Sel.
20
0
Yield
Tests were conducted to investigate the recyclability of the
OMS-2 catalyst. After each run, the catalyst was filtered out and
washed several times with acetone or alcohol before being dried
0
7
0
80
90
100
110
◦
at 110 C overnight. Across the four runs, there is only insignificant
o
Reaction temperature ( C)
change in p-chlorotoluene conversion and p-chlorobenzaldehyde
selectivity (Fig. S2). The XRD patterns (Fig. 1b) and N2 adsorp-
tion/desorption isotherms (Table 1) of used catalysts indicate that
the main characteristics of the catalyst are preserved during the
four cycles of oxidation reaction.
Fig. 7. Effect of reaction temperature on p-chlorotoluene oxidation (reaction condi-
tions: catalyst 50 mg, substrate 1 mL, solvent 10 mL, HBr (40 wt.%) 0.02 g, water 2 g,
oxygen flow rate 50 mL/min, time 10 h).
the most suitable for this liquid-phase p-chlorotoluene oxidation
system.
3.5. Possible reaction mechanism
The liquid-phase catalytic oxidation of p-chlorotoluene to
3
.3.3. Effect of reaction temperature
The influence of reaction temperature on the conversion of
p-chlorobenzoic acid in an acetic acid media is typical of aromatic
hydrocarbon oxidation processes, which belongs to the kind of
classical free-radical chain reactions. Researchers such as Kamiya
[37], Suresh et al. [38], and Wang et al. [39] studied the mechanism
of liquid-phase catalytic oxidation of aromatic hydrocarbons for
many years. The initiation mechanism is predominantly hydrogen
abstraction from methyl groups by bromine atoms. The Mn and/or
Co ions oxidize the resulted bromine ions to bromine atoms thus
ensuring the availability of bromine atoms for initiation.
◦
p-chlorotoluene was investigated from 70 to 110 C (Fig. 7). It
can be seen that the yield of p-chlorobenzaldehyde comes to
a maximum at 100 C. A rise of temperature above 100 C (e.g.
◦
◦
◦
1
10 C which is close to the boiling point) results in poor yield
of p-chlorobenzaldehyde. The phenomenon is a result of low O2
solubility in the liquid phase at such a high temperature [36].
We investigated the oxidation reaction of p-chlorotoluene with
3
.3.4. Effect of reaction time
The change of p-chlorotoluene, p-chlorobenzaldehyde, and
HBr in N , and obtained only 10.8% conversion. The finding cor-
2
roborates that the aerobic benzyl alcohol oxidation over OMS-2
catalysts proceeds according to a Mars-van Krevelen mechanism
16,40]. Thus, the conversion observed in the N2 stream can be
p-chlorobenzoic acid in the reaction system with time was inves-
tigated (Fig. 8) One can see rapid rise of p-chlorobenzaldehyde
selectivity whereas the rise of selectivity to p-chlorobenzoic acid
is moderate. Simultaneously, the selectivity to p-chlorobenzyl bro-
mide and p-chlorobenzyl alcohol decreases rapidly due to the
corresponding consecutive reactions VIII, IX and III, IV (Fig. 4). With
prolonged reaction, there is constant rise of p-chlorobenzaldehyde
concentration, leading to further oxidation and the formation of p-
chlorobenzoic acid (reaction VI, Fig. 4) as reflected in the decrease
[
explained by substrate oxidation with oxygen supplied from the
solid manganese-oxide-based OMS-2 catalyst. Without further
supply of gas-phase oxygen to reoxidize the catalyst, the reac-
tion slows down and essentially stops. According to the results
of this study and the mechanism depicted in literatures, we pro-
pose a mechanism for the liquid-phase catalytic oxidation of
p-chlorotoluene to p-chlorobenzaldehyde over OMS-2. When the
oxidation reaction of p-chlorotoluene is performed without HBr,
there is no p-chlorotoluene conversion. We hence deduce that at
the beginning the Mn(IV) ions oxidize bromine ions to bromine rad-
ical thus ensuring the availability of bromine radicals for initiation
100
Conv.
80
Sel.to p-chlorobenzaldehyde
Sel.to p-chlorobenzoic acid
Yield
8
0
0
0
0
(
see Eq. (1)) [41].
6
4
2
0
0
0
0
Mn(IV) + Br → Mn(III) + Br•
−
(1)
6
4
2
•
The Br radical initiates the oxidation of p-chlorotoluene by
hydrogen abstraction according to the following radical reaction:
Br + p-Cl-Ph-CH → p-Cl-Ph-CH + H+ + Br−
•
•
(2)
3
2
(
“Ph” denotes benzene ring and “p” denotes the para-substituted
•
group). The generated p-Cl-Ph-CH2 radical has high reactivity and
•
combines with molecular oxygen to generate p-CH -Ph-CH O
3
2
2
radical by the reaction:
0
p-Cl-Ph-CH2 + O2 → p-Cl-Ph-CH2O2•
•
(3)
2
4
6
8
10
12
14
Reaction time (h)
Then the following radical reaction occurs:
p-Cl-ph-CH O2 + Mn(III) + H+ → p-Cl-Ph-CHO + Mn(IV) + H O
•
Fig. 8. Effect of reaction time on p-chlorotoluene oxidation (reaction conditions:
2
2
catalyst 50 mg, substrate 1 mL, solvent 10 mL, HBr (40 wt.%) 0.02 g, water 2 g, oxygen
◦
flow rate 50 mL/min, temperature 100 C).
(4)

Y.-Q. Deng et al. / Applied Catalysis A: General 467 (2013) 117–123
123
The major chain termination reactions are that between two
radicals [42]. At early stage of reaction, we detected a higher con-
centration of p-chlorobenzyl bromide by GCMS, but at the last
period of the reaction p-chlorobenzyl bromide disappeared, via
the corresponding consecutive reactions VIII, IX (Fig. 4). We hence
Universities. C.T. Au thanks the Hunan University for an adjunct
professorship.
Y.Q. Deng and T. Zhang contributed equally.
Appendix A. Supplementary data
·
deduce that Br radicals have an important role to act in chain initi-
·
ation and propagation [33,42]. The following reaction between Br
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.apcata.2013.
radical and other radicals are considered as the chain termination
reaction.
0
7.015.
•
•
Br + p-Cl-Ph-CH → p-Cl-Ph-CH Br
(5)
2
2
References
The aromatic aldehyde group also can be oxidized to the cor-
·
responding peroxide radical RCO3 in the same manner as that of
methyl group, and then reacts with low valance metal to generate
aromatic peracid RCOOOH [43]. The reactions are as follows:
[1] R.A. Sheldon, H. Van Bekkum, Fine Chemicals through Heterogeneous Catalysis,
Wiley-VCH, Weinheim, 2001.
[
[
[
2] P. Anastas, J. Warner, Green Chemistry: Theory Practice, Oxford University
Press, New York, 1998.
3] F.M. Bautista, D. Luna, J. Luque, J.M. Marinas, J.F. Sánchez-Royo., Appl. Catal. A
•
Br + p-Cl-Ph-CHO → p-Cl-Ph-CO + H+ + Br−
•
(6)
(7)
(8)
352 (2009) 251–258.
4] F. Konietzni, U. Kolb, U. Dingerdissen, W.F. Maier, J. Catal. 176 (1998)
527–535.
•
•
p-Cl-Ph-CO + O → p-Cl-Ph-CO3
2
[
[
5] J. Wang, H. Fang, Y. Li, J. Li, Z. Yan, J. Mol. Catal. A: Chem. 250 (2006) 75–79.
6] YooF J.S., Appl. Catal., A 135 (1996) 261–271.
p-Cl-Ph-CO3 + Mn(III) + H+ → p-Cl-ph-COOOH + Mn(IV)
•
[7] H. Borchert, T. Gerdau, J. Weiguny, US Patent 5,648,551 (1997).
[
8] F.M. Bautista, J.M. Campelo, D. Luna, J. Luque, J.M. Marinas, Catal. Today 128
(2006) 183–190.
The reducibility of aldehyde group is strong and the follow-
ing self-catalyzed reaction between aldehyde group and aromatic
peracid occurs to generate aromatic acid [41]:
[
9] F.M. Bautista, J.M. Campelo, D. Luna, J. Luque, J.M. Marinas, Appl. Catal., A 325
(
2007) 336–344.
[
[
10] A.P. Singh, T. Selvam, Appl. Catal., A 143 (1996) 111–124.
11] R. Fatemeh, H.C. James, K. Babak, J.M. Duncan, Org. Biomol. Chem. 3 (2005)
725–726.
p-Cl-Ph-COOOH + p-Cl-Ph-CHO → 2p-Cl-Ph-COOH
(9)
[
12] T. Punniyamurthy, S. Velusamy, J. Iqbal, Chem. Rev. 105 (2005) 2329–2363.
Besides the generation of aromatic acid via Baeyer-Villiger reac-
tion (reaction Eq. (9)), peroxyacid reacts with Mn(III) to give the
aromatic acid according to the following reaction:
[13] Fabrizio.F C., J. Chem. Technol. Biotechnol. 85 (2010) 1175–1183.
[14] D.W. Liu, B.B. Garcia, Q.F. Zhang, Q. Guo, Y.H. Zhang, S. Sepehri, G.Z. Cao, Adv.
Funct. Mater. 19 (2009) 1015–1023.
[
15] Z.S. Wu, W. Ren, D.W. Wang, F. Li, B. Liu, H.M. Cheng, ACS Nano 4 (2010)
5835–5842.
p-Cl-Ph-COOOH + 2Mn(III) + 2H+
[
16] Y.C. Son, V.D. Makwana, A.R. Howell, S.L. Suib, Angew. Chem. Int. Ed. 40 (2001)
4280–4283.
→
p-Cl-Ph-COOH + 2Mn(IV) + H O
(10)
2
[
[
17] S.L. Suib, Acc. Chem. Res. 41 (2008) 479–487.
18] M.A. Peluso, L.A. Gambaro, E. Pronsato, D. Gazzoli, H.J. Thomas, J.E. Sambeth,
Catal. Today 133 (2008) 487–492.
[
[
19] V.D. Makwana, Y.C. Son, A.R. Howell, S.L. Suib, J. Catal. 210 (2002) 46–52.
20] G. Leofanti, M. Padovan, G. Tozzola, B. Venturelli, Catal. Today 41 (1998)
4
. Conclusion
Liquid-phase catalytic oxidation of p-chlorotoluene with molec-
2
07–219.
[
21] X. Tang, Y. Li, J. Chen, Y. Xu, W. Shen, Microporous Mesoporous Mater. 103
(2005) 250–256.
ular oxygen using synthesized OMS-2 octahedral molecular sieve
as catalyst results in promising conversion and selectivity in the
adopted atmospheric reflux system. The TPD and TGA results
confirm that there is a large amount of highly mobile lat-
tice oxygen in OMS-2, and the activation of p-chlorotoluene
is promoted as a result. Under optimum conditions, an 86.0%
conversion of p-chlorotoluene along with 68.7% selectivity to
p-chlorobenzaldehyde is obtained. The OMS-2 catalyst shows
promising reusability and can be recycled. A free-radical reac-
tion mechanism is proposed. The OMS-2 catalyst with such
activity, selectivity, and recyclability opens up new possi-
bilities for the conversion of aromatic hydrocarbons to the
[22] J. Luo, Q. Zhang, J. Garcia-Martinez, S.L. Suib, J. Am. Chem. Soc. 130 (2008)
198–3207.
3
[
[
23] Z.M. Wang, S. Tezuka, H. Kanoh, Chem. Mater. 13 (2001) 530–537.
24] H. Sun, S. Chen, P. Wang, X. Quan, Chem. Eng. J. 178 (2011) 191–196.
[25] H.C. Genuino, S. Dharmarathna, E.C. Njagi, M.C. Mei, S.L. Suib, J. Phys. Chem. C
16 (2012) 12066–12078.
26] V.P. Santos, M.F.R. Pereira, J.J.M. Órfão, Figueiredo.F J.L., Appl. Catal., B 99 (2010)
53–363.
[27] J. Luo, Q. Zhang, A. Huang, S.L. Suib, Microporous Mesoporous Mater. 35–36
2000) 209–217.
1
[
3
(
[
[
28] R. Wang, J. Li, Environ. Sci. Technol. 44 (2010) 4282–4287.
29] F. Wang, H. Dai, J. Deng, G. Bai, Y. Liu, K. Ji, Environ. Sci. Technol. 46 (2012)
4034–4041.
[30] E.R. Stobbe, B.A. Boer, J.W. Geus, Catal. Today 47 (1999) 161–167.
[31] A. Hu, C. Lu, B. Li, Huo.F T., Ind. Eng. Chem. Res. 45 (2006) 5688–5692.
[32] W. Partenheimer, J. Mol, Catal. A: Chem. 206 (2003) 105–119.
corresponding aldehydes with O in liquid-phase atmospheric sys-
[33] W. Partenheimer, Adv. Synth. Catal. 346 (2004) 297–306.
2
[
34] A.R. Li, S.W. Tang, P.H. Tan, C.J. Liu, B. Liang, J. Chem. Eng. Data 52 (2007)
339–2344.
tems.
2
[
[
35] J.F. Black, J. Am. Chem. Soc. 100 (1978) 527–535.
36] N. Takeo, I. Toshiyuki, M. Masatoshi, I. Takatoshi, M. Takumi, Tetrahedron Lett.
Acknowledgments
51 (2010) 2225–2227.
[
[
[
[
37] Y. Kamiya, J. Catal. 33 (1974) 480–485.
38] A.K. Suresh, M. Sharma, T. Sridhar, Ind. Eng. Chem. Res. 39 (2000) 3958–3997.
39] Q. Wang, Y. Zhang, Y. Cheng, X. Li, AIChE J. 54 (2008) 2674–2688.
40] F. Schurz, J.M. Bauchert, T. Merker, T. Schleid, H. Hasse, R. Glaser, Appl. Catal.,
A 355 (2009) 42–49.
The project was financially supported by NSFC (Grant Nos.
1273067, U1162109, 20873038), the program for New Century
Excellent Talents in Universities (NCET-10-0371), Program for
Changjiang Scholars and Innovative Research Team in Univer-
sity (IRT1238), Hunan Provincial Natural Science Foundation of
China (10JJ1003), and the Fundamental Research Funds for Central
2
[
[
[
41] P. Raghavendrachar, S. Ramachadran, Ind. Eng. Chem. Res. 31 (1992) 453–462.
42] M. Hronec, Z. Hrabe, Ind. Eng. Chem. Prod. Res. Dev. 25 (1986) 257–261.
43] L.J. Wang, X. Li, G. Xie, Y.W. Cheng, J. Siman, J. Chem. Ind. Eng. 54 (2003)
946–952.